Total synthesis and stereochemical reassignment of mandelalide A

Article abstract of DOI:10.1002/anie.201403542

The total synthesis of the tunicate metabolite mandelalide A and the correction of its originally assigned stereochemistry are reported. Key features of the convergent, fully stereocontrolled route include the use of a Prins cyclization for the diastereoselective construction of the tetrahydropyran subunit, Rychnovsky-Bartlett cyclization for the preparation of the tetrahydrofuran moiety, Suzuki coupling, Horner-Wadsworth-Emmons macrocyclization, and glycosylation to append the L-rhamnose-derived pyranoside.

Full text of DOI:10.1002/anie.201403542

Angewandte
Chemie
DOI: 10.1002/anie.201403542
Total Synthesis
Hot Paper
Total Synthesis and Stereochemical Reassignment of Mandelalide A**
Honghui Lei, Jialei Yan, Jie Yu, Yuqing Liu, Zhuo Wang, Zhengshuang Xu,* and Tao Ye*
Abstract: The total synthesis of the tunicate metabolite
mandelalide A and the correction of its originally assigned
stereochemistry are reported. Key features of the convergent,
fully stereocontrolled route include the use of a Prins cycliza-
tion for the diastereoselective construction of the tetrahydro-
pyran subunit, Rychnovsky–Bartlett cyclization for the prep-
aration of the tetrahydrofuran moiety, Suzuki coupling,
Horner–Wadsworth–Emmons macrocyclization, and glycosy-
lation to append the l-rhamnose-derived pyranoside.
mandelalide A and the resulting reassignment of the stereo-
chemical configuration of the natural product.
Our retrosynthetic analysis of mandelalide A 1 is shown in
Scheme 1. It is envisaged that a late-stage glycosylation[3] of
M
andelalide A (1) is an extraordinary glycosylated macro-
lide that was recently isolated from a new species of
Lissoclinum ascidian, collected from Algoa Bay, South
Africa.[1] The assignment of the relative configuration was
accomplished by considering the homonuclear and hetero-
nuclear coupling constants in tandem with ROESY data. The
absolute configuration of mandelalide A was assigned
through chiral GC-MS analysis of the hydrolyzed monosac-
charide and correlation with ROESY data.[1] Intriguing
structural features of mandelalide A include a 24-membered
a,b-unsaturated macrolactone, which entails a conjugated
diene, a trisubstituted tetrahydrofuran (THF) moiety, and
a trisubstituted tetrahydropyran (THP) fragment appended
with an unusual carbohydrate unit, 2-O-methyl-a-l-rham-
nose. Furthermore, a total of nine stereogenic centers are
present in the carbon backbone of mandelalide A. Mandela-
lide A exhibited potent cytotoxicity to human NCI-H460 lung
cancer cells (IC50: 12 nm) and mouse Neuro-2A neuroblas-
toma cells (IC50: 29 nm).[1] We have been engaged in a program
devoted to the total synthesis of biologically active marine
natural products.[2] Herein, we disclose the total synthesis of
[*] H. Lei, J. Yan, J. Yu, Prof. Z. S. Xu
Laboratory of Chemical Genomics
Scheme 1. Retrosynthetic analysis of mandelalide A (1).
School of Chemical Biology and Biotechnology
Peking University Shenzhen Graduate School
Xili, Nanshan District, Shenzhen, 518055 (China)
E-mail: xuzs@pkusz.edu.cn
the aglycone fragment 2 with the l-rhamnose-derived thio-
glycosyl donor 3 would produce the natural product in its
protected form. Careful inspection of the complete aglycone
framework reveals that the 24-membered macrocycle could
be assembled from two subunits, that is 4 and 5, of comparable
complexity through Suzuki coupling and Horner–Wads-
worth–Emmons (HWE) macrocyclization.[4] Fragment 5,
which entails the tetrahydrofuran motif, could be accessed
through a Rychnovsky–Bartlett cyclization[5] of a suitable
alkene (8, 9, or 10). Subunit 4, which contains the tetrahy-
dropyran ring, would arise from Prins cyclization[6] of
aldehyde 6 and alcohol 7.
Dr. Y. Liu, Z. Wang, Prof. T. Ye
Department of Applied Biology & Chemical Technology
The Hong Kong Polytechnic University
Kowloon, Hong Kong (China)
E-mail: tao.ye@polyu.edu.hk
tao_ye35@hotmail.com
[**] We acknowledge financial support from the Hong Kong Research
Grants Council (PolyU 5040/10P, PolyU 5037/11P, PolyU 5020/12P,
PolyU5030/13P, and PolyU153035/14P), the Fong Shu Fook Tong
Foundation, the Joyce M. Kuok Foundation, the National Natural
Science Foundation of China (21072007, 21272011, and 21133002),
and the Shenzhen Science and Technology Development Fund
(JCYJ20130329175740481 and ZYC201105170351A).
The synthesis of fragment 5 commenced from the known
homoallylic alcohol 11.[7] As shown in Scheme 2, homoallylic
alcohol 11 was protected as its 2,6-dichlorobenzyl ether, and
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201403542.
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

.
Angewandte
Communications
Scheme 3. Synthesis of subunit 5. a) DMP, NaHCO3, CH2Cl2, 08C!
RT; b) LiCl, trimethyl phosphonoacetate, DIPEA, CH3CN, RT, 95%;
c) DIBAL-H, THF, ꢀ788C!ꢀ408C, 98%; d) I2, CH3CN, 08C!RT,
94%; e) K2CO3, MeOH, RT, 84%; f) CuI, vinylmagnesium bromide,
THF, ꢀ788C!ꢀ208C, 95%; g) TBSOTf, 2,6-lutidine, CH2Cl2, ꢀ788C!
ꢀ308C, 98%; h) DDQ, CH2Cl2, RT, 89%; i) DMP, NaHCO3, CH2Cl2,
08C!RT, 96%; j) [ICH2PPh3]I, NaHMDS, HMPA, THF, ꢀ788C, 82%;
k) AD-mix-a, tBuOH/H2O, 08C, 20/21=1:2, 82%; l) TBSCl, imidazole,
DMAP, CH2Cl2, RT, 96%; m) dimethylphosphonoacetic acid, 2,4,6-
trichlorobenzoyl chloride, Et3N; then DMAP, toluene, 08C!RT, 92%.
DDQ=2,3-dichloro-5,6-dicyano-1,4-benzoquinone, DIBAL-H=diisobu-
tylaluminum hydride, DIPEA=N,N-diisopropylethylamine, DMAP=4-
dimethylaminopyridine, HMPA=hexamethylphosphoramide,
NaHMDS=sodium bis(trimethylsilyl)amine, TBSCl=tert-butyldime-
thylsilyl chloride, TBSOTf=tert-butyldimethylsilyl trifluoromethanesul-
fonate, THF=tetrahydrofuran.
Scheme 2. Rychnovsky–Bartlett cyclization of 8 and 9. a) NaH, 2,6-
dichlorobenzyl bromide, Bu4NI, THF, 08C!RT, 94%; b) 9-BBN, 08C!
RT; then NaOH, H2O2, reflux, 96%; c) DMP, NaHCO3, CH2Cl2, 08C!
RT, 98%; d) 13, KHMDS, DME, ꢀ788C, 90%; e) I2, CH3CN, 08C!RT,
61%; f) CSA, MeOH, RT, 91%; g) diethyl carbonate, K2CO3, 808C,
95%; h) I2, CH3CN, 08C!RT, 95%. 9-BBN=9-borabicyclo-
[3.3.1]nonane, CSA=camphorsulfonic acid, DME=dimethoxyethane,
DMP=Dess–Martin periodinane; KHMDS=potassium bis(trimethylsi-
lyl)amide.
the terminal alkene was subjected to hydroboration with 9-
BBN to afford the corresponding primary alcohol 12 in 90%
yield. Dess–Martin oxidation of alcohol 12 afforded the
corresponding aldehyde, which was then coupled with chiral
bromide in the presence of catalytic amounts of CuI. The
secondary alcohol of 17 was protected as its TBS ether, and
the benzyl ether was cleaved according to the procedure of
Mori and co-workers[12] to afford alcohol 18 in 89% yield.
Dess–Martin oxidation of alcohol 18 afforded the corre-
sponding aldehyde, which was converted into the requisite
(Z)-vinyl iodide 19 in 82% yield according to the Wittig–
Stork–Zhao olefination protocol.[13] A selective dihydroxyla-
tion of the terminal olefin of 19 using the Sharpless AD-mix-
a reagent[14] provided diol 21 (55% yield) together with its
minor diastereoisomer 20 (27% yield).[15] The primary
alcohol of 21 was protected as its TBS ether, and the
secondary alcohol was condensed with dimethylphosphono-
acetic acid under the Yamaguchi conditions[16] to produce
phosphonate 5 in 88% yield.
sulfone 13[8] through a Kocienski–Julia olefination[9] to give 8
´
in 88% yield. Unfortunately, Rychnovsky–Bartlett cycliza-
tion[5] of alkene 8 led to the undesired 2,4-disubstituted
tetrahydrofuran 14 as the major product (61% yield), which is
formed through iodoetherification of the acetonide moiety.[10]
Removal of the acetonide moiety in 8, followed by re-
protection of the resulting diol gave rise to cyclic carbonate 9
in 86% yield. When 9 was submitted to the conditions for
a Rychnovsky–Bartlett cyclization, the expected tetrahydro-
furan 15 was obtained in 95% yield. However, attempts to
convert this iodide into the corresponding alcohol
(AgCO2CF3, DME, then H2O) met with failure.[11]
Bearing in mind the problems encountered with precur-
sors 8 and 9, we embarked on the ultimately successful route
towards the construction of tetrahydrofuran 5 (Scheme 3).
Alcohol 12 was homologated into allylic alcohol 10 in 91%
yield by a three-step sequence that included Dess–Martin
oxidation, Horner–Wadsworth–Emmons olefination, and
reduction of the resulting a,b-unsaturated ester with
DIBAL-H. 2,5-Dichlorobenzyl ether 10 was treated with
iodine in acetonitrile at low temperature to afford the desired
2,5-cis-disubstituted tetrahydrofuran 16 as the sole stereoiso-
mer in 94% yield. Tetrahydrofuran 16 was then converted
into allylic alcohol 17 in 80% yield by a two-step sequence,
namely base-promoted epoxide formation followed by nucle-
ophilic opening of the resulting epoxide with vinylmagnesium
We next explored an intermolecular Prins cyclization for
the construction of the tetrahydropyran subunit 4 (Scheme 4).
Scheme 4. Prins cyclization of aldehyde 6 and homoallylic alcohol 7.
a) TFA, pentane, ꢀ58C; b) K2CO3, MeOH, RT, 54% over 2 steps.
TFA=trifluoroacetic acid.
2
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
These are not the final page numbers!

Angewandte
Chemie
Condensation of aldehyde 6[17] with homoallylic alcohol 7[18] in
the presence of trifluoroacetic acid induced the Prins cycliza-
tion[6] and afforded the corresponding tetrahydropyranyl
trifluoroacetate, which was not isolated, but immediately
treated with potassium carbonate in methanol to give rise to
tetrahydropyran 22 in good yield. As shown in Table 1,
Scheme 6. Synthesis of glycosyl donor 3. a) trimethyl orthoformate,
2,2,3,3-tetramethoxybutane, CSA, MeOH, reflux, 98%; b) NaH, MeI,
DMF, 08C!RT, 88%; c) TFA, CH2Cl2/H2O, 08C!RT, 97%; d) TBSOTf,
2,6-lutidine, CH2Cl2, 08C, 95%; e) m-CPBA, CH2Cl2, ꢀ788C, 90%. m-
CPBA=meta-chloroperoxybenzoic acid.
Table 1: Prins cyclization of aldehyde 6 and homoallylic alcohol 7.
Entry
t [h]
Solvent
T [8C]
d.r.[a]
Yield [%]
1
2
3
4
3
24
1
CH2Cl2
CH2Cl2
CH2Cl2
pentane
0
ꢀ20
ꢀ5
5:1
1:1
6:1
7:1
44
41
42
54
(26)[22] as its bis-acetal 27 (Scheme 6).[23] The remaining
hydroxy group was immediately methylated (NaH, MeI) to
afford the corresponding methyl ether in 88% yield.[24]
Removal of the bis-acetal (TFA, CH2Cl2/H2O) followed by
re-protection of the resulting diol afforded the bis-TBS ether
28, which was then submitted to a m-CPBA oxidation to
afford sulfoxide 3 in 90% yield.[25]
Our fragment assembly started with Suzuki reaction of
vinyl boronate 4 with vinyl iodide 5, which proceeded
smoothly to give the desired diene 29 in 88% yield
(Scheme 7). The primary hydroxy group was selectively
oxidized using the Piancatelli protocol[26] to provide the
macrocyclization precursor, which was then subjected to an
intramolecular HWE reaction under Roush–Masamune con-
ditions[27] to afford aglycone fragment 2 in 44% yield over two
steps. Coupling of 2 and 3 by a Kahne glycosylation[25] through
sulfoxide activation[28] furnished 30 in 66% yield as a single
diastereomer. Global desilylation of 30 using tris(dimethyla-
1
ꢀ5
[a] d.r. determined by NMR analysis.
lowering the reaction temperature to ꢀ208C led to a decrease
in reactivity, and prolonged reaction times resulted in
disappointingly low diastereoselectivity (entry 2).[19] Under
the optimized conditions (entry 4; TFA, pentane, ꢀ58C, 1 h),
22 was obtained in 54% yield with a diastereomeric ratio of
7:1.
The synthesis of vinyl boronate 4 commenced with
hydrogenolysis of the benzyl ether in 22 to afford the
corresponding diol, which was re-protected as the bis-PMB-
protected ether 23 (Scheme 5). The fully protected tetrahy-
Scheme 5. Synthesis of vinyl boronate 4. a) Pd/C, H2, MeOH, RT, 97%;
b) NaH, PMBBr, Bu4NI, THF, 08C!RT, 97%; c) TBAF, THF, RT, 97%;
d) DMP, NaHCO3, CH2Cl2, 08C!RT, 98%; e) 24, K2CO3, MeOH/THF,
08C, 94%; f) pinacolborane, dicyclohexylborane, THF, 08C!RT;
g) DDQ, CH2Cl2/buffer (pH 7), RT, 68% over 2 steps. DDQ=2,3-
dichloro-5,6-dicyano-1,4-benzoquinone, PMBBr=p-methoxybenzyl bro-
mide.
dropyran 23 was converted into terminal alkyne 25 in 89%
yield through a three-step sequence that involved TBAF-
mediated desilylation, Dess–Martin oxidation, and subse-
quent homologation of the resulting aldehyde with the Ohira–
Bestmann reagent (24).[20] Dicyclohexylborane-mediated
hydroboration of terminal alkyne 25 with pinacolborane[21]
gave rise to the corresponding (E)-1-alkenylboronic acid
pinacol ester, which was treated with DDQ in dichloro-
methane/ buffer solution (pH 7.1) to provide fragment 4 in
68% yield over two steps.
Scheme 7. Synthesis of proposed mandelalide A (1). a) [Pd(PPh3)4],
Ag2O, THF/H2O, RT, 88%; b) TEMPO, PhI(OAc)2, CH2Cl2, RT; c) LiCl,
DIPEA, CH3CN, RT, 44% over 2 steps; d) 3, M.S. (4 ꢀ), DTBMP, Tf2O,
ꢀ788C!ꢀ358C, 66%; e) TASF, DMF/THF, 08C. (1: 43% yield; 31:
9% yield). DTBMP=2,6-di-tert-butyl-4-methylpyridine, TASF=tris(di-
methylamino)sulfur (trimethylsilyl)difluoride, M.S.=molecular sieves,
TEMPO=2,2,6,6-tetramethylpiperidine-1-oxyl.
The synthesis of glycosyl donor 3 began with the selective
protection of known phenyl-1-thio-a-l-rhamnopyranoside
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

.
Angewandte
Communications
mino)sulfur (trimethylsilyl)difluoride (TASF) provided the
targeted product 1 (43% yield)[29] along with the isomer
obtained through macrolide ring expansion 31 (9% yield).
1
Unfortunately, neither the H nor the 13C NMR spectra of
1 were identical with those reported for natural mandelali-
de A,[1] which suggested that the reported structure (i.e., 1)
must be incorrect.[30] As shown in Figure 1, the 13C NMR
chemical shifts of the carbohydrate unit of 1 matched closely
Figure 1. Differences in the 13C NMR chemical shifts between natural
mandelalide A and synthetic samples.
with the values reported for the natural product. However,
there were obvious discrepancies between the chemical shifts
in the regions of the tetrahydrofuran and tetrahydropyran
subunits; in particular, the observed 13C NMR chemical shifts
for the C11 stereogenic center and its appended methyl
substituent (C25) significantly deviated from those for the
natural product.
Figure 2. Structures of proposed mandelalide A (1), madeirolide A
(32), and the two diastereomers 1a and 1b.
With many building blocks already in hand, the next step
was to prepare a batch of ent-11, and this was readily achieved
by following the same synthetic procedure as for 11, but using
the enantiomer of the previously employed crotylation agent.
As shown in Scheme 8, ent-11 was transformed into ent-19 by
following the same synthetic procedure as for 19. A selective
dihydroxylation of the terminal olefin of ent-19 using the
Sharpless AD-mix-b reagent[14] provided the desired diol ent-
21 together with its minor diastereoisomer ent-20.[15] Further
elaboration of ent-21 and ent-20 into 1a and 1b included the
Having established that the published structure was
incorrect, we wished to elucidate the correct structure of
mandelalide A. The actual structure of mandelalide A
appeared to be epimeric to the proposed structure (1) at
one or more stereogenic centers in the ring system. As the
relative stereochemistry of 1 was assigned by considering
homonuclear and heteronuclear coupling constants in tandem
with ROESY data measured on a macrolide possessing
considerable flexibility, an error in the relative stereochem-
istry between the tetrahydrofuran subunit and the tetrahy-
dropyran fragment seemed most likely. As shown in Figure 2,
mandelalide A shares several structural features with the
scarce marine sponge metabolite madeirolide A.[31] A com-
parison of the structure proposed for mandelalide A (1) with
that of madeirolide A (32) reveals a critical difference in the
proposed relative stereochemistry between the tetrahydro-
furan moiety and the tetrahydropyran fragment of the
molecule. The methyl-substituted stereogenic centers and
the tetrahydropyran rings of mandelalide A and madeiroli-
de A have the same absolute configuration, whereas the
tetrahydrofuran moieties are enantiomeric. On the basis of
the original data in conjunction with our synthetic efforts and
biosynthetic considerations, we postulated that the correct
structure of mandelalide A was a diastereomer of 1 for which
the whole tetrahydrofuran moiety had been inverted. As the
configuration of the C23 stereogenic center could be ambig-
uous, we chose to synthesize both the C23-(S) and C23-(R)
epimers of mandelalide A with an inverted tetrahydrofuran
moiety (1a and 1b; Figure 2). Given our convergent
approach, testing this hypothesis was straightforward.
Scheme 8. Synthesis of revised mandelalide 1a and its C23 epimer 1b.
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2014, 53, 1 – 6
4
www.angewandte.org
These are not the final page numbers!

Angewandte
Chemie
Chen, L. Song, Z. Xu, T. Ye, Org. Lett. 2010, 12, 2036; f) B. Chen,
L. Dai, H. Zhang, W. Tan, Z. Xu, T. Ye, Chem. Commun. 2010,
46, 574.
incorporating of vinyl boronate 4 under the previously
described conditions, through Suzuki coupling, HWE macro-
cyclization, and glycosylation. These were readily achieved,
and 1a and 1b were obtained in 5.2% and 1.9% overall yield
starting from ent-11, respectively. Unlike the proposed
structure of mandelalide A (1), which readily underwent
ring expansion to afford 31, both 1a and 1b were found to be
stable under the desilylation conditions. Gratifyingly, 1H, 13C,
and mass spectra of 1a were found to be completely identical
to those of natural mandelalide A, which led us to the
conclusion that 1a indeed corresponds to the actual stereo-
chemistry of mandelalide A. Interestingly, the optical rotation
[3] H. Pellissier, Tetrahedron 2005, 61, 2947.
[4] J. A. Bisceglia, L. R. Orelli, Curr. Org. Chem. 2012, 16, 2206.
[5] S. D. Rychnovsky, P. A. Bartlett, J. Am. Chem. Soc. 1981, 103,
3964.
[6] a) X. Han, G. Peh, P. E. Floreancig, Eur. J. Org. Chem. 2013,
1193; b) I. M. Pastor, M. Yus, Curr. Org. Chem. 2012, 16, 1277.
[7] A. B. Smith III, G. K. Friestad, J. Barbosa, E. Bertounesque,
K. G. Hull, M. Iwashima, Y. Qiu, B. A. Salvatore, P. G. Spoors,
J. J.-W. Duan, J. Am. Chem. Soc. 1999, 121, 10468.
[8] a) L. A. Paquette, S. K. Chang, Org. Lett. 2005, 7, 3111; b) C.
Nilewski, N. R. Deprez, T. C. Fessard, D. B. Li, R. W. Geisser,
E. M. Carreira, Angew. Chem. 2011, 123, 8087; Angew. Chem.
Int. Ed. 2011, 50, 7940.
[9] a) P. R. Blakemore, W. J. Cole, P. J. Kocienski, A. Morley, Synlett
1998, 26; b) P. R. Blakemore, J. Chem. Soc. Perkin Trans. 1 2002,
2563.
[10] L. Zhu, D. R. Mootoo, Org. Lett. 2003, 5, 3475.
[11] P. A. Bartlett, P. C. Ting, J. Org. Chem. 1986, 51, 2230.
[12] H. Furuta, Y. Hasegawa, M. Hase, Y. Mori, Chem. Eur. J. 2010,
16, 7586.
[13] G. Stock, K. Zhao, Tetrahedron 1989, 30, 2173.
[14] H. C. Kolb, M. S. VanNieuwenhze, K. B. Sharpless, Chem. Rev.
1994, 94, 2483.
[15] For detailed assignment of the stereochemistry of the newly
formed secondary alcohol 21, see the Supporting Information.
[16] a) J. Inanaga, K. Hirata, H. Saeki, T. Katsuki, M. Yamaguchi,
Bull. Chem. Soc. Jpn. 1979, 52, 1989; b) M. Hikota, Y. Sakurai,
K. Horita, O. Yonemitsu, Tetrahedron Lett. 1990, 31, 6367.
[17] F. Keyling-Bilger, G. Schmitt, A. Beck, B. Luu, Tetrahedron
1996, 52, 14891.
[18] F. Fu, T.-P. Loh, Tetrahedron Lett. 2009, 50, 3530.
[19] The mechanism of the racemization in the Prins cyclization
reactions has been elucidated; see: R. Jasti, S. D. Rychnovsky, J.
Am. Chem. Soc. 2006, 128, 13640.
[20] S. Mꢀller, B. Liepold, G. J. Roth, H. J. Bestmann, Synlett 1996,
521.
20
value of our synthetic sample [1a; ½aꢁD ¼ꢀ34.6 (c = 0.25,
MeOH)] is higher than that reported for the natural material
23
[½aꢁD ¼ꢀ9 (c = 0.25, MeOH)].[1] The reason for the low optical
rotation value for the natural mandelalide A is unclear at
present, although it might be attributed to contamination by
a small amount of a highly optically active impurity in the
natural sample.
An initial cytotoxicity evaluation of the synthetic man-
delalide A and its diastereomers (including one aglycone) was
performed across a panel of ten cancer cell lines of different
histological origins.[32] These studies revealed no significant
cytotoxicities for the tested synthetic mandelalide A or its
analogues. It should be noted that the related structural
analogue madeirolide A, which was isolated by Winder and
Wright, also exhibited very minimal inhibition of prolifera-
tion against the AsPC-1 and PANC-1 pancreatic cancer cell
lines at 10 mgmLꢀ1.[31a]
In summary, we have achieved the total synthesis of the
proposed and revised structures of mandelalide A (1 and 1a)
along with two analogues (31 and 1b), which enabled us to
revise the structure that was originally proposed for natural
mandelalide A. The convergent approach features a highly
diastereoselective Prins cyclization for the construction of the
tetrahydropyran subunit and Rychnovsky–Bartlett cycliza-
tion for the preparation of the tetrahydrofuran moiety. Suzuki
coupling, Horner–Wadsworth–Emmons macrocyclization,
and glycosylation also served as key reactions for the total
synthesis. The application of this strategy to the synthesis of
mandelalide B is in progress, and the results will be reported
in due course.
[21] K. Shirakawa, A. Arase, M. Hoshi, Synthesis 2004, 1814.
[22] M. Fragoso-Serrano, G. Guillꢁn-Jaramillo, R. Pereda-Miranda,
C. M. Cerda-Garcꢂa-Rojas, J. Org. Chem. 2003, 68, 7167.
[23] S. V. Ley, H. W. M. Priepke, S. L. Warrine, Angew. Chem. Int.
Ed. Engl. 1994, 33, 2290; Angew. Chem. 1994, 106, 2410.
[24] a) S. V. Ley, D. R. Owen, K. E. Wesson, J. Chem. Soc. Perkin
Trans. 1 1997, 2805; b) A. Cirla, A. R. McHale, J. Mann,
Tetrahedron 2004, 60, 4019.
[25] D. Kahne, S. Walker, Y. Cheng, D. Van Engen, J. Am. Chem. Soc.
1989, 111, 6881.
[26] A. De Mico, R. Margarita, L. Parlanti, A. Vescovi, G. Piancatelli,
J. Org. Chem. 1997, 62, 6974.
[27] M. A. Blanchette, W. Choy, J. T. Davis, A. P. Essenfeld, S.
Masamune, W. R. Roush, T. Sakai, Tetrahedron Lett. 1984, 25,
2183.
[28] H. Wehlan, M. Dauber, M. T. M. Fernaud, J. Schuppan, S.
Keiper, R. Mahrwald, M.-E. J. Garcia, U. Koert, Chem. Eur. J.
2006, 12, 7378.
[29] J. Willwacher, A. Fꢀrstner, Angew. Chem. Int. Ed. 2014, 53, 4217;
Angew. Chem. 2014, 126, 4301.
Received: March 20, 2014
Published online: && &&, &&&&
Keywords: configuration determination · diastereoselectivity ·
.
marine macrolides · structure elucidation · total synthesis
[1] J. Sikorska, A. M. Hau, C. Anklin, S. Parker-Nance, M. T.
Davies-Coleman, J. E. Ishmael, K. L. McPhail, J. Org. Chem.
2012, 77, 6066.
[30] See the Supporting Information.
[2] a) L. Song, J. Liu, H. Gui, C. Hui, J. Zhou, Y. Guo, P. Zhang, Z.
Xu, T. Ye, Chem. Asian J. 2013, 8, 2955; b) H. Liu, Y. Liu, Z.
Wang, X. Xing, A. R. Maguire, H. Luesch, H. Zhang, Z. Xu, T.
Ye, Chem. Eur. J. 2013, 19, 6774; c) L. Dai, B. Chen, H. Lei, Z.
Wang, Y. Liu, Z. Xu, T. Ye, Chem. Commun. 2012, 48, 8697; d) S.
Li, Z. Chen, Z. Xu, T. Ye, Chem. Commun. 2010, 46, 4773; e) Z.
[31] a) P. L. Winder, Ph.D. Thesis, Florida Atlantic University, 2009;
b) I. Paterson, G. W. Haslett, Org. Lett. 2013, 15, 1338.
[32] For an initial cytotoxicity evaluation of the synthetic compounds,
see the Supporting Information.
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!

.
Angewandte
Communications
Communications
Total Synthesis
Structural revision: A revised configura-
tional assignment for the marine macro-
lide mandelalide A is proposed and vali-
dated by total synthesis. This study is one
of several recent examples in a growing
list of investigations that correct misas-
signed structures of natural products by
stereocontrolled total synthesis.
H. Lei, J. Yan, J. Yu, Y. Liu, Z. Wang,
Z. S. Xu,* T. Ye*
&&&& — &&&&
Total Synthesis and Stereochemical
Reassignment of Mandelalide A
6
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
These are not the final page numbers!