Synthesis and biological properties of the cytotoxic 14-membered macrolides aspergillide A and B

Article abstract of DOI:10.1002/chem.201001682

Total, stereoselective syntheses of the naturally occurring, cytotoxic macrolides aspergillide A and B are described. Olefin metatheses and asymmetric allylations were key steps in the synthetic sequences. Cytotoxicity assays against several tumor cell li

Full text of DOI:10.1002/chem.201001682

FULL PAPER
DOI: 10.1002/chem.201001682
Synthesis and Biological Properties of the Cytotoxic 14-Membered
Macrolides Aspergillide A and B
Santiago Dꢀaz-Oltra,[a] Cꢁsar A. Angulo-Pachꢂn,[a] Juan Murga,[a] Eva Falomir,[a]
Miguel Carda,*[a] and J. Alberto Marco*[b]
Dedicated to Professor Josꢀ Barluenga on the occasion of his 70th birthday
Abstract: Total, stereoselective syntheses of the naturally occurring, cytotoxic mac-
rolides aspergillide A and B are described. Olefin metatheses and asymmetric ally-
lations were key steps in the synthetic sequences. Cytotoxicity assays against sever-
al tumor cell lines have been performed for the two aspergillides and some of the
intermediates or side products of the synthetic sequence. One of these intermedi-
ates has been found markedly active against the human leukemia cancer cell line
HL-60, with an IC50 value comparable with that of the clinical drug fludarabine.
Keywords: antitumor
asymmetric allylation · cytotoxicity ·
macrolides · olefin metathesis
agents
·
Introduction
originally published structures of aspergillides A and B, as
well as to a confirmation of the structure of aspergillide C.
The aspergillides A, B and C (Scheme 1, 1–3) are three 14-
membered macrolides isolated from a strain of the marine-
derived fungus Aspergillus ostianus cultivated in a bromine-
modified medium.[1,2] The compounds showed cytotoxic ac-
tivity in the micromolar range against mouse lymphocytic
leukemia cells (L1210). Their stereostructures show some
unusual features. For instance, only two recent examples
have been reported of naturally occurring, 14-membered
macrolides that possess a tetrahydropyran ring not forming
part of a hemiacetal or acetal moiety.[3,4] This and the afore-
mentioned bioactivities prompted us and other groups to
perform total syntheses of these compounds. As commented
below, these synthetic efforts have led to a correction of the
Scheme 1. Correct stereostructures of aspergillides A (1), B (2) and C
(3).
In the beginning of 2009, Uenishi et al. published their
synthesis of a compound with structure 2, which at that time
was believed to correspond to aspergillide A.[5] As a matter
of fact, these authors found that their synthetic compound
had spectral properties identical with those reported for as-
pergillide B. The latter compound was thus assigned struc-
ture 2, which led to the need of a revised structure for asper-
gillide A. Shortly afterwards, we published our own synthe-
sis of compound 2, which was found to be identical with as-
pergillide B,[6] therefore confirming the findings of Uenishi
and his group.[5] Eventually, the group who isolated these
natural compounds succeeded in carrying out X-ray diffrac-
tion analyses[7] of suitable derivatives of aspergillides A and
B. This led to the assignment of 1 and 2, respectively, as the
correct stereostructures of these natural compounds.[8] Very
recently, we have communicated our own synthesis of asper-
[a] Dr. S. Dꢀaz-Oltra, C. A. Angulo-Pachꢁn, Dr. J. Murga,
Dr. E. Falomir, Prof. Dr. M. Carda
Depart. de Q. Inorgꢂnica y Orgꢂnica, Univ. Jaume I
Avda. Sos Baynat s/n, 12071 Castellꢁn (Spain)
Fax : (+34)964728214
E-mail: mcarda@qio.uji.es
[b] Prof. Dr. J. A. Marco
Depart. de Q. Orgꢂnica, Univ. de Valencia, c/D. Moliner, 50
46100 Burjassot, Valencia (Spain)
Fax : (+34)963544328
E-mail: alberto.marco@uv.es
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201001682.
Chem. Eur. J. 2011, 17, 675 – 688
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
675

gillide A, which confirms the structure deduced from the
diffraction analysis.[9] Finally, the structure of aspergillide C
has recently been confirmed by two total syntheses.[10]
In the present manuscript, we wish to publish the full de-
tails of our total syntheses of aspergillides A (1) and B (2).
We will also report on the cytotoxic activity of these two
compounds and of some of the intermediates of the synthet-
ic sequence against several cancer cell lines.
tion to aldehyde 9 and retro-oxidative cleavage of the ole-
finic bond finally leads to the known compound 10.[19]
Scheme 3 shows the details of the synthesis. Alcohol 10
was benzylated to 11 under mild, nonbasic conditions.[20]
Ozonolytic cleavage of the olefinic bond yielded aldehyde 9,
which was first purified and then subjected to Brown’s asym-
metric allylboration.[21] This gave homoallyl alcohol 12 in a
very high diastereomeric purity (d.r. >95:5 by NMR), which
was subsequently protected as its triethylsilyl derivative 8.
Isomerization of the terminal olefinic bond to the internal
position was achieved by means of Wipf’s adaptation of the
catalytic method reported by Roy et al.[22] With the aid of
this procedure, compound 8 was converted into its isomer 7
in high yield.[23] Compound 7 was obtained as a 9:1 E/Z mix-
ture[24] which proved difficult to separate and was thus car-
ried as such until the CM step. Cleavage of the two silyl
groups and selective oxidation of the primary alcohol with
Results and Discussion
At the onset of our research, aspergillide A was still as-
sumed to have structure 2. Its retrosynthetic analysis is de-
picted in Scheme 2. Hydrolytic opening of the lactone ring
and inverse metathesis yields the known alcohol 4[11] and tet-
rahydropyran 5. The reconstitution of the lactone ring from
precursors 4 and 5 was feasible in principle either by means
PhIACHTNUTGRNEUNG
(OAc)2/TEMPO[16g,25] afforded d-lactone 6. Reduction of
ꢀ
of esterification (intermolecular C O bond formation) fol-
lowed by ring-closing metathesis (RCM, intramolecular C=
6 with DIBAL followed by acetylative quenching yielded
the acetylated lactol 14 as a mixture of stereoisomers,[26]
which were not separated. The mixture was subsequently
treated with the trimethylsilyl enolate of tert-butyl thioace-
tate[27] in the presence of the Lewis acids BF3-etherate and
TMSOTf.[15a,28] This furnished the trans-2,6-disubstituted[16,29]
tetrahydropyran 15a in 55% yield, accompanied by its
epimer at C-3, 15b (21%). After chromatographic separa-
tion, alkaline hydrolysis of 15a provided acid 5 in high yield.
Treatment of 5 with five equivalents of olefinic alcohol 4
in the presence of 20% of the second-generation ruthenium
catalyst Ru-II caused cross-metathesis[13] and afforded hy-
droxy acid 16 in 89% yield as a 7:3 E/Z mixture. Macrolac-
tonization was performed on the mixture by means of the
Yamaguchi procedure[30] and gave a separable mixture of
(E)-17 and (Z)-17. Cleavage of the benzyl group in (E)-17
was performed with DDQ[31] in wet CH2Cl2 and yielded lac-
tone 2 in 51% yield.[32,33] The synthetic compound showed
physical and spectral properties identical to those published
for aspergillide B,[1] as also found by Hande and Uenishi.[5]
Even though the yield of the CM step 5 ! 16 was good
(89%), both the stereoselectivity of this step and the chemi-
cal yield of the macrolactonization step 16 ! 17 were not as
satisfactory as desired. For this reason, we decided to inves-
tigate whether the change of order of these two steps might,
despite our previous concern, lead to higher yields of lac-
tone (E)-17. The results of these efforts are shown in
Scheme 4. Yamaguchi esterification[30] of acid 5 with alcohol
4 gave ester 18 in 75% yield. RCM of this ester catalyzed
by the first-generation Grubbs catalyst Ru-I led to a mixture
of lactones (E+Z)-17 with predominance of the undesired
Z isomer. When catalyst Ru-II was used, two lactones were
formed, too, together with decomposition products. Surpris-
ingly, while one of the lactones was (Z)-17, the other was
found to have structure 19 (~85:15 E/Z mixture), a product
of double bond migration (from D8,9 to D9,10, aspergillide
numbering). The longer the reaction time, the higher the
percentage of 19 and also of decomposition products. Lac-
tone (E)-17 was formed, if at all, in a very small amount.
C-bond formation) or, conversely, through initial cross meta-
thesis (CM) followed by intramolecular C O-bond forma-
ꢀ
tion (macrolactonization). We opted for the latter alterna-
tive because of the stereochemical uncertainties of the E/Z
stereoselectivity in the RCM step, CM being expected to
give a higher proportion of the required E configuration of
the olefinic bond.[12,13]
Scheme 2. Retrosynthetic analysis of 2, later shown to be aspergillide B.
Bn = benzyl; TPS = tert-butyldiphenylsilyl; TES = triethylsilyl.
We planned the stereoselective preparation of the 2,6-
trans-disubstituted tetrahydropyran[14–16] 5 by means of a
Mukaiyama-type C-glycosidation[17] of a suitable lactol de-
rivative prepared through reduction of lactone 6. The latter
was to be obtained by functional modification of 7, in turn
derived from 8 through a metal-catalyzed double bond mi-
gration.[18] Following this, retrosynthetic asymmetric allyla-
676
www.chemeurj.org
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 675 – 688

Aspergillide A and B
FULL PAPER
Scheme 4. Attempts at an improved preparation of intermediate (E)-17.
to 2, a cis-2,6-disubstituted tetrahydropyran moiety. Under
the assumption that aspergillide A and B could possibly be
epimeric at C-7 (Scheme 1), we tried to adapt to the prepa-
ration of aspergillide A the same route that was successful
for aspergillide B (Scheme 3). Thus, aldehyde 9 was subject-
ed as previously to Brown’s asymmetric allylation but now
with the chiral allylborane prepared from allylmagnesium
bromide and (+)-Ipc2BCl (Scheme 5). This gave secondary
alcohol 20, with the opposite configuration at C-7 (aspergil-
Scheme 3. Synthesis of aspergillide B (2). Ipc = isopinocampheyl; TBAF
= tetra-n-butylammonium fluoride; DMAP = 4-(N,N-dimethylamino)-
pyridine; Tf
=
trifluoromethanesulfonyl; TMS
=
trimethylsilyl;
TEMPO = 2,2,6,6-tetramethylpiperidine-1-oxyl; DIBAL = diisobutyl-
aluminum hydride; DDQ = 2,3-dichloro-5,6-dicyano-1,4-benzoquinone.
Furthermore, when subjected to RCM reaction conditions
with Ru-II, lactone (Z)-17 was progressively converted into
(E+Z)-19. These facts suggest that lactone 19 is thermody-
namically more stable than either (E)-17 or (Z)-17, a con-
clusion supported by theoretical calculations.[34] Indeed, such
double-bond migrations and other various “non-metathesis”
side reactions in Ru-catalyzed processes have been attribut-
ed to the in situ formation of ruthenium hydrides, this being
more frequent with second-generation catalysts like Ru-
II.[35] In any case, this alternative route does not lead to an
improvement in the yield of the required (E)-17.
When we initiated the synthesis of aspergillide A, we
were still unaware of the actual structure of this natural
macrolide. A comparative study of the NMR data of several
of our synthetic compounds with those of the natural prod-
uct suggested that aspergillide A might contain, in contrast
Scheme 5. Initial route towards possible structures of aspergillide A.
Chem. Eur. J. 2011, 17, 675 – 688
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
677

M. Carda, J. A. Marco et al.
lide numbering) as compared with 12 (Scheme 3). Alcohol
20 was then subjected to the same reaction sequence depict-
ed in Scheme 3 for 12. However, this led to a disappoint-
ment, as the C-glycosidation step provided in 70% overall
yield a mixture 26a/26b in which the targeted cis-2,6-disub-
stituted tetrahydropyran 26b was the minor component.
Variation of the reaction conditions did not bring much
change in the diastereomeric ratio.
Just at this moment, we learned the results of X-ray dif-
fraction analyses[7] of aspergillide A, which showed that the
compound had actually structure 1 (Scheme 1). Thus, asper-
gillides A and B are epimeric at C-3, not at C-7. This appa-
rently minor stereochemical difference between 1 and 2 af-
fects the retrosynthetic concept in a crucial way, as it in-
volves the methodology needed to create the relative config-
uration of the stereocenters C-3/C-7.[14–16]
The retrosynthetic analysis as depicted in Scheme 6 relies
on that proposed for aspergillide B (2).[6] Retrosynthetic
opening of the lactone ring by means of hydrolytic cleavage
of the ester moiety and olefin metathesis[12,13] gives rise to
the known alcohol 5[11] as well as to acid 27. Although the
results depicted in Scheme 5 suggested that the formation of
the tetrahydropyran ring[14] of 27 through a stereocontrolled
C-glycosidation[17] of a lactol derivative prepared via reduc-
tion of d-lactone 7[6] was not a promising way, we made
some attempts in this direction. However, all attempts to
obtain 27 or a similar derivative using a Mukaiyama-type C-
glycosidation on a lactol derivative prepared from 6 were
not satisfactory. Mixtures of cis-/trans-2,6-disubstituted tetra-
hydropyrans were always obtained, with the desired cis
isomer never being the major compound.[36]
Scheme 7. Synthesis of aspergillide A (1). LiHMDS = lithium hexame-
thyldisilazide.
provided 29, which displayed an exclusively cis relationship
at the stereocenter pair C-3/C-7 (aspergillide numbering,
Scheme 1), as demonstrated by the observation of a NOE
between H-3 and H-7.[9a] As the precursor lactone 6, 29 was
also a ~9:1 E/Z mixture. Ester 29 was then converted into
aspergillide A (1) along a reaction sequence described in
our previous paper.[9a]
Some particular aspects of the syntheses disclosed above
deserve comment. One of these aspects is the stereochemi-
cal features observed in the formation of tetrahydropyran
rings (Scheme 8). C-Glycosidation of the lactol derivative 14
gives rise mainly to tetrahydropyran 15a (Scheme 3), where
the acetic acid chain and the propenyl group are trans with
respect to the ring. As regards to lactol derivative 25, epi-
meric of 14 at C-7 (aspergillide numbering), the same reac-
tion conditions give rise mainly to tetrahydropyran 26a,
where the acetic acid moiety and the propenyl group are
also trans with respect to the ring. Furthermore, treatment
of lactol 28 with Et3SiH/BF3·Et2O gives only tetrahydropyr-
an 29 (Scheme 8), where the acetic acid moiety and the pro-
penyl group are cis with respect to the ring. In contrast,
lactol 30, obtained from lactone 24 (Scheme 5), was found
to yield under the same reaction conditions tetrahydropyran
31 (Scheme 8), with the acetic acid moiety and the propenyl
group trans with respect to the ring.
These results should be examined in the light of previous
findings of Kishi and co-workers,[15a] as well as of other
groups. According to the common view, Mukaiyama-type C-
glycosidations and reductions of lactols with Et3SiH, both
Lewis-acid mediated reactions, take place via a kinetically
controlled, nucleophilic attack of the enol silane and the sili-
con hydride, respectively, on an intermediate cyclic oxocar-
benium cation.[37,38] The reactions are markedly exergonic
and are thus believed to occur through reactant-like transi-
tion states.[39] Thus, lactol 28 and lactol acetate 14, both de-
rived from d-lactone 6 (Scheme 9), are transformed into the
corresponding oxocarbenium cations where the main half-
chair conformations, A and C, respectively, display the pro-
penyl and benzyloxy groups in the energetically more favor-
able pseudoequatorial positions. Furthermore, these confor-
mations are favored by the pseudoaxial hydrogen at C-2, as
Scheme 6. Retrosynthetic analysis of aspergillide A (1).
In the strategy we have followed until now for the conver-
sion of the d-lactone moiety into a tetrahydropyran ring, we
first introduced the hydrogen atom (DIBAL reduction to
the lactol) and then the acetic acid side chain (Mukaiyama
C-glycosidation). We then investigated the result of a
change in the order of steps, that is, by introducing first the
acetic acid chain and then the hydrogen atom. In accordance
with this idea, we finally reached success in the way shown
in Scheme 7. Addition of the lithium enolate of ethyl acetate
to lactone 6 at low temperature gave lactol 28 as a mixture
of stereoisomers at the olefinic bond and at the hemiketal
carbon. Treatment of this mixture with Et3SiH/BF3·Et2O[15a,b]
a sCꢀ bond is a better hyperconjugative donor to the C=O+
H
bond than a sCꢀ bond.[39a] It may be expected that reactions
O
occur preferentially via axial attack of the nucleophile on
678
www.chemeurj.org
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 675 – 688

Aspergillide A and B
FULL PAPER
tinuum of mechanisms which range from SN2-like to SN1-
like processes.[39e,g] Key-influencing factors are the type of
nucleophile and Lewis acid used. Low diastereoselectivities
may be due to competition between these two mechanisms,
this being in turn related to the formation of ion pairs[39g,i]
with counteranions formed from the leaving group or else to
nucleophilic attacks taking place at the diffusion limit. Even
the possibility of nonaxial attacks to the oxocarbenium ion
cannot be excluded, either.[39e,g] In the present case, the ap-
preciable electronic and steric differences between the enol
silane, a strong p-nucleophile,[39g,40] and the weak nucleo-
phile Et3SiH may play a relevant role in the stereochemical
outcome of the reactions discussed here. Another factor
worth considering is the fact that the C-glycosidation is car-
ried out at ꢀ188C whereas the lactol reduction is performed
at ꢀ808C.[41] Furthermore, differences in solvent polarity
(MeCN in the C-glycosidation vs CH2Cl2 in the lactol reduc-
tion) may also play a role, as observed in previous instan-
ces.[42]
The same reactions were carried out on the corresponding
lactol derivatives prepared from lactone 24, epimeric of 6 at
C-7. Lactol 30, the epimer of 28, gave the trans-2,6-substitut-
ed tetrahydropyran 31 as the sole reaction product. As re-
gards to lactol acetate 25, the epimer of 14, it gave mainly
the trans-2,6-substituted tetrahydropyran 26a under the
same C-glycosidation conditions as above. In order to ex-
plain the formation of 26a and its minor isomer 26b, we
may assume that axial attack of the enol silane takes place
on an oxocarbenium ion having the two half-chair confor-
mations G and H, both showing a pseudoaxial and a pseu-
doequatorial substituent. According to the previous reason-
ing,[39] the former should be slightly predominant (pseudoax-
ial hydrogen at C-2) but, at the same time, less reactive due
to the developing 1,3-diaxial interaction between the incom-
ing nucleophile and the 1-propenyl group in the transition
state. Apparently, these two factors compensate each other,
as no high stereoselectivity is observed (d.r. ~1.9:1).
More surprising, and as yet unexplained, is the stereo-
chemical outcome of the silane reduction of lactol 30, where
only the trans-2,6-substituted tetrahydropyran 31 is
formed.[43] On the basis of axial hydride attack on an inter-
mediate oxocarbenium cation, only conformation E can ex-
plain the observed result. As commented above, E should
be slightly predominant but less reactive than F. No convinc-
ing reasons therefore can be presented for the observed ste-
reochemical outcome but, once again, differences in temper-
ature and solvent[41,42] between both reaction types may
have an influence.
Scheme 8. Comparison of stereochemical outcomes in the formation of
C-glycosides carried out with different methods on different substrates.
either of these two conformations, rather than on the alter-
native half-chair conformations B and D, where the same
groups are in a pseudoaxial orientation. Axial attack at C-1
in conformations A and C may appear counterintuitive at
first sight, as it takes place syn to the benzyloxy group and
gives rise to a developing gauche interaction between the
latter and the nucleophile. However, such an interaction is
not considered to be quantitatively very important in the
case of a benzyloxy group.[39]
All these considerations fit well with the result of the re-
duction of lactol 28 with Et3SiH/BF3·Et2O, where the cis-2,6-
substituted tetrahydropyran 29 was the sole product isolat-
ed. Since reactions mediated by BF3·Et2O are believed to
occur via an SN1-like mechanism, oxocarbenium ion with
conformation A is assumed to be the key intermediate. In
fact, no product formed through axial attack on conforma-
tion B is detected. Not only ground state but also transition-
state effects (Curtin–Hammett kinetics) disfavor the last
possibility, as axial attack on B would develop a 1,3-diaxial
interaction.[39] Likewise, enol silane attack on the corre-
sponding oxocarbenium cation C leads to the trans-2,6-sub-
stituted tetrahydropyran 15a as the major product. Accord-
ingly, the minor cis-2,6-substituted tetrahydropyran 15b may
be formed through a pathway via conformation D, which is
less favorable for the same reasons discussed above with B.
This mechanistic rationale notwithstanding, it is not yet
clear why the C-glycosidation is much less stereoselective
than the lactol reduction with triethylsilane. Various explan-
ations have been advanced to explain such situations in nu-
cleophilic substitutions of cyclic acetals. It has been pro-
posed that this type of reactions takes place through a con-
Another aspect worth discussing is the result of some of
the metathesis reactions, those depicted in Schemes 4 and
10, the latter showing the key RCM step of the synthetic se-
quence leading to 1.[9a] The very high stereoselectivity of the
conversion of 32 into (Z)-33 (Scheme 10) deserves some
mention since the stereochemical outcome of RCM reac-
tions that give rise to medium-sized or large rings (ꢁ10
atoms)[44] is not always predictable with full certainty. In the
present case, molecular mechanics calculations showed that
Chem. Eur. J. 2011, 17, 675 – 688
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
679

M. Carda, J. A. Marco et al.
Scheme 9. Mechanistic discussion of the reactions depicted in Scheme 8 (SN1-like processes with formation of intermediate oxocarbenium ions are as-
sumed).
namically controlled. However, even though the macrolide
ring of 1 is formally 14-membered from the point of view of
the atom periphery and thus relatively flexible, the presence
of the tetrahydropyran ring may impose certain conforma-
tional restraints which make the formation of (E)-33 kineti-
cally disfavored. Thus, the formation of (Z)-33 in the RCM
may also be kinetically controlled as well. Indeed, com-
Scheme 10. Key RCM step in the synthesis of 1 (ref. [9a]).
pound (Z)-33 is obtained in very high yield when using cata-
(Z)-33 is much more stable than (E)-33.[34] This suggests that
the formation of (Z)-33 in the RCM process is thermody-
lyst Ru-I, which is assumed to work under kinetic condi-
tions.[12] The more active, second-generation catalyst Ru-II
680
www.chemeurj.org
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 675 – 688

Aspergillide A and B
FULL PAPER
gave also (Z)-33 but in a lower yield (67%). Catalyst Ru-II
is known to cause in some cases E/Z isomerizations,[35] thus
approaching thermodynamic (equilibrium) conditions. Fur-
thermore, compound (E)-33 may show some instability
under these conditions, what could explain the eroded yield
observed when using Ru-II. Finally, it also worth noting
that, in contrast to that observed with ester 18 (Scheme 4),
no C=C bond migration took place in the RCM of 32 with
catalyst Ru-II. This may be related to the fact that, accord-
ing to our theoretical calculations, C=C bond migration
leads in the present case to a less stable compound.[34]
cell types, with aspergillide B (2) being at least 1.5 times
more active throughout the entire panel. Consistent in all
series is also the fact that the HL-60 (human leukemia) cell
line showed the highest sensitivity to treatment with the dif-
ferent aspergillide analogues, whereas the non-transformed
endothelial cells (BAE) showed the lowest sensitivity to the
tested compounds.
Most of the tested non-natural aspergillide derivatives
were found to be more active than the natural metabolites 1
and 2. The best results were obtained with compound 34,
which showed the highest cytotoxicity for all cell types as-
sayed, most particularly for human leukemia (18 times more
cytotoxic than aspergillide B and 45 times more cytotoxic
than aspergillide A) and breast cancer cells (24 times more
cytotoxic than aspergillide B and 34 times more cytotoxic
than aspergillide A). In the case of aspergillide B (2), the
saturation of the olefinic bond (compound 35) leads to a
small decrease of the cytotoxicity in a degree which depend-
ed of the cell line tested.
Another aspect to note is the influence of the O-benzyl
group. For aspergillide B, its cytotoxicity is very similar to
those of its benzylated derivative (E)-17 and the stereoiso-
mer (Z)-17 (except in the case of the HL-60 human leuke-
mia cell line, where the two benzylated derivatives are 2.5–3
times more active than 2). In contrast, the benzylated deriv-
ative of aspergillide A, (E)-33, and its stereoisomer (Z)-33,
are at least three times more active than the parent lactone
1 against all cell lines tested, and in some cases much more.
The observation that all tested compounds showed their
highest activity against leukemia cells suggests a potential
usefulness as antileukemia drugs, most particularly in the
case of lactone 34, which shows a cytotoxic activity compa-
rable with that of the clinical drug fludarabine.[45] In view of
the potential interest of this and other aspergillide analogues
for cancer treatment, we are now preparing a series of such
compounds in order to establish more extended structure–
activity relationships.
Biological assays: Aspergillides A (1) and B (2) have been
described as moderately cytotoxic against murine lymphoma
L1210 cells.[1] A more detailed characterization of their ac-
tivity and selectivity profile, however, is still missing. In
order to further characterize the biological profile of these
two natural macrolides, their cytotoxicity has now been ex-
amined against five different human cancer cell lines: pro-
myelocytic leukemia, breast carcinoma, fibrosarcoma, colon
adenocarcinoma and osteosarcoma, as well as in a primary
culture of non-transformed bovine aorta endothelial cells. In
addition, some of the synthetic intermediates prepared in
the courses of the syntheses have also been tested. Lactone
34 is the Z stereoisomer of aspergillide A,[9a] while lactone
35 is the dihydro derivative of aspergillide B (its preparation
is described in the Experimental Section). The results of the
cytotoxicity tests are shown in Table 1.
The data of Table 1 indicate that aspergillide A (1) is the
compound that exhibits the lowest cytotoxicity for all the
Experimental Section
General methods: 1H/13C NMR spec-
tra were recorded at 500/125 MHz in
the indicated solvent at 308C. Signals
of the deuterated solvent (CDCl3 in all
but one case) were taken as the refer-
ence (in CDCl3, the singlet at d 7.27
and the triplet centered at 77.0 ppm
for 1H and 13C NMR, respectively; in
C6D6, the singlet at d 7.16 and the trip-
let centered at 128.0 ppm for 1H and
13C NMR, respectively). Carbon atom
types (C, CH, CH2, CH3) were deter-
mined with the DEPT pulse sequence.
1H and 13C signals were assigned with
the aid of 2D methods. Mass spectra
were run by the electron impact mode
(EIMS, 70 eV), by the fast atom bom-
bardment mode (FABMS, m-nitroben-
zyl alcohol matrix) or by the electro-
Table 1. Cytotoxic activity IC50 [mgmLꢀ1] of macrolides 1, 2 and some synthetic intermediates against several
cell types.[a]
Cell lines[b]
Compound
HL-60
MDA-MB-231
HT-1080
HT-29
U2OS
BAE
1
2
81.2ꢂ17.5
32.8ꢂ7.6
12.3ꢂ1.5
11.4ꢂ0.2
12.3ꢂ0.3
6.7ꢂ0.9
99.0ꢂ7.9
68.2ꢂ6.5
46.7ꢂ5.7
77.4ꢂ19.0
18.5ꢂ2.6
16.2ꢂ2.2
2.9ꢂ0.2
84.3ꢂ13.7
45.0ꢂ4.5
48.1ꢂ2.7
60.3ꢂ2.1
27.9ꢂ2.9
19.7ꢂ3.3
6.5ꢂ0.4
107.7ꢂ9.3
64.3ꢂ1.5
54.6ꢂ5.1
51.3ꢂ2.8
31.1ꢂ2.9
20.4ꢂ0.8
6.7ꢂ0.8
>100
>100
54.5ꢂ6.6
53.8ꢂ4.1
72.8ꢂ5.2
22.8ꢂ3.7
18.3ꢂ3.5
4.2ꢂ0.9
>100
60.6ꢂ6.5
33.4ꢂ3.0
41.7ꢂ4.7
34.3ꢂ2.2
21.5ꢂ2.4
7.7ꢂ1.3
>100
(E)-17
(Z)-17
(E)-33
(Z)-33
34
1.8ꢂ0.2
35
62.8ꢂ10.9
71.2ꢂ10.2
92.3ꢂ13.3
95.0ꢂ8.7
[a] Cells were incubated for 72 h in the presence of each compound, and the ratios of viable cells were deter-
mined by MTT assay. The drug concentration required to inhibit cell growth by 50% was determined from
semilogarithmic dose-response plots, and results represent the means + SDs of three independent experi-
ments with triplicate samples each. [b] The cell lines used were: HL-60 (human promyelocytic leukemia),
MDA-MB-231 (human breast carcinoma), HT1080 (human fibrosarcoma), HT29 (human colon adenocarcino-
ma), U2OS (human osteosarcoma), and the non-transformed bovine aorta endothelial (BAE) cells.
Chem. Eur. J. 2011, 17, 675 – 688
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
681

M. Carda, J. A. Marco et al.
spray mode (ESMS). IR data are given only for compounds with signifi-
cant functions (OH, C=O) and were recorded as oily films on NaCl
plates (oils) or as KBr pellets (solids). Optical rotations were measured
at 258C. Reactions which required an inert atmosphere were carried out
under N2 with flame-dried glassware. Et2O and THF were freshly dis-
tilled from sodium/benzophenone and transferred via syringe. Dichloro-
methane was freshly distilled from CaH2. Tertiary amines were freshly
distilled from KOH. Toluene was freshly distilled from sodium wire.
Commercially available reagents were used as received. Unless detailed
otherwise, “work-up” means pouring the reaction mixture into brine, fol-
lowed by extraction with the solvent indicated in parenthesis. If the reac-
tion medium was acidic, an additional washing with 5% aq NaHCO3 was
performed. If the reaction medium was basic, an additional washing with
aq NH4Cl was performed. New washing with brine, drying over anhy-
drous Na2SO4 and elimination of the solvent under reduced pressure
were followed by chromatography on a silica gel column (60–200 mm)
and elution with the indicated solvent mixture. Where solutions were fil-
tered through a Celite pad, the pad was additionally washed with the
same solvent used, and the washings incorporated to the main organic
layer.
(200 mL), MeOH (200 mL) and 30% H2O2 (100 mL). After stirring for
30 min at room temperature, the mixture was poured onto satd. aq
NaHCO3 and worked up (extraction with EtOAc). Column chromatogra-
phy on silica gel (hexane/EtOAc 9:1) afforded 12, still contaminated with
boron-containing side products, which was used as such in the next step.
An aliquot was carefully purified for analytical purposes. Colourless oil.
1
[a]D = ꢀ16.7 (c = 1.6, CHCl3); H NMR: d = 7.75–7.70 (m, 4H), 7.45–
7.30 (brm, 11H), 5.83 (ddt, J=17.2, 10.3, 7 Hz, 1H), 5.15–5.10 (m, 2H),
4.69 (d, J=11.8 Hz, 1H), 4.54 (d, J=11.8 Hz, 1H), 3.82 (dd, J=10.5,
5.6 Hz, 1H), 3.70 (dd, J=10.5, 5 Hz, 1H), 3.65–3.60 (m, 1H), 3.60–3.55
(m, 1H), 2.30–2.25 (m, 1H), 2.20–2.15 (m, 1H), 2.00 (brs, 1H, OH),
1.80–1.75 (m, 1H), 1.70–1.60 (m, 2H), 1.50–1.45 (m, 1H), 1.10 ppm (s,
9H); 13C NMR: d = 138.7, 133.5 (x 2), 19.2 (C), 135.6 (ꢄ4), 135.0, 129.6
(ꢄ2), 128.3 (ꢄ2), 127.8 (ꢄ2), 127.7 (ꢄ4), 127.5, 79.8, 70.8 (CH), 117.8,
72.1, 66.1, 41.9, 32.6, 27.8 (CH2), 26.9 ppm (ꢄ3) (CH3); IR: nmax
=
3420 cmꢀ1 (br, OH); HR FABMS: m/z: calcd for C31H40O3Si: 488.2746;
found: 488.2730 [M+].
AHCTUNGTRENGN(UN 4R,7S)-7-Benzyloxy-8-(tert-butyldiphenylsilyloxy)-4-(triethylsilyloxy)oct-
1-ene (8): The alcohol 12 obtained above was dissolved in dry CH2Cl2
(400 mL), cooled to ꢀ808C and treated dropwise with Et3N (12.5 mL,
90 mmol) and TESOTf (17 mL, 75 mmol). The mixture was stirred for
15 min at the same temperature and then for 2 h at room temperature.
Work-up (extraction with CH2Cl2) was followed by column chromatogra-
phy of the residue on silica gel (hexanes/Et2O 49:1) to yield 8 (13.93 g,
77% overall from 9) as a colourless oil. [a]D = ꢀ9 (c = 1.65, CHCl3);
1H NMR: d = 7.75–7.70 (m, 4H), 7.45–7.30 (brm, 11H), 5.82 (ddt, J=
17.2, 10.3, 7 Hz, 1H), 5.10–5.00 (m, 2H), 4.67 (d, J=11.7 Hz, 1H), 4.52
(d, J=11.7 Hz, 1H), 3.78 (dd, J=10.7, 5.4 Hz, 1H), 3.72 (brquint, J
~6 Hz, 1H), 3.70 (dd, J=10.7, 4.7 Hz, 1H), 3.50 (m, 1H), 2.24 (dd, J=7,
6 Hz, 2H), 1.80–1.70 (m, 1H), 1.70–1.60 (m, 1H), 1.60–1.50 (m, 1H),
1.45–1.35 (m, 1H), 1.10 (s, 9H), 0.97 (t, J=7.7 Hz, 9H), 0.62 ppm (q, J=
(S)-5-Benzyloxy-6-(tert-butyldiphenylsilyloxy)-hex-1-ene (11): A solution
of alcohol 10 (23.04 g, 65 mmol) in dry Et2O (250 mL) was treated under
N2 with benzyl bromide (35.7 mL, 300 mmol) and Ag2O (45 g, ca.
195 mmol). The reaction mixture was then protected from light and
stirred for 3 d at room temperature. After this time, the mixture was fil-
tered through Celite, and the volatiles were removed under reduced pres-
sure. Column chromatography of the residue on silica gel (hexanes/tolu-
ene 7:3) furnished benzyl ether 11 (24 g, 83%) as a colourless oil. [a]D
=
ꢀ20.6 (c = 0.86, CHCl3); 1H NMR: d = 7.70–7.65 (m, 4H), 7.45–7.30
(brm, 11H), 5.83 (ddt, J=17.2, 10.3, 6.5 Hz, 1H), 5.05–4.95 (m, 2H), 4.70
(d, J=11.7 Hz, 1H), 4.54 (d, J=11.7 Hz, 1H), 3.81 (dd, J=10.5, 5.5 Hz,
1H), 3.71 (dd, J=10.5, 5.5Hz, 1H), 3.58 (m, 1H), 2.30–2.10 (brm, 2H),
1.75–1.65 (brm, 2H), 1.11 ppm (s, 9H); 13C NMR: d = 139.0, 133.6 (ꢄ2),
19.2 (C), 138.7, 135.6 (ꢄ4), 129.7 (ꢄ2), 128.3 (ꢄ2), 127.8 (ꢄ2), 127.7 (ꢄ4),
127.5, 79.2 (CH), 114.6, 72.2, 66.2, 31.0, 29.7 (CH2), 26.9 ppm (ꢄ3) (CH3);
HR FABMS: m/z: calcd for C29H36O2Si: 444.2484; found: 444.2465 [M+].
7.7 Hz, 6H); 13C NMR: d
= 139.0, 133.6 (ꢄ2), 19.2 (C), 135.6 (ꢄ4),
135.2, 129.6 (ꢄ2), 128.2 (ꢄ2), 127.7 (ꢄ2), 127.6 (ꢄ4), 127.3, 80.1, 72.2
(CH), 116.8, 72.0, 66.2, 42.2, 32.7, 27.7, 5.1 (ꢄ3) (CH2), 26.9 (ꢄ3),
7.0 ppm (ꢄ3) (CH3); HR FABMS: m/z: calcd for C37H53O3Si2: 601.3533;
found: 601.3520 [MꢀH+].
(S)-4-(Benzyloxy)-5-(tert-butyldiphenylsilyloxy)pentanal (9): A solution
of olefin 11 (22.23 g, 50 mmol) in CH2Cl2 (500 mL) was cooled to ꢀ808C.
A stream of ozone-containing air was then bubbled through the solution
until complete consumption of the starting material (ca. 30 min, TLC
monitoring). Ozone residues were then eliminated by bubbling a stream
of N2 for 5 min and the mixture was allowed to reach room temperature.
After addition of PPh3 (26.23 g, 100 mmol), the mixture was then stirred
for 2 h. After solvent removal under reduced pressure, the crude residue
was stirred for 15 min under pentane (3ꢄ300 mL) and filtered. The solu-
tion was then concentrated under reduced pressure and the crude residue
was purified by means of column chromatography on silica gel (pentane/
Et2O 9:1). This provided aldehyde 9 (17.42 g, 78%) as a colourless oil.
[a]D = ꢀ26.3 (c = 0.86, CHCl3); 1H NMR: d = 9.70 (brs, 1H), 7.70–
7.65 (m, 4H), 7.45–7.30 (brm, 11H), 4.65 (d, J=11.6 Hz, 1H), 4.48 (d,
J=11.6 Hz, 1H), 3.82 (dd, J=10.7, 5.2 Hz, 1H), 3.70 (dd, J=10.7, 5.2 Hz,
1H), 3.55 (m, 1H), 2.49 (brt, J=7.3 Hz, 2H), 2.05–1.95 (m, 1H), 1.90–
1.80 (m, 1H), 1.11 ppm (s, 9H); 13C NMR: d = 138.4, 133.3, 133.2, 19.2
(C), 202.2, 135.6 (ꢄ4), 129.7 (ꢄ2), 128.3 (ꢄ2), 127.8 (ꢄ2), 127.7 (ꢄ4),
127.5, 78.5 (CH), 72.0, 65.7, 39.8, 24.3 (CH2), 26.8 ppm (ꢄ3) (CH3); IR:
ACHTUNGTRNE(NUNG 4R,7S)-7-Benzyloxy-8-(tert-butyldiphenylsilyloxy)-4-(triethylsilyloxy)oct-
2E,Z-ene (7): N-Tritylallylamine (12 g, 40 mmol) and catalyst Ru-II
(850 mg, 1 mmol) were dissolved under N2 in dry, deoxygenated CH2Cl2
(50 mL). Following this, a solution of compound 8 (12.06 g, 20 mmol) and
iPr2NEt (3.5 mL, 20 mmol) in dry, deoxygenated CH2Cl2 (350 mL) were
added dropwise under N2, and the mixture was heated at reflux for 16 h.
After removal of all volatiles under reduced pressure, the residue was pu-
rified by means of column chromatography on silica gel (hexanes/Et2O
49:1) to give 7 (11.58 g, 96%) as a ca. 9:1 E/Z mixture as a colourless oil.
1H NMR (signals from the major E isomer): d =7.75–7.70 (m, 4H), 7.45–
7.30 (brm, 11H), 5.60–5.50 (m, 1H), 5.50–5.40 (m, 1H), 4.68 (d, J=
11.7 Hz, 1H), 4.56 (d, J=11.7 Hz, 1H), 4.04 (m, 1H), 3.79 (brdd, J=
10.8, 5.5 Hz, 1H), 3.70 (brdd, J=10.8, 4.8 Hz, 1H), 3.54 (brm, 1H), 1.70
(brd, J ~6.3 Hz, 3H), 1.70–1.55 (brm, 4H), 1.11 (s, 9H), 0.97 (t, J=
7.7 Hz, 9H), 0.62 ppm (q, J=7.7 Hz, 6H); 13C NMR (signals from the
major E isomer): d = 139.0, 133.6 (ꢄ2), 19.2 (C), 135.6 (ꢄ4), 134.8, 129.6
(ꢄ2), 128.2 (ꢄ2), 127.7 (ꢄ2), 127.6 (ꢄ4), 127.3, 125.2, 80.1, 73.9 (CH),
72.0, 66.2, 34.2, 27.7, 5.1 (ꢄ3) (CH2), 26.8 (ꢄ3), 17.5, 6.9 ppm (ꢄ3) (CH3);
HR FABMS: m/z: calcd for C37H53O3Si2: 601.3533; found: 601.3544
[MꢀH+].
nmax
446.2277; found: 446.2299 [M+].
(4R,7S)-7-Benzyloxy-8-(tert-butyldiphenylsilyloxy)-oct-1-en-4-ol (12): Al-
=
1710 cmꢀ1 (C=O); HR FABMS: m/z: calcd for C28H34O3Si:
AHCTUNGRETG(NNUN 2S,5R)-2-(Benzyloxy)-oct-6E,Z-ene-1,5-diol (13): A solution of com-
G
pound 7 (10.8 g, ca. 18 mmol) in dry THF (180 mL) was treated at room
temperature under N2 with solid TBAF trihydrate (12.6 g, 40 mmol). The
mixture was stirred for 16 h at room temperature. After removal of all
volatiles under reduced pressure, the residue was purified by means of
column chromatography on silica gel (hexanes/EtOAc 1:1) to yield 13
(4.41 g, 98%) as a ~9:1 E/Z mixture: colourless oil. 1H NMR (signals
from the major E isomer): d = 7.30–7.20 (brm, 5H), 5.60–5.50 (m, 1H),
5.40–5.35 (m, 1H), 4.50 (AB system, J=11.7 Hz, 2H), 3.92 (m, 1H), 3.59
(brdd, J=11.3, 2 Hz, 1H), 3.50–3.40 (brm, 2H), 2.70 (brs, 1H, OH), 2.50
(brs, 1H, OH), 1.62 (brd, J ~6.3 Hz, 3H), 1.60–1.45 ppm (brm, 4H);
13C NMR (signals from the major E isomer): d = 138.3 (C), 133.9, 128.3
lylmagnesium bromide (commercial 1m solution in Et2O, 37.5 mL,
37.5 mmol) was added dropwise under N2 at 08C via syringe to a solution
of (ꢀ)-Ipc2BCl (14.4 g, 45 mmol) in dry Et2O (250 mL). The mixture was
further stirred for 1 h at 08C. The solution was then allowed to stand,
which caused precipitation of magnesium chloride. The supernatant solu-
tion was then carefully transferred to another flask via cannula. After
cooling this flask at ꢀ908C, a solution of aldehyde 9 (13.4 g, 30 mmol) in
dry Et2O (60 mL) was added dropwise via syringe. The resulting solution
was further stirred at the same temperature for 2 h. The reaction mixture
was then quenched through addition of phosphate pH 7 buffer solution
682
www.chemeurj.org
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 675 – 688

Aspergillide A and B
FULL PAPER
(ꢄ2), 127.7 (ꢄ2), 127.6, 126.6, 79.6, 72.7 (CH), 71.5, 63.8, 32.8, 26.7
(CH2), 17.6 ppm (CH3); IR: nmax = 3400 cmꢀ1 (br, OH); HR ESMS: m/z:
calcd for C15H22O3Na: 273.1467; found: 273.1466 [M+Na+].
11.7 Hz, 1H), 4.50–4.40 (m, 2H), 4.20 (m, 1H), 3.53 (m, 1H), 2.90 (dd,
J=15.2, 8.4 Hz, 1H), 2.75 (dd, J=15.2, 5.3 Hz, 1H), 1.95–1.70 (brm,
3H), 1.69 (brd, J ~6.4 Hz, 3H), 1.45 (s, 9H), 1.45–1.40 ppm (m, 1H);
13C NMR (signals from the major E isomer): d = 198.1, 138.3, 48.0 (C),
130.6, 128.3 (ꢄ2), 127.7, 127.6 (ꢄ2), 127.5, 73.5, 70.8, 70.7 (CH), 70.6,
43.4, 27.3, 23.5 (CH2), 29.8 (ꢄ3), 17.8 ppm (CH3); IR: nmax = 1682 cmꢀ1
(C=O); HR ESMS: m/z: calcd for C21H30O3SNa: 385.1813; found:
385.1808 [M+Na+]. The NOE indicated above between H-7 (aspergil-
lide numbering) and the acetate side-chain methylene is diagnostic of the
relative configuration at the pair of stereocenters C-3/C-7 in this mole-
cule.
ACHTUNGTRENNUNG(3S,6R)-3-(Benzyloxy)-6-(prop-1E,Z-enyl)tetrahydro-2H-pyran-2-one (6):
A solution of diol 13 (3.75 g, 15 mmol) in dry CH2Cl2 (150 mL) was
cooled under N2 to 08C and treated sequentially with TEMPO (470 mg,
3 mmol) and then, portionwise, with PhIACTHNUTRGNENUG(OAc)2 (10.6 g, 33 mmol). The
mixture was stirred for 16 h at room temperature. The reaction was then
quenched by addition of saturated aqueous Na2S2O3 (20 mL) and worked
up (extraction with CH2Cl2). CAUTION: evaporation of volatiles under
reduced pressure has to be performed at room temperature, due to the
appreciable volatility of lactone 6. Column chromatography on silica gel
(pentane/Et2O 9:1) furnished 6 (3.1 g, 84%) as a ~9:1 E/Z mixture as a
15b: colourless oil; 1H NMR (signals from the major E isomer): d =
7.35–7.25 (brm, 5H), 5.70 (dqd, J=15.7, 6.4, 1.5 Hz, 1H), 5.47 (ddq, J=
1
colourless oil. H NMR (signals from the major E isomer): d = 7.40–7.25
(brm, 5H), 5.77 (dqd, J=15.6, 6.5, 1 Hz, 1H), 5.49 (ddq, J=15.6, 6.7,
1.5 Hz, 1H), 4.94 (m, 1H), 4.92 (d, J=11.8 Hz, 1H), 4.66 (d, J=11.8 Hz,
1H), 3.96 (dd, J=7.5, 6 Hz, 1H), 2.20–1.95 (brm, 3H), 1.72 (ddd, J=6.5,
1.5, 1 Hz, 3H), 1.70 ppm (m, 1H); 13C NMR (signals from the major E
isomer): d = 170.7, 137.4 (C), 129.7, 128.9, 128.4 (ꢄ2), 128.0 (ꢄ2), 127.8,
80.3, 73.6 (CH), 72.6, 27.4, 26.3 (CH2), 17.6 ppm (CH3); IR: nmax
=
1744 cmꢀ1 (C=O); HR EIMS: m/z (%): calcd for C15H19O3: 247.1334;
found: 247.1324 [M+H+] (4), 91 (100).
(2R/S,3S,6R)-3-(Benzyloxy)-6-(prop-1E,Z-enyl)tetrahydro-2H-pyran-2-yl
acetate (14): A solution of lactone 6 (2.46 g, 10 mmol) in dry CH2Cl2
(100 mL) was cooled under N2 to ꢀ808C and treated dropwise with
DIBAL (1m solution in CH2Cl2, 13 mL, 13 mmol). The reaction mixture
was then stirred at ꢀ808C for 2 h. After this time, a solution of DMAP
(7.33 g, 60 mmol) in dry CH2Cl2 (60 mL) was slowly added dropwise, fol-
lowed by addition of Ac2O (8.5 mL, 90 mmol). After stirring overnight at
ꢀ808C, the temperature was allowed to reach 08C and the reaction was
quenched by addition of saturated aqueous NH4Cl (100 mL) and saturat-
ed sodium potassium tartrate (65 mL), followed by stirring at room tem-
perature for 30 min. Work-up (extraction with CH2Cl2) and column chro-
matography on silica gel (hexane/EtOAc 19:1) afforded the acetylated
lactol 14 (2.70 g, 93%) as a mixture of stereoisomers at both the anome-
ric carbon (~2:1) and the olefinic bond (~9:1) as a colourless oil.
15.7, 5.8, 1.5 Hz, 1H), 4.64 (d, J=11.7 Hz, 1H), 4.46 (d, J=11.7 Hz, 1H),
3.82 (m, 2H), 3.16 (td, J=9.7, 4.2 Hz, 1H), 3.00 (dd, J=15.2, 4 Hz, 1H),
2.62 (dd, J=15.2, 8.3 Hz, 1H), 2.28 (m, 1H), 1.78 (m, 1H), 1.69 (brd,
J ~6.4 Hz, 3H), 1.47 (s, 9H), 1.55–1.35 ppm (brm, 2H); 13C NMR (sig-
nals from the major E isomer): d = 197.6, 138.2, 47.7 (C), 131.2, 128.3 (ꢄ
2), 127.6 (ꢄ2), 127.5, 126.5, 77.4, 77.2, 76.6 (CH), 70.5, 47.7, 30.8, 29.0
(CH2), 29.7 (ꢄ3), 17.7 ppm (CH3); IR: nmax = 1684 cmꢀ1 (C=O); HR
ESMS: m/z: calcd for C21H30O3SNa: 385.1813; found: 385.1811 [M+Na+
]. Due to the overlapping of the signals of H-3 and H-7 at d 3.80, this
highly diagnostic NOE was not amenable to measurement. However,
when a chair conformation is assumed for 15b, what seems very likely in
view of the three equatorial substituents, the observed NOE between H-
4 and the acetate side-chain methylene strongly suggests that the mole-
cule shows the configuration depicted below.
1H NMR: d
= 7.35–7.25 (brm, 5H), 6.38 (d, 1H, J=3.1 Hz, minor
anomer, E isomer)/6.33 (d, 1H, J=3.1 Hz, minor anomer, Z isomer),
5.64 (d, 1H, J=7.9 Hz, major anomer, E isomer)/5.59 (d, 1H, J=7.9 Hz,
major anomer, Z isomer), 5.75–5.65 (brm, 1H), 5.50–5.40 (brm, 1H),
4.64 (s, 2H, major anomer)/4.63, 4.53 (AB system, 2H, J=11.7 Hz, minor
anomer), 4.24 (m, 1H, minor anomer)/4.04 (m, 1H, major anomer), 3.57
(m, 1H, minor anomer)/3.38 (m, 1H, major anomer), 2.20 (m, 1H, major
anomer), 2.15 (s, 3H, minor anomer), 2.10 (s, 3H, major anomer), 1.92
(m, 1H, minor anomer), 1.68 (brd, J ~6.4 Hz, 3H), 1.80–1.45 ppm (brm,
3H); IR: nmax = 1752 cmꢀ1 (C=O).
2-[(2S,3S,6R)-3-(Benzyloxy)-6-(prop-1E,Z-enyl)-tetrahydro-2H-pyran-2-
yl]acetic acid (5): A solution of thioester 15a (362 mg, 1 mmol) in
MeOH (200 mL) was treated at room temperature with a solution of
NaOH (8 g, 200 mmol) in water (50 mL). The reaction mixture was then
stirred at room temperature for 16 h, then concentrated under reduced
pressure to eliminate most MeOH and washed with CH2Cl2 to eliminate
neutral impurities. The aqueous layer was then acidified with HCl and
extracted with EtOAc. The organic layer was dried over dry MgSO4 and
evaporated under reduced pressure. The oily residue was acid 5 (273 mg,
94%, ~9:1 E/Z mixture), pure enough for the next reaction. 1H NMR
(signals from the major E isomer): d = 11.00 (brs, 1H, OH), 7.45–7.35
(brm, 5H), 5.80 (m, 1H), 5.60 (m, 1H), 4.70 (d, J=11.8 Hz, 1H), 4.58 (d,
J=11.8 Hz, 1H), 4.60–4.50 (m, 1H), 4.32 (brs, 1H), 3.67 (dt, J=7.8,
3.8 Hz, 1H), 2.90 (dd, J=15.7, 8.8 Hz, 1H), 2.77 (dd, J=15.7, 5.4 Hz,
1H), 2.10–1.80 (brm, 3H), 1.80 (d, J ~6.5 Hz, 3H), 1.55 ppm (m, 1H);
13C NMR (signals from the major E isomer): d = 177.2, 138.0 (C), 130.3,
128.2 (ꢄ2), 127.8, 127.5 (ꢄ2), 127.4, 73.1, 71.0, 70.2 (CH), 70.6, 33.7, 26.9,
S-tert-Butyl 2-[(2S,3S,6R)-3-(benzyloxy)-6-(prop-1E,Z-enyl)tetrahydro-
2H-pyran-2-yl]ethanethioate (15a) and S-tert-butyl 2-[(2R,3S,6R)-3-(ben-
zyloxy)-6-(prop-1E,Z-enyl)tetrahydro-2H-pyran-2-yl]ethanethioate
(15b): The trimethylsilyl enolate of tert-butyl thioacetate, CH2=C-
ACHTUNGTRENNUNG
(OStBu)OSiMe3, was prepared according to a literature procedure.[27]
This enol silane (2.04 g, 10 mmol) and compound 14 (1.16 g, 4 mmol)
were dissolved under N2 in dry MeCN (20 mL), mixed with powdered,
activated 4 ꢅ molecular sieves (2 g) and then cooled to ꢀ308C. Following
this, a solution of freshly distilled BF3-etherate (1 mL, ca. 8 mmol) and
TMSOTf (145 mL, 0.8 mmol) in dry MeCN (3 mL) was added dropwise
to the reaction mixture, which was subsequently stirred at ꢀ188C for
16 h. Work-up (extraction with CH2Cl2) and column chromatography on
silica gel (hexanes/Et2O 49:1 to 19:1) furnished first 15b (305 mg, 21%)
and then 15a (798 mg, 55%), both as ~9:1 E/Z mixtures.
23.1 (CH2), 17.7ppm (CH3); IR: nmax
= 3500–2500 (br, COOH),
1714 cmꢀ1 (C=O); HR ESMS: m/z: calcd for C17H22O4Na: 313.1416;
found: 313.1419 [M+Na+].
2-[(2S,3S,6R)-3-(Benzyloxy)-6-[(S)-6-hydroxyhept-1E,Z-enyl]tetrahydro-
2H-pyran-2-yl]acetic acid (16): Compound 4[11] (342 mg, 3 mmol) and cat-
alyst Ru-II (100 mg, 0.12 mmol) were dissolved under N2 in dry, deoxy-
genated CH2Cl2 (20 mL). The mixture was then heated at reflux. A solu-
tion of acid 5 (175 mg, 0.6 mmol) in dry, deoxygenated CH2Cl2 (5 mL)
was added dropwise, and the mixture was maintained at reflux for 6 h.
After cooling to room temperature, the reaction was quenched through
addition of DMSO[46] (0.4 mL) followed by stirring overnight. Removal
of all volatiles under reduced pressure and filtration of the residue
through silica gel (CH2Cl2/MeOH 19:1) yielded acid 16 (193 mg, 89%) as
a ~70:30 E/Z mixture, which was used as such in the next step. An ali-
15a: colourless oil; 1H NMR (signals from the major E isomer): d =
7.35–7.25 (brm, 5H), 5.75–5.60 (m, 1H), 5.50–5.40 (m, 1H), 4.56 (d, J=
Chem. Eur. J. 2011, 17, 675 – 688
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
683

M. Carda, J. A. Marco et al.
quot was rechromatographed for analytical purposes, which led to an en-
richment in the E isomer: colourless oil; 1H NMR: d = 7.35–7.25 (m,
5H), 5.75–5.65 (m, 1H), 5.55–5.45 (m, 1H), 4.80–4.65 (brs, 2H, OH),
4.62 (d, J=12 Hz, 1H), 4.48 (d, J=12 Hz, 1H), 4.35 (dt, J=8.5, 4 Hz,
1H), 4.30 (brq, J ~5 Hz, 1H), 3.80 (m, 1H), 3.52 (brq, J ~4.5 Hz, 1H),
2.80 (dd, J=15.7, 9.2 Hz, 1H), 2.56 (dd, J=15.7, 4.5 Hz, 1H), 2.15–2.00
(brm, 3H), 1.82 (m, 2H), 1.55–1.40 (m, 5H), 1.18 ppm (d, J=6.5 Hz,
3H); 13C NMR: d = 175.3, 138.2 (C), 133.5, 129.1, 128.4 (ꢄ2), 127.8 (ꢄ
2), 127.7, 73.2, 71.6, 70.2, 68.1 (CH), 70.8, 38.4, 34.7, 32.2, 26.1, 25.0, 23.0
(CH2), 23.3 ppm (CH3); IR: nmax = 3400–2500 (br, COOH), 1722 cmꢀ1
(C=O); HR ESMS: m/z: calcd for C21H30O5Na: 385.1991; found:
385.1989 [M+Na+].
Yields and relative proportions of the two lactones depended on the re-
action time as indicated in Scheme 4.
When compound (Z)-17 was subjected to the same reaction conditions
with catalyst Ru-II (reaction time 16 h), 19 was formed in 76% yield
(based on recovered starting material) together with unreacted (Z)-17.
1
19: colourless oil; H NMR (signals from the major E isomer): d = 7.35–
7.25 (m, 5H), 5.45–5.30 (m*, 2H), 4.87 (m, 1H), 4.68 (d, J=12 Hz, 1H),
4.38 (d, J=12 Hz, 1H), 4.03 (brdt, J ~13, 4.5 Hz, 1H), 3.97 (brd, J
~11.5 Hz, 1H), 3.28 (brs, 1H), 2.77 (dd, J=14, 12 Hz, 1H), 2.68 (brq, J
~12 Hz, 1H), 2.30 (m, 1H), 2.20–1.90 (brm, 5H), 1.80–1.60 (brm, 4H),
1.17 ppm (d, J=6.5 Hz, 3H) (* the signals of each of the two olefinic pro-
tons may be interpreted as double doublets, each peak being further sub-
divided by many small coupling constants of less than 2 Hz; this makes
the extraction of the individual J values difficult, see Supporting Informa-
tion); 13C NMR (signals from the major E isomer): d = 170.6, 138.5 (C),
133.3, 128.3 (ꢄ2), 127.9 (ꢄ2), 127.6, 125.0, 73.3, 72.3, 70.0, 67.3 (CH),
(1S,5S,11R,14S)-14-(Benzyloxy)-5-methyl-4,15-dioxabicycloACTHNUGRTNEUNG[9.3.1]penta-
dec-9E-en-3-one [(E)-17] and (1S,5S,11R,14S)-14-(benzyloxy)-5-methyl-
4,15-dioxabicyclo [9.3.1]pentadec-9Z-en-3-one [(Z)-17]: A solution of
acid 16 (E/Z mixture, 181 mg, 0.5 mmol) in dry THF (4 mL) was cooled
to 08C under N2. Then, triethylamine (700 mL, 5 mmol) and 2,4,6-trichlor-
obenzoyl chloride (470 mL, 3 mmol) were added dropwise, followed by
stirring at room temperature for 1.5 h. The reaction mixture was then di-
luted with dry toluene (100 mL) and added dropwise along 6 h over a so-
lution of DMAP (733 mg, 6 mmol) in dry toluene (250 mL) at 508C.
Work-up (extraction with EtOAc) and column chromatography on silica
gel (hexanes/EtOAc 19:1 to 9:1) furnished first (Z)-17 (29 mg, 17%) and
then (E)-17 (69 mg, 40%). Physical and spectral properties of both com-
pounds described in ref. [6].
70.5, 39.3, 35.0, 34.0, 31.6, 22.1, 21.5 (CH2), 21.2 ppm (CH3); IR: nmax
=
1728 cmꢀ1 (C=O); HR ESMS: m/z: calcd for C21H28O4Na: 367.1885;
found: 367.1886 [M+Na+].
A
(20):
Asymmetric allylation of aldehyde 9 was carried out under the same con-
ditions followed for the preparation of 12, except that (+)-Ipc2BCl was
now the chiral reagent. Column chromatography on silica gel (hexanes/
EtOAc 9:1) afforded alcohol 20, still contaminated with boron-containing
side products, which was used as such in the next step. An aliquot was
carefully purified for analytical purposes: colourless oil. [a]D = ꢀ20.9 (c
= 2, CHCl3); 1H NMR: d = 7.75–7.70 (m, 4H), 7.50–7.30 (brm, 11H),
5.85 (ddt, J=17.2, 10.3, 7 Hz, 1H), 5.20–5.10 (m, 2H), 4.72 (d, J=
11.5 Hz, 1H), 4.56 (d, J=11.5 Hz, 1H), 3.85 (dd, J=10.6, 5.5 Hz, 1H),
3.74 (dd, J=10.6, 5 Hz, 1H), 3.65–3.55 (m, 2H), 2.30–2.25 (m, 1H), 2.20–
2.15 (m, 1H), 2.00 (brs, 1H, OH), 1.80–1.70 (m, 2H), 1.60–1.50 (m, 2H),
1.14 ppm (s, 9H); 13C NMR: d = 138.7, 133.5 (ꢄ2), 19.2 (C), 135.6 (ꢄ4),
134.9, 129.6 (ꢄ2), 128.3 (ꢄ2), 127.8 (ꢄ2), 127.7 (ꢄ4), 127.5, 79.5, 70.6
(CH), 117.8, 72.0, 66.1, 41.9, 32.4, 27.7 (CH2), 26.9 ppm (ꢄ3) (CH3); IR:
nmax = 3415 cmꢀ1 (br, OH); HR ESMS: m/z: calcd for C31H40O3SiNa:
511.2644; found: 511.2645 [M+Na+].
(1S,5S,11,14S)-14-Hydroxy-5-methyl-4,15-dioxabicycloACTHNUTRGNE[NUG 9.3.1]pentadec-
9E-en-3-one (aspergillide B, 2): A solution of lactone (E)-17 (62 mg,
0.18 mmol) was dissolved in CH2Cl2/H2O 10:1 (15 mL) and treated with
DDQ (1.22 g, 5.4 mmol). The reaction mixture was stirred for 20 h at
room temperature. Work-up (extraction with CH2Cl2) and column chro-
matography on silica gel (hexane/EtOAc 7:3) furnished 2 (24 mg, 51%).
Physical and spectral properties described in ref. [6] (see also correction).
(S)-Hept-6-en-2-yl 2-[(2S,3S,6R)-3-(benzyloxy)-6-(prop-1E,Z-enyl)tetra-
hydro-2H-pyran-2-yl]acetate (18):
A solution of acid 5 (174 mg,
0.6 mmol) in dry THF (15 mL) was cooled to 08C under N2. Then, trie-
thylamine (210 mL, 1.5 mmol) and 2,4,6-trichlorobenzoyl chloride
(188 mL, 1.2 mmol) were added dropwise, followed by stirring at room
temperature for 2 h. Alcohol 4 (82 mg, 0.72 mmol) and DMAP (183 mg,
1.5 mmol) were dissolved in dry THF (9 mL) and added slowly via sy-
ringe to the reaction mixture, with further stirring for 16 h at room tem-
perature. Work-up (extraction with Et2O) and column chromatography
on silica gel (hexanes/EtOAc 19:1) furnished 18 (174 mg, 75%) as a ca.
9:1 E/Z mixture: colourless oil; 1H NMR (signals from the major E
isomer): d = 7.35–7.25 (m, 5H), 5.85–5.60 (m, 2H), 5.50–5.40 (m, 1H),
5.05–4.90 (m, 3H), 4.60 (d, J=12 Hz, 1H), 4.55–4.40 (m, 2H), 4.20 (m,
1H), 3.58 (m, 1H), 2.76 (dd, J=15.2, 9 Hz, 1H), 2.60 (dd, J=15.2, 5 Hz,
1H), 2.05 (m, 2H), 1.95–1.60 (brm, 3H), 1.69 (brd, J ~6.2 Hz, 3H), 1.60–
1.30 (brm, 5H), 1.29 ppm (d, J=6.3 Hz, 3H); 13C NMR (signals from the
major E isomer): d = 171.5, 138.4 (C), 138.5, 130.8, 128.3 (ꢄ2), 127.6 (ꢄ
2), 127.5, 127.4, 73.7, 70.9 (ꢄ2), 70.6 (CH), 114.7, 70.7, 35.4, 34.1, 33.5,
27.7, 24.6, 23.6 (CH2), 20.0, 17.8 ppm (CH3); IR: nmax = 1730 cmꢀ1 (C=
O); HR ESMS: m/z: calcd for C24H34O4Na: 409.2355; found: 409.2357
[M+Na+].
AHCTUNGTRENGN(UN 4S,7S)-7-Benzyloxy-8-(tert-butyldiphenylsilyloxy)-4-(triethylsilyloxy)oct-
1-ene (21): The silylation of alcohol 20 was carried out under the same
conditions followed for alcohol 12. Column chromatography on silica gel
(hexanes/Et2O 49:1) furnished 21 (69% overall yield from 9). Colourless
oil. [a]D = ꢀ10.5 (c = 2.4, CHCl3); 1H NMR: d = 7.75–7.70 (m, 4H),
7.45–7.30 (brm, 11H), 5.82 (ddt, J=17.2, 10.3, 7 Hz, 1H), 5.10–5.00 (m,
2H), 4.68 (d, J=11.7 Hz, 1H), 4.54 (d, J=11.7 Hz, 1H), 3.78 (dd, J=
10.6, 5.5 Hz, 1H), 3.75–3.65 (m, 2H), 3.53 (m, 1H), 2.25–2.20 (m, 2H),
1.65–1.45 (m, 4H), 1.10 (s, 9H), 0.98 (t, J=8 Hz, 9H), 0.62 ppm (q, J=
8 Hz, 6H); 13C NMR: d = 139.1, 133.6 (ꢄ2), 19.2 (C), 135.6 (ꢄ4), 135.2,
129.6 (ꢄ2), 128.2 (ꢄ2), 127.7 (ꢄ2), 127.6 (ꢄ4), 127.4, 80.0, 72.0 (CH),
116.7, 71.9, 66.3, 41.9, 32.5, 27.3, 5.1 (ꢄ3) (CH2), 26.9 (ꢄ3), 7.0 ppm (ꢄ3)
(CH3); HR ESMS: m/z: calcd for C37H54O3Si2Na: 625.3509; found:
625.3510 [M+Na+].
AHCTUNGTRENGN(UN 4S,7S)-7-Benzyloxy-8-(tert-butyldiphenylsilyloxy)-4-(triethylsilyloxy)oct-
2E,Z-ene (22): The double-bond isomerization in olefin 21 to yield 22
was carried out under the same conditions followed for the conversion of
8 into 7. Column chromatography on silica gel (hexanes/Et2O 49:1) gave
22 (88%) as a ca. 9:1 E/Z mixture: colourless oil. 1H NMR: d = 7.75–
7.70 (m, 4H), 7.50–7.30 (brm, 11H), 5.55–5.45 (m, 1H), 5.40–5.30 (m,
1H), 4.65 (d, J=11.8 Hz, 1H), 4.50 (d, J=11.8 Hz, 1H), 4.00 (m, 1H),
3.73 (dd, J=10.6, 3.5 Hz, 1H), 3.64 (dd, J=10.6, 4.4 Hz, 1H), 3.50 (m,
1H), 1.65 (brd, J ~6.3 Hz, 3H), 1.65–1.40 (brm, 4H), 1.05 (s, 9H), 0.92
(t, J=7.7 Hz, 9H), 0.57 ppm (q, J=7.7 Hz, 6H); 13C NMR (signals from
the major E isomer): d = 139.1, 133.6 (ꢄ2), 19.2 (C), 135.6 (ꢄ4), 134.7,
129.6 (ꢄ2), 128.2 (ꢄ2), 127.7 (ꢄ2), 127.6 (ꢄ4), 127.4, 125.3, 80.0, 73.6
(CH), 72.0, 66.3, 34.1, 27.3, 5.0 (ꢄ3) (CH2), 26.8 (ꢄ3), 17.5, 6.9 ppm (ꢄ3)
(CH3); HR ESMS: m/z: calcd for C37H54O3Si2Na: 625.3509; found:
625.3502 [M+Na+].
Ring-closing metathesis of 18: a) With catalyst Ru-I: Grubbs ruthenium
catalyst Ru-I (16.5 mg, ca. 0.02 mmol) was dissolved under N2 in dry, de-
oxygenated CH2Cl2 (90 mL). After heating the solution to reflux, diene
18 (39 mg, ca. 0.1 mmol) dissolved in dry, deoxygenated CH2Cl2 (10 mL)
was added slowly via syringe (within 1 h) to the reagent solution. The re-
action mixture was then additionally stirred at reflux for 2 h. After cool-
ing to room temperature, the reaction was quenched through addition of
DMSO[46] (75 mL) followed by stirring overnight. Removal of all volatiles
under reduced pressure and column chromatography of the residue on
silica gel (hexanes/EtOAc 19:1 to 9:1) yielded first (Z)-17 (16.5 mg,
48%) and then (E)-17 (11 mg, 33%).
b) With catalyst Ru-II: the reaction was carried out under the same con-
ditions as above. Column chromatography on silica gel (hexanes/EtOAc
19:1 to 9:1) gave first (Z)-17 and then 19 as a ca. 85:15 E/Z mixture.
AHCTUNGTERG(NNUN 2S,5S)-2-(Benzyloxy)-oct-6E,Z-ene-1,5-diol (23): Desilylation of com-
pound 22 was carried out under the same conditions followed for the
preparation of 13. Column chromatography on silica gel (hexanes/EtOAc
684
www.chemeurj.org
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 675 – 688

Aspergillide A and B
FULL PAPER
1:1) afforded diol 23 as a ca. 9:1 E/Z mixture (91%). Colourless oil;
1H NMR (signals from the major E isomer): d = 7.35–7.25 (brm, 5H),
5.60–5.55 (m, 1H), 5.45–5.40 (m, 1H), 4.54 (s, 2H), 3.95 (m, 1H), 3.62
(dd, J=11.5, 3.8 Hz, 1H), 3.52 (dd, J=11.5, 5.5 Hz, 1H), 3.47 (m, 1H),
3.10 (brs, 1H, OH), 2.90 (brs, 1H, OH), 1.67 (brd, J ~6.5 Hz, 3H), 1.65–
1.45 ppm (brm, 4H); 13C NMR (signals from the major E isomer): d =
138.2 (C), 133.9, 128.2 (ꢄ2), 127.6 (ꢄ2), 127.4, 126.2, 79.3, 72.3 (CH),
71.2, 63.5, 32.4, 26.4 (CH2), 17.4 ppm (CH3); IR: nmax = 3380 cmꢀ1 (br,
OH); HR ESMS: m/z: calcd for C15H22O3Na: 273.1467; found: 273.1468
[M+Na+].
(brd, J ~15 Hz, 1H), 1.65 (brq, J ~13.5 Hz, 1H), 1.58 (brd, J ~6.5 Hz,
3H), 1.50–1.30 (m, 2H), 1.35ppm (s, 9H); 13C NMR (signals from the
major E isomer): d = 198.2, 138.4, 47.8 (C), 131.9, 128.2 (ꢄ2), 128.0 (ꢄ
2), 127.5, 126.8, 78.5, 76.0, 71.2 (CH), 70.9, 46.5, 26.0, 25.8 (CH2), 29.7 (ꢄ
3), 17.7 ppm (CH3); IR: nmax = 1679 cmꢀ1 (C=O); HR ESMS: m/z: calcd
for C21H30O3SNa: 385.1813; found: 385.1800 [M+Na+]. The NOE indi-
cated above is diagnostic of the relative configuration at C-3 and C-7.
ACHTUNGTRENNUNG(3S,6S)-3-(Benzyloxy)-6-(prop-1E,Z-enyl)tetrahydro-2H-pyran-2-one
(24): The oxidation of diol 23 to lactone 24 was carried out under the
same conditions followed for the preparation of 6. CAUTION: evapora-
tion of volatiles under reduced pressure has to be performed at room
temperature, due to the appreciable volatility of lactone 24. Column
chromatography on silica gel (pentane/Et2O 9:1) provided 24 as a ca.
~9:1 E/Z mixture (84%). Colourless oil; 1H NMR (signals from the
major E isomer): d = 7.45–7.30 (brm, 5H), 5.79 (brdq, J=15.5, 6.8 Hz,
1H), 5.49 (ddq, J=15.5, 6.5, 1.5 Hz, 1H), 4.98 (d, J=11.8 Hz, 1H), 4.72
(m, 1H), 4.65 (d, J=11.8 Hz, 1H), 4.05 (t, J=8.2 Hz, 1H), 2.25–2.20 (m,
1H), 2.00–1.85 (brm, 3H), 1.73 ppm (brd, J=7 Hz, 3H); 13C NMR (sig-
nals from the major E isomer): d = 171.5, 137.5 (C), 130.0, 128.6, 128.4
(ꢄ2), 128.0 (ꢄ2), 127.8, 78.7, 71.5 (CH), 72.4, 26.6, 25.1 (CH2), 17.6 ppm
Ethyl
2-[(2R,3S,6S)-3-(benzyloxy)-6-(prop-1E,Z-enyl)tetrahydro-2H-
pyran-2-yl]acetate (31): A solution of ethyl acetate (0.1 mL, ~1 mmol) in
dry THF (3 mL) was cooled under N2 to ꢀ808C and treated dropwise
(CH3); IR: nmax
=
1745 cmꢀ1 (C=O); HR ESMS: m/z: calcd for
C15H18O3Na: 269.1154; found: 269.1154 [M+Na+].
(2R/S,3S,6S)-3-(Benzyloxy)-6-(prop-1E,Z-enyl)tetrahydro-2H-pyran-2-yl
acetate (25): Reduction of lactone 24 to lactol acetate 25 was carried out
under the same conditions followed for the preparation of 14. Column
chromatography on silica gel (hexane/EtOAc 19:1) afforded 25 (93%) as
a mixture of stereoisomers at both the anomeric carbon and the olefinic
bond: colourless oil; 1H NMR: d
= 7.35–7.20 (brm, 5H), 5.75–5.50
with lithium hexamethyldisilazide (1m in hexane, 0.7 mL, 0.7 mmol).
After stirring at this temperature for 30 min, a solution of lactone 24
(25 mg, ca. 0.1 mmol) in dry THF (1 mL) was added dropwise. The reac-
tion mixture was then stirred at ꢀ808C for 15 min. Work-up (extraction
with EtOAc) gave crude lactol 30, which was dried and used directly in
the next step.
(brm, 3H, olefinic and anomeric protons), 4.64 (brs, 2H), 4.05 (ddd, J=
10, 6.5, 2.5 Hz, 1H), 3.52 (brs, 1H), 2.10–2.00 (m, 1H), 2.06 (s, 3H),
1.85–1.75 (m, 1H), 1.65 (brd, J ~6 Hz, 3H), 1.65–1.55 (m, 1H), 1.50–
1.40 ppm (m, 1H); IR: nmax = 1750 cmꢀ1 (C=O).
S-tert-Butyl 2-[(2R,3S,6S)-3-(benzyloxy)-6-(prop-1E,Z-enyl)tetrahydro-
2H-pyran-2-yl]ethanethioate (26a) and S-tert-butyl 2-[(2S,3S,6S)-3-(ben-
zyloxy)-6-(prop-1E,Z-enyl)tetrahydro-2H-pyran-2-yl]ethanethioate
(26b): The C-glycosidation of 25 was carried out under the same condi-
tions followed for the preparation of thiol esters 15a and 15b. Work-up
(extraction with CH2Cl2) and column chromatography on silica gel (hex-
anes/Et2O 49:1 to 19:1) furnished first 26b (24%) and then 26a (46%),
both as ~9:1 E/Z mixtures.
The oily lactol from above was dissolved under N2 in dry CH2Cl2 (2 mL)
and treated with triethylsilane (160 mL, 1 mmol). After cooling to ꢀ808C,
freshly distilled BF3·Et2O (62 mL, 0.5 mmol) was added dropwise. The re-
action mixture was then stirred at the same temperature for 16 h. Work-
up (extraction with CH2Cl2) and column chromatography on silica gel
(hexanes/EtOAc 9:1) afforded 31 (29 mg, 91%) as a ca. 9:1 E/Z mixture
as a colourless oil. 1H NMR (signals from the major E isomer): d =
7.35–7.25 (brm, 5H), 5.75–5.65 (m, 1H), 5.57 (ddq, J=15.5, 3, 1.7 Hz,
1H), 4.62 (d, J=11.6 Hz, 1H), 4.47 (d, J=11.6 Hz, 1H), 4.30 (m, 1H),
4.15–4.05 (m, 3H), 3.22 (td, J=8, 4 Hz, 1H), 2.80 (dd, J=15.3, 4 Hz,
1H), 2.44 (dd, J=15.3, 9 Hz, 1H), 2.02 (m, 1H), 1.85–1.65 (brm, 3H),
1.72 (brd, J ~6.5 Hz, 3H), 1.24 ppm (t, J=7.3 Hz, 3H); 13C NMR (signals
from the major E isomer): d = 171.5, 138.3 (C), 130.0, 128.3 (ꢄ2), 128.1,
127.7 (ꢄ2), 127.6, 76.5, 72.0, 70.7 (CH), 70.5, 60.3, 38.0, 27.2, 24.4 (CH2),
18.0, 14.2 ppm (CH3); IR: nmax = 1735 cmꢀ1 (C=O); HR ESMS: m/z:
calcd for C19H26O4Na: 341.1729; found: 341.1728 [M+Na+]. The NOE in-
dicated above (see Supporting Information) is diagnostic of the relative
configuration at C-3 and C-7.
26a: oil; 1H NMR (signals from the major E isomer): d = 7.35–7.20 (m,
5H), 5.75–5.65 (brm, 1H), 5.55–5.50 (m, 1H), 4.60 (d, J=12 Hz, 1H),
The experimental procedures for the preparation of compounds 28, 29,
32, (Z)-33 and 34, as well as the corresponding spectral data, can be
taken from ref. [9a].
4.47 (d, J=12 Hz, 1H), 4.30 (brs, 1H), 4.15 (td, J=7.7, 4.5 Hz, 1H), 3.20
(m, 1H), 2.90 (dd, J=14.7, 4.4 Hz, 1H), 2.60 (dd, J=14.7, 8.3 Hz, 1H),
1.95 (m, 1H), 1.80 (m, 1H), 1.75–1.60 (brm, 2H), 1.71 (brd, J ~6.5 Hz,
3H), 1.44 ppm (s, 9H); 13C NMR (signals from the major E isomer): d =
197.5, 138.2, 47.8 (C), 130.1, 128.2 (ꢄ2), 127.8, 127.5 (ꢄ2), 127.4, 75.8,
71.6, 71.0 (CH), 70.3, 47.1, 26.9, 24.2 (CH2), 29.6 (ꢄ3), 17.8 ppm (CH3);
(1S,5S,11S,14S)-14-Hydroxy-5-methyl-4,15-dioxabicyclo-
AHCTUNGERTG[NNUN 9.3.1]pentadecan-3-one (35): Palladium catalyst (10% Pd/C, 10 mg) was
suspended in EtOAc (1 mL) and stirred under H2 at room temperature
for 10 min. Subsequently, a solution of compound 17, either as a single
stereoisomer or as the E/Z mixture (17 mg, 0.05 mmol) in EtOAc (2 mL)
was added via syringe, followed by stirring under H2 at room temperature
for 2 h. The reaction mixture was then filtered through Celite, and the
solvent was removed under reduced pressure. Column chromatography
of the residue on silica gel (hexanes/EtOAc 7:3) yielded 35 (12 mg, 95%)
as a colourless oil. [a]D = ꢀ28.1 (c = 0.1, CHCl3); 1H NMR: d = 5.02
(m, 1H), 4.20 (brd, J ~11.3 Hz, 1H), 3.94 (brdd, J ~11, 6 Hz, 1H), 3.60
IR: nmax
=
1683 cmꢀ1 (C=O); HR ESMS: m/z: calcd for C21H31O3S:
363.1994; found: 363.1999 [M+H+]. The NOE indicated above is diag-
nostic of the relative configuration at C-3 and C-7.
26b: oil; 1H NMR (signals from the major E isomer): d = 7.30–7.15 (m,
5H), 5.60 (brdq, J=15.2, 6.5 Hz, 1H), 5.42 (ddq, J=15.2, 6.5, 1 Hz, 1H),
4.54 (d, J=12 Hz, 1H), 4.35 (d, J=12 Hz, 1H), 3.81 (brt, J ~6.5 Hz, 1H),
3.76 (brdd, J=10.8, 6 Hz, 1H), 3.30 (brs, 1H), 2.80–2.70 (m, 2H), 2.05
Chem. Eur. J. 2011, 17, 675 – 688
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
685

M. Carda, J. A. Marco et al.
(m, 1H), 2.66 (brt, J ~13 Hz, 1H), 2.42 (brd, J ~13 Hz, 1H), 2.15–2.05
(brm, 3H), 1.90–1.75 (brm, 2H), 1.75–1.50 (brm, 6H), 1.40–1.20 (brm,
overlapping a methyl doublet, 4H), 1.20 ppm (m, 1H); 13C NMR: d =
170.7 (C), 75.1, 69.8, 69.4, 66.8 (CH), 39.3, 32.9, 26.6, 26.0, 25.6, 25.1, 23.9,
21.6 (CH2), 20.0 ppm (CH3); IR: nmax = 3450 (br, OH), 1735 cmꢀ1 (C=
O); HR ESMS: m/z: calcd for C14H22O4Na: 279.1572; found: 279.1575
[M+Na+].
Guzman, D. Harmody, P. Linley, P. J. McCarthy, T. P. Pitts, S. A.
Pomponi, J. K. Reed, J. Nat. Prod. 2007, 70, 412–416. The structure
reported in this paper has been later corrected as a consequence of
synthetic efforts of several groups. For a review on this compound,
see: b) J. Gallon, S. Reymond, J. Cossy, C. R. Chim. 2008, 11, 1463–
1476. For more recent syntheses of neopeltolide, see: c) D. W.
Custar, T. P. Zabawa, J. Hines, C. M. Crews, K. A. Scheidt, J. Am.
Chem. Soc. 2009, 131, 12406–12414; d) X. Guinchard, E. Roulland,
Org. Lett. 2009, 11, 4700–4703; e) H. Fuwa, A. Saito, M. Sasaki,
Angew. Chem. 2010, 122, 3105–3108; Angew. Chem. Int. Ed. 2010,
49, 3041–3044.
Materials and methods for the biological work
Cell culture: Cell culture media were purchased from Gibco (Grand
Island, NY, USA) and Cambrex (Walkersville, MD, USA). Fetal bovine
serum (FBS) was a product of Harlan-Seralab (Belton, UK). Supple-
ments and other chemicals not listed in this section were obtained from
Sigma Chemicals Co. (St. Louis, Mo., USA). Plastics for cell culture were
supplied by NUNC (Roskilde, Denmark). Aspergillides and their ana-
logues (for structures, see Schemes 1, 3 and 10) were dissolved in DMSO
and stored at ꢀ208C until use.
[4] Pochonin J (benzofused 14-membered macrolide containing a trans-
2,6-disubstituted tetrahydropyran ring): H. Shinonaga, Y. Kawa-
mura, A. Ikeda, M. Aoki, N. Sakai, N. Fujimoto, A. Kawashima, Tet-
rahedron Lett. 2009, 50, 108–110.
[5] S. M. Hande, J. Uenishi, Tetrahedron Lett. 2009, 50, 189–192.
[6] S. Dꢀaz-Oltra, C. A. Angulo-Pachꢁn, M. N. Kneeteman, J. Murga,
M. Carda, J. A. Marco, Tetrahedron Lett. 2009, 50, 3783–3785.
When we started our work, the synthesis of Hande and Uenishi had
not yet been reported. Our initial synthetic target therefore was as-
pergillide A, still believed to be 2. We apologize for a typographical
mistake in the 1H NMR data of our synthetic compound: the signal
indicated at d=5.64 ppm (brdd, J=15.5, 4 Hz, 1H) appears actually
at d=5.40 ppm.
Bovine aortic endothelial (BAE) cells were obtained by collagenase di-
gestion and maintained in Dulbeccoꢆs modified Eagleꢆs medium
(DMEM) containing glucose (1 gLꢀ1), glutamine (2 mm), penicillin
(50 IUmLꢀ1),
streptomycin
(50 mgmLꢀ1),
and
amphoterycin
(1.25 mgmLꢀ1), supplemented with 10% FBS. All the cancer cell lines
used in this study were obtained from the American Type Culture Collec-
tion (ATCC). Human fibrosarcoma HT1080 cells were maintained in
DMEM containing glucose (4.5 gLꢀ1), glutamine (2 mm), penicillin (50
IUmLꢀ1), streptomycin (50 mgmLꢀ1), and amphoterycin (1.25 mgmLꢀ1),
supplemented with 10% FBS. Human colon adenocarcinoma HT29 and
human osteosarcoma U2-OS cells were maintained in McCoyꢆs 5A
medium containing glutamine (2 mm), penicillin (50 IUmLꢀ1), streptomy-
cin (50 mgmLꢀ1), and amphoterycin (1.25 mgmLꢀ1), supplemented with
10% FBS. Human breast cancer carcinoma MDA-MB-231 and human
promyelocytic leukemia HL60 cells were maintained in RPMI1640
medium containing glutamine (2 mm), penicillin (50 IUmLꢀ1), streptomy-
cin (50 mgmLꢀ1), and amphoterycin (1.25 mgmLꢀ1), supplemented with 10
and 20% FBS, respectively.
[7] R. Ookura, K. Kito, Y. Saito, T. Kusumi, T. Ooi, Chem. Lett. 2009,
384.
[8] Further syntheses of aspergillide B have recently been reported:
a) J. Liu, K. Xu, J.-M. He, L. Zhang, X.-F. Pan, X.-G. She, J. Org.
Chem. 2009, 74, 5063–5066; b) T. Nagasawa, S. Kuwahara, Biosci.
Biotechnol. Biochem. 2009, 73, 1893–1894; c) A. J. Mueller-Hendrix,
M. P. Jennings, Tetrahedron Lett. 2010, 51, 4260–4262; d) H. Fuwa,
H. Yamaguchi, M. Sasaki, Tetrahedron 2010, 66, 7492–7503.
[9] a) S. Dꢀaz-Oltra, C. A. Angulo-Pachꢁn, J. Murga, M. Carda, J. A.
Marco, J. Org. Chem. 2010, 75, 1775–1778; b) a second synthesis of
aspergillide A has recently appeared: T. Nagasawa, S. Kuwahara,
Tetrahedron Lett. 2010, 51, 875–877, see also reference [8d].
[10] a) T. Nagasawa, S. Kuwahara, Org. Lett. 2009, 11, 761–764; b) J. D.
Panarese, S. P. Waters, Org. Lett. 2009, 11, 5086–5088.
Cytotoxicity assays: The 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazo-
lium bromide (MTT; Sigma Chemical Co., St. Louis, MO) dye reduction
assay in 96-well microplates was performed by Drug Discovery Biotech
S.L. (Mꢂlaga, Spain).[47] 3ꢄ103 BAE or 2 ꢄ 103 tumor cells in a total
volume of 100 mL of their respective growth media were incubated with
serial dilutions of the tested compounds. After 3 d of incubation (378C,
5% CO2 in a humid atmosphere) 10 mL of MTT (5 mgmLꢀ1 in PBS)
were added to each well and the plate was incubated for further 4 h
(378C). The resulting formazan was dissolved in 0.04n HCl in isopropa-
nol (150 mL) and read at 550 nm. All determinations were carried out in
triplicate. IC50 values were calculated from semilogarithmic dose-re-
sponse plots as the concentration of compound yielding a 50% of cell
survival.
[11] D. J. Dixon, S. V. Ley, E. W. Tate, J. Chem. Soc. Perkin Trans. 1
2000, 2385–2394.
[12] For reviews on ring-closing metathesis, see: a) A. Fꢇrstner, Angew.
Chem. 2000, 112, 3140–3172; Angew. Chem. Int. Ed. 2000, 39, 3012–
3043; b) L. Jafarpour, S. P. Nolan, Adv. Org. Chem. 2000, 46, 181–
222; c) T. M. Trnka, R. H. Grubbs, Acc. Chem. Res. 2001, 34, 18–29;
d) J. A. Love in Handbook of Metathesis, Vol. 2 (Ed.: R. H.
Grubbs), Wiley-VCH, Weinheim, 2003, pp. 296–322; e) R. H.
Grubbs, Tetrahedron 2004, 60, 7117–7140; f) D. Astruc, New J.
Chem. 2005, 29, 42–56; g) A. H. Hoveyda, A. R. Zhugralin, Nature
2007, 450, 243–251.
[13] For reviews on cross-metathesis, see: a) S. J. Connon, S. Blechert,
Angew. Chem. 2003, 115, 1944–1968; Angew. Chem. Int. Ed. 2003,
42, 1900–1923; b) A. K. Chatterjee, T.-L. Choi, D. P. Sanders, R. H.
Grubbs, J. Am. Chem. Soc. 2003, 125, 11360–11370; c) A. J. Vernall,
A. D. Abell, Aldrichimica Acta 2003, 36, 93–105; d) Y. Schrodi,
R. L. Pederson, Aldrichimica Acta 2007, 40, 45–52.
[14] For reviews on various synthetic methods for the preparation of tet-
rahydropyran derivatives, see: a) T. L. B. Boivin, Tetrahedron 1987,
43, 3309–3362; b) H. Kotsuki, Synlett 1992, 97–106; c) P. A. Clarke,
S. Santos, Eur. J. Org. Chem. 2006, 2045–2053; d) V. Piccialli, Syn-
thesis 2007, 2585–2607; e) A. B. Smith, III, R. J. Fox, T. M. Razler,
Acc. Chem. Res. 2008, 41, 675–687.
Acknowledgements
Financial support has been granted by the Spanish Ministry of Education
and Science (project CTQ2008-02800), by the Consellerꢀa d’Empresa,
Universitat i Ciencia de la Generalitat Valenciana (project ACOMP09/
113) and by the BANCAJA-UJI Foundation (projects P1-1B2002-06 and
P1-1B-2008-14).
[15] For examples of synthesis of cis-2,6-disubstituted tetrahydropyran
rings, see: a) M. D. Lewis, J. K. Cha, Y. Kishi, J. Am. Chem. Soc.
1982, 104, 4976–4978; b) W. Zheng, J. A. DeMattei, J.-P. Wu, J. J.-W.
Duan, L. R. Cook, H. Oinuma, Y. Kishi, J. Am. Chem. Soc. 1996,
118, 7946–7968; c) D. A. Evans, P. H. Carter, E. M. Carreira, A. B.
Charette, J. A. Prunet, M. Lautens, J. Am. Chem. Soc. 1999, 121,
7540–7552; d) D. J. Kopecky, S. D. Rychnovsky, J. Am. Chem. Soc.
2001, 123, 8420–8421; e) S. M. Berberich, R. J. Cherney, J. Colucci,
[1] K. Kito, R. Ookura, S. Yoshida, M. Namikoshi, T. Ooi, T. Kusumi,
Org. Lett. 2008, 10, 225–228.
[2] These compounds should not be confused with a group of other
structurally unrelated metabolites of the same name isolated from
Aspergillus terreus, see: B. T. Golding, R. W. Rickards, Z. Vanek, J.
Chem. Soc. Perkin Trans. 1 1975, 1961–1963.
[3] Neopeltolide (14-membered macrolide containing a cis-2,6-disubsti-
tuted tetrahydropyran ring): a) A. E. Wright, J. C. Botelho, E.
686
www.chemeurj.org
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 675 – 688

Aspergillide A and B
FULL PAPER
C. Courillon, L. S. Geraci, T. A. Kirkland, M. A. Marx, M. F.
Schneider, S. F. Martin, Tetrahedron 2003, 59, 6819–6832; f) G. Pat-
tenden, M. A. Gonzꢂlez, P. B. Little, D. S. Millan, A. T. Plowright,
J. A. Tornos, T. Ye, Org. Biomol. Chem. 2003, 1, 4173–4208; g) S.
Das, L.-S. Li, S. C. Sinha, Org. Lett. 2004, 6, 123–126; h) E. J. Kang,
E. J. Cho, M. K. Ji, Y. E. Lee, D. M. Shin, S. Y. Choi, Y. K. Chung,
J. S. Kim, H.-J. Kim, S.-G. Lee, M. S. Lah, E. Lee, J. Org. Chem.
2005, 70, 6321–6329; i) D.-R. Li, D.-H. Zhang, C.-Y. Sun, J.-W.
Zhang, L. Yang, J. Chen, B. Liu, C. Su, W.-S. Zhou, G.-Q. Lin,
Chem. Eur. J. 2006, 12, 1185–1204; j) D. M. Troast, J. Yuan, J. A.
Porco Jr. , Adv. Synth. Catal. 2008, 350, 1701–1711; k) T. A. Mitch-
ell, C. Zhao, D. Romo, J. Org. Chem. 2008, 73, 9544–9551; l) S. Ha-
nessian, X. Mi, Synlett 2010, 761–764. See also reference [16g].
[16] For examples of synthesis of trans-2,6-disubstituted tetrahydropyran
rings, see: a) K. Horita, Y. Oikawa, O. Yonemitsu, Chem. Pharm.
Bull. 1989, 37, 1698–1704; b) Y. Wang, S. A. Babirad, Y. Kishi, J.
Org. Chem. 1992, 57, 468–481; c) T. Haneda, P. G. Goekjian, S. H.
Kim, Y. Kishi, J. Org. Chem. 1992, 57, 490–498; d) G. C. Micalizio,
A. N. Pinchuk, W. R. Roush, J. Org. Chem. 2000, 65, 8730–8736;
e) A. Fettes, E. M. Carreira, J. Org. Chem. 2003, 68, 9274–9283;
f) D. R. Williams, S. Patnaik, S. V. Plummer, Org. Lett. 2003, 5,
5035–5038; g) I. Paterson, M. Tudge, Tetrahedron 2003, 59, 6833–
6849; h) J. De Vicente, J. R. Huckins, S. D. Rychnovsky, Angew.
Chem. 2006, 118, 7416–7420; Angew. Chem. Int. Ed. 2006, 45, 7258–
7262. See also references [15a,c,k].
[23] Protection of the free OH group in compound 12 is essential. At-
tempts at double bond isomerization in 12 under various conditions
gave only ketone iii in low yield.
[24] Both E/Z isomers were synthetically productive. However, in order
to avoid working with mixtures, we tried to convert compound 7
into its nor-methyl derivative (vinyl instead of propenyl) by means
of cross metathesis under an ethylene atmosphere. This goal was
achieved but yields were not satisfactory (50–55% at <1 mmol
scale) and difficult to reproduce at a larger scale.
[25] a) A. De Mico, R. Margarita, L. Parlanti, A. Vescovi, G. Piancatelli,
J. Org. Chem. 1997, 62, 6974–6977; b) I. Larrosa, M. I. Da Silva,
P. M. Gꢁmez, P. Hannen, E. Ko, S. R. Lenger, S. R. Linke, A. J. P.
White, D. Wilton, A. G. M. Barrett, J. Am. Chem. Soc. 2006, 128,
14042–14043.
[26] Compound 14 was obtained a ca. 2:1 mixture of anomers, each of
them being a ꢃ9:1 mixture of E/Z isomers (see the Experimental
Section).
[27] a) G. Simchen, W. West, Synthesis 1977, 247–248; b) D. A. Evans,
K. A. Scheidt, J. N. Johnston, M. C. Willis, J. Am. Chem. Soc. 2001,
123, 4480–4491.
[17] a) T. Mukaiyama, Aldrichimica Acta 1996, 29, 59–76; b) T. Mu-
kaiyama, J.-I. Matsuo in Modern Aldol Reactions, Vol. 1 (Ed.: R.
Mahrwald), Wiley-VCH, Weinheim, 2004, Chapter 3; c) For
a
[28] I. Paterson, C. A. Luckhurst, Tetrahedron Lett. 2003, 44, 3749–3754.
[29] We have found one precedent of this Mukaiyama-type C-glycosida-
tion method by using a ketene O,S-acetal: J. P. Vitale, S. A. Wolck-
enhauer, N. M. Do, S. D. Rychnovsky, Org. Lett. 2005, 7, 3255–3258.
[30] J. Inanaga, K. Hirata, H. Saeki, T. Katsuki, M. Yamaguchi, Bull.
Chem. Soc. Jpn. 1979, 52, 1989–1993.
review on synthesis of C-glycosides, see: Y. Du, R. J. Linhardt, I. R.
Vlahov, Tetrahedron 1998, 54, 9913–9959. See also reference [15k].
[18] The presence in compounds 5–7 of a 1-propenyl moiety instead of
the more logical vinyl residue has its reasons. We made attempts at
obtaining compound ii, similar to 12 but bearing a vinyl fragment,
by means of asymmetric ethynylation of aldehyde 9 (D. E. Frantz,
R. Fꢈssler, E. M. Carreira, J. Am. Chem. Soc. 2000, 122, 1806–
1807), followed by partial hydrogenation of the triple bond. Un-
fortunately, the ethynylation step to give i proved too slow (mainly
starting materials in the reaction mixture after 4 d) and was aban-
doned. Therefore, the alternative sequence 9 ! 5 was investigated.
[31] a) N. Ikemoto, S. L. Schreiber, J. Am. Chem. Soc. 1992, 114, 2524–
2536; b) M. T. Crimmins, K. A. Emmitte, Org. Lett. 1999, 1, 2029–
2032.
[32] The use of BCl3 for benzyl cleavage[33] led to complete decomposi-
tion.
[33] T. W. Greene, P. G. M. Wuts, Protective Groups in Organic Synthesis,
3rd ed., Wiley, New York, 1999, p. 82.
[34] The energy contents and relative stabilities of synthetic lactones
(E)-17, (Z)-17, (E)-19, (E)-33 and (Z)-33, as well as of the nonpre-
pared lactone (E)-36, have been evaluated with the aid of theoreti-
cal calculations. For the results of these, see the Supporting Informa-
tion.
[19] Compound 10 was prepared in two steps from (R)-glycidol (silyla-
tion and epoxide opening with an allylcopper reagent): D. J. Dixon,
S. V. Ley, D. J. Reynolds, Chem. Eur. J. 2002, 8, 1621–1636.
[20] a) K. Mislow, R. E. O’Brien, H. Schaefer, J. Am. Chem. Soc. 1962,
84, 1940–1944; b) K. Takai, C. H. Heathcock, J. Org. Chem. 1985,
50, 3247–3251; c) J. A. Marshall, B. M. Seletsky, G. P. Luke, J. Org.
Chem. 1994, 59, 3413–3420; d) M.-D. Chen, M.-Z. He, X. Zhou, L.-
Q. Huang, Y.-P. Ruan, P.-Q. Huang, Tetrahedron 2005, 61, 1335–
1344; e) F. Yokokawa, A. Inaizumia, T. Shioiri, Tetrahedron 2005,
61, 1459–1480.
[21] a) P. V. Ramachandran, G.-M. Chen, H. C. Brown, Tetrahedron Lett.
1997, 38, 2417–2420; b) P. V. Ramachandran, Aldrichimica Acta
2002, 35, 23–35.
[22] a) Y.-J. Hu, R. Dominique, S. K. Das, R. Roy, Can. J. Chem. 2000,
78, 838–845; b) P. Wipf, S. R. Rector, H. Takahashi, J. Am. Chem.
Soc. 2002, 124, 14848–14849; c) B. Sieng, O. L. Ventura, V. Bellosta,
J. Cossy, Synlett 2008, 1216–1218.
[35] For reviews on non-metathetic reactions catalyzed by ruthenium
complexes, see: a) B. M. Trost, F. D. Toste, A. B. Pinkerton, Chem.
Rev. 2001, 101, 2067–2096; b) B. Schmidt, Angew. Chem. 2003, 115,
5146–5149; Angew. Chem. Int. Ed. 2003, 42, 4996–4999; c) B. Al-
caide, P. Almendros, Chem. Eur. J. 2003, 9, 1258–1262; d) B.
Chem. Eur. J. 2011, 17, 675 – 688
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
687

M. Carda, J. A. Marco et al.
Schmidt, Eur. J. Org. Chem. 2004, 1865–1880; e) M. Arisawa, Y.
Terada, K. Takahashi, M. Nakagawa, A. Nishida, Chem. Rec. 2007,
7, 238–253.
5322–5323; c) C. G. Lucero, K. A. Woerpel, J. Org. Chem. 2006, 71,
2641–2647; d) D. M. Smith, K. A. Woerpel, Org. Biomol. Chem.
2006, 4, 1195–1201; e) J. R. Krumper, W. A. Salamant, K. A. Woer-
pel, Org. Lett. 2008, 10, 4907–4910; f) M. T. Yang, K. A. Woerpel, J.
Org. Chem. 2009, 74, 545–553; g) J. R. Krumper, W. A. Salamant,
K. A. Woerpel, J. Org. Chem. 2009, 74, 8039–8050; h) M. G. Beaver,
K. A. Woerpel, J. Org. Chem. 2010, 75, 1107–1118; i) For the impor-
tance of ion pair formation in glycosylation processes, see: D. Crich,
Acc. Chem. Res. 2010, 43, 1144–1153.
[36] With respect to the reaction conditions used in our previous synthe-
sis of aspergillide B from lactol 14,[6] where the trans (major) to cis
(minor) ratio was about 2.6:1, changes in temperature, solvent and
Lewis acid did not lead to mixtures in which the desired cis-3,7-dis-
ubstituted tetrahydropyran (aspergillide numbering) was clearly the
major component. The enolsilanes of acetic acid and ethyl acetate
were tried in addition to tert-butyl thioacetate. In all cases, mixtures
of the cis and trans isomers were obtained but the cis/trans ratio was
not ameliorated in this way. Finally, we tried to equilibrate the unde-
sired C-3 epimer of 29 with 29 via a base-catalyzed retro-Michael/
Michael sequence (equilibration). Again, mixtures of variable com-
position were obtained but no synthetically satisfactory percentages
of 29.
[40] H. Mayr, B. Kempf, A. R. Ofial, Acc. Chem. Res. 2003, 36, 66–77.
[41] The C-glycosidation reactions on 14 and 25 did not take place in
CH2Cl2 and were very slow if carried out in MeCN below ꢀ258C.
Furthermore, the use of a mixture of BF3·Et2O and TMSOTf was
necessary, as separate reactions with either of these two Lewis acids
failed.
[42] An influence of the solvent in the stereoselectivity of C-glycosida-
tions has been previously noted: T. Mukaiyama, H. Uchiro, N.
Hirano, T. Ishikawa, Chem. Lett. 1996, 629–630.
[43] We have found only two examples where a trans-2,6-substituted tet-
rahydropyran was formed as a minor stereoisomer during the reduc-
tion of a lactol with Et3SiH under acid catalysis: a) T. Saleh, G.
Rousseau, Tetrahedron 2002, 58, 2891–2897; b) B. A. Ellsworth,
A. G. Doyle, M. Patel, J. Caceres-Cortes, W. Meng, P. P. Deshpande,
A. Pullockaran, W. N. Washburn, Tetrahedron: Asymmetry 2003, 14,
3243–3247.
[44] For reviews on the uses of this reaction for the synthesis of macrocy-
cles, see: a) J. Prunet, Angew. Chem. 2003, 115, 2932–2936; Angew.
Chem. Int. Ed. 2003, 42, 2826–2830; b) A. Gradillas, J. Pꢉrez-Cas-
tells, Angew. Chem. 2006, 118, 6232–6247; Angew. Chem. Int. Ed.
2006, 45, 6086–6101; c) K. C. Majumdar, H. Rahaman, B. Roy, Curr.
Org. Chem. 2007, 11, 1339–1365.
˘
˘
[45] A. L. Dogan, A. Dogan, H. Canpınar, ꢊ. DꢇzgꢇnÅınar, E. Demi-
rpenÅe, Chemotherapy 2004, 50, 283–288.
[46] Y.-M. Anh, K. Yang, G. I. Georg, Org. Lett. 2001, 3, 1411–1413.
[47] T. Mosmann, J. Immunol. Methods 1983, 65, 55–63.
[37] P. Deslongchamps, Stereoelectronic Effects in Organic Chemistry,
Pergamon Press, New York, 1983, Chapter 6.
[38] The same type of mechanistic considerations has been applied to nu-
cleophilic attacks at cyclic iminium cations: R. V. Stevens, Acc.
Chem. Res. 1984, 17, 289–296.
[39] a) L. Ayala, C. G. Lucero, J. A. C. Romero, S. A. Tabacco, K. A.
Woerpel, J. Am. Chem. Soc. 2003, 125, 15521–15528; b) S. Chamber-
land, J. W. Ziller, K. A. Woerpel, J. Am. Chem. Soc. 2005, 127,
Received: April 19, 2010
Revised: September 3, 2010
Published online: November 9, 2010
688
www.chemeurj.org
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 675 – 688