A ring-closing metathesis (RCM)-based approach to mycolactonesa A/B

Article abstract of DOI:10.1002/chem.201101799

The total synthesis of the mycobacterial toxins mycolactones A/B (1 a/b) has been accomplished based on a strategy built around the construction of the mycolactone core through ring-closing metathesis. By employing the Grubbs second-generation catalyst, t

Full text of DOI:10.1002/chem.201101799

FULL PAPER
DOI: 10.1002/chem.201101799
A Ring-Closing Metathesis (RCM)-Based Approach to Mycolactones A/B
Philipp Gersbach,[a] Andrea Jantsch,[a, c] Fabian Feyen,[a, d] Nicole Scherr,[b]
Jean-Pierre Dangy,[b] Gerd Pluschke,[b] and Karl-Heinz Altmann*[a]
Abstract: The total synthesis of the
mycobacterial toxins mycolactones A/B
(1a/b) has been accomplished based on
a strategy built around the construction
of the mycolactone core through ring-
closing metathesis. By employing the
Grubbs second-generation catalyst, the
12-membered core macrocycle of my-
colactones, with a functionalized C2
handle attached to C11, was obtained
in 60–80% yield. The C-linked upper
side chain (comprising C12–C20) was
completed by a highly efficient modi-
fied Suzuki coupling between C13 and
C14, while the attachment of the C5-
O-linked polyunsaturated acyl side
chain was achieved by Yamaguchi
esterification. Surprisingly, a diene con-
taining a simple isopropyl group at-
tached to C11 could not be induced to
undergo ring-closing metathesis. By
employing fluorescence microscopy
and flow cytometry techniques, the syn-
thetic mycolactones A/B (1a/b) were
demonstrated to display similar apop-
tosis-inducing and cytopathic effects as
mycolactones A/B extracted from My-
cobacterium ulcerans. In contrast, a
simplified analogue with truncated
upper and lower side chains was found
to be inactive.
Keywords:
Buruli
ulcer
·
metathesis · mycolactones · natural
products · total synthesis
Introduction
Saharan western Africa, where the prevalence is highest.[7]
Aquatic insects[8] as well as mosquitos[9] have been implicat-
ed as vectors for disease transmission, but the exact mecha-
nism of Buruli infections in humans has not been elucidated.
If left untreated, extensive infection with M. ulcerans can
lead to loss of vision, permanent impairment of joints, or
the need for amputation of affected limbs. Until 2004, the
only existing treatment option for all stages of BU was sur-
gical removal of the lesions; since then, pharmacotherapy
with a combination of oral rifampicin and intramuscular
streptomycin has been recommended by the World Health
Organization (WHO),[10,11] although this approach can be as-
sociated with severe side effects, especially in children. Im-
munization with the Mycobacterium bovis-derived BCG vac-
cine appears to provide partial, short-lasting protection
against BU, but no effective vaccine against the disease is
currently available.[12]
Mycolactones are a group of polyketide-derived 12-mem-
bered macrolides that are produced by different strains of
Mycobacterium ulcerans and that are causally involved in
the pathology associated with M. ulcerans infections in
humans, commonly known as Buruli ulcer (BU).[1,2] BU is a
severe necrotizing skin disease and it represents the third
most common mycobacterial infection in humans, after tu-
berculosis and leprosy.[3] In its initial phase the disease is
characterized by small painless nodules, which may heal
spontaneously, but frequently develop into large ulcerations
displaying massive tissue destruction with minimal inflam-
mation.[4,5] BU is generally associated with wetland and riv-
erine areas[6] in regions of tropical and subtropical climates
in Asia, the Western Pacific region, Latin America, and sub-
Contrary to infections with the closely related Mycobacte-
rium tuberculosis, bacteria in Buruli lesions are located
largely in extracellular microcolonies;[5] as indicated above,
their pathogenic effect is related to the secretion of a family
of related cytotoxins of polyketide origin, collectively re-
ferred to as mycolactones. So far, seven different mycolac-
tones (mycolactones A/B, C, D, E, and F/dia-F; Scheme 1)
have been isolated from different strains of M. ulcerans, as
well as from the related fish and frog pathogens Mycobacte-
rium marinum and Mycobacterium liflandii, respectively.[13]
[a] P. Gersbach, Dr. A. Jantsch, Dr. F. Feyen, Prof. Dr. K.-H. Altmann
Department of Chemistry and Applied Biosciences
Institute of Pharmaceutical Sciences
Swiss Federal Institute of Technology (ETH) Zꢀrich
HCI H405, Wolfgang-Pauli-Str. 10, 8093 Zꢀrich (Switzerland)
Fax : (+41)44-6331369
E-mail: karl-heinz.altmann@pharma.ethz.ch
[b] Dr. N. Scherr, J.-P. Dangy, Prof. Dr. G. Pluschke
Swiss Tropical and Public Health Institute
Socinstrasse 57, 4002 Basel (Switzerland)
Structurally, all known mycolactones are based on
a
[c] Dr. A. Jantsch
common macrolide core to which two side chains are at-
Current affiliation: CSL Behring AG, Bern (Switzerland)
ꢀ
tached via
a
C C and an ester bond, respectively
[d] Dr. F. Feyen
(Scheme 1). Structural variations between individual myco-
lactones have only been found in the polyunsaturated fatty
acid side chain attached to C5-O, whereas the extended core
Current affiliation: Carbogen-Amcis AG, Bubendorf (Switzerland)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201101799.
Chem. Eur. J. 2011, 17, 13017 – 13031
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
13017

elaborated first-,[12] second-,[19] and third-generation[20] ap-
proaches to the synthesis of mycolactones A/B (1a/b) and it
has also developed syntheses for mycolactones C,[21] E,[22]
and F.[23] Beyond the intrinsic synthetic challenges posed by
the complex molecular architecture of mycolactones, the
synthetic work has been driven by the desire to provide suf-
ficient amounts of pure material for systematic biological
studies. The most recent rendition of Kishiꢂs mycolactone
synthesis featured the construction of the core macrocycle
(with C12-C13 already appended) by Yamaguchi macrolac-
tonization, establishment of the complete C12–C20 side
chain through formation of the C13–C14 bond by Negishi
coupling, and attachment of the C1’–C16’ acyl side chain by
means of Yamaguchi esterification.[20] The synthesis of the
acyl fragment relied heavily on Horner–Wadsworth–
Emmons (HWE) chemistry for double-bond construction.
More recently, an elegant total synthesis of mycolactones A/
B (1a/b) has been reported by Negishi and co-workers,[24]
who made extensive use of Negishi-type coupling reactions
for the construction of the C1’–C16’ acyl fragment, as well
as for elaboration of the extended core macrocycle. As in all
of Kishiꢂs syntheses of different mycolactone variants, ring
closure was based on Yamaguchi macrolactonization of a
seco acid precursor, in this case comprising the entire C1–
C20 carbon framework. A formal total synthesis of mycolac-
tones has been reported by Burkart and co-workers,[25] who
prepared macrocyclic iodide 2, an advanced intermediate in
Kishiꢂs third-generation mycolactone synthesis.[13] In contrast
to Kishiꢂs and Negishiꢂs work, ring closure to the 12-mem-
bered macrocycle in Burkartꢂs case was based on ring-clos-
ing olefin metathesis (RCM), rather than a macrolactoniza-
tion reaction; the RCM reaction proceeded in excellent
yield and with exclusive E selectivity.
Scheme 1. Molecular structures of mycolactones A–F. The macrolactone
core is highlighted in red, the C-linked C12–C20 side chain is colored in
blue, and the polyunsaturated fatty acid side chain is shown in black. We
refer to the macrolactone core with the C12–C20 side chain attached
(red and blue parts) as the “extended core (structure)”.1 The stereochem-
istry of mycolactone D has not been fully elucidated.
structure (i.e., the macrolide ring with the C-linked side
chain) is fully conserved.1 Importantly, mycolactones A/B
represent a rapidly equilibrating mixture of Z-D4’,5’ (myco-
lactone A, 1a) and E-D4’,5’ (mycolactone B, 1b) geometric
isomers; this mixture is isolated as such from bacterial cul-
tures. Mycolactones exhibit apoptosis-inducing[13] and immu-
nosuppressive[14] properties, but marked differences in po-
tency levels have been reported between individual struc-
tures. The most potent variants are mycolactones A/B (1a/
b), with the other mycolactones becoming progressively less
potent in alphabetical order.[15] Although mycolactones A/B
(1a/b) are produced by nearly all strains of M. ulcerans,
highly virulent African and Malaysian strains produce the
largest quantities and greatest diversity of mycolactones.[16]
Injection of mycolactones A/B into guinea pigs produces ne-
crotic lesions that resemble BU pathology; infection with
mycolactone-negative M. ulcerans fails to induce ulceration
in this animal model, but chemical complementation with
purified mycolactone restores wild-type pathology.[1]
Our own synthetic efforts on mycolactones[26] have been
driven by the desire to access sufficient quantities of homo-
geneous material for biological studies; we also envisaged
the possibility of producing antibodies against these toxins,
which might be useful for the development of diagnostic
tools for the early detection of M. ulcerans infections. At
the same time, the chemistry developed for the synthesis of
natural mycolactones was to provide the basis for the prepa-
ration of analogues for structure–activity relationship (SAR)
studies, which so far have been largely limited to natural
mycolactone variants. Our synthetic strategy towards myco-
lactones was designed around the formation of the core
macrolactone by RCM between C8 and C9 and subsequent
The chemistry of mycolactones has been pioneered by
Kishi and co-workers, who have also established the relative
and absolute configurations of these natural products by ex-
tensive NMR spectroscopic studies.[18] The Kishi group has
1
In a slight deviation from Kishiꢂs terminology, we refer to the C1–C11
ꢀ
macrolactone ring (including the OH group on C5, but without the C
C-linked C12–C20 side chain) as the mycolactone core (structure) and
to the entire C1–C20 fragment as the extended core (structure).
13018
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 13017 – 13031

Total Synthesis of Mycolactones A/B Based on RCM
FULL PAPER
incorporation of the C- and O-linked side chains by suitable
fragment couplings. In this context, we have previously com-
municated (prior to the publication of Kishiꢂs third-genera-
tion approach to mycolactones) the synthesis of macrocyclic
alkyl iodide 3,[26] which is a differently protected version of
Kishiꢂs advanced intermediate 2.
In this paper, we now disclose full details on the opti-
mized synthesis of 3 and its elaboration into mycolactone-
s A/B (1a/b). The apoptosis-inducing and cytopathic effects
of synthetic 1a/b were investigated by means of fluorescence
microscopy and flow cytometry. We have also prepared ana-
logue 4 (see above), which has allowed assessment of the in-
herent biological activity of the mycolactone core macrocy-
cle.
4 was envisioned to be accomplished by a Pd-mediated cou-
pling reaction, either of the Negishi- or the (modified)
Suzuki-type. The coupling product, after selective deprotec-
tion of the C5 hydroxyl group, would then be esterified with
I-2 to provide fully protected 1a/b. Vinyl iodide I-4 and acid
I-2 have been described previously by Kishi and co-work-
ers[19] and we planned to follow the published syntheses of
these intermediates with some modifications; for I-2, this
would involve Horner–Wadsworth–Emmons (HWE) cou-
pling of a phosphonate I-5 and aldehyde I-6 (Scheme 2),
which display similar levels of complexity. Alkyl iodide I-3,
comprising the core macrolactone of mycolactones with an
appended C2 fragment at C11, was to be obtained through
ring-closing olefin metathesis (RCM)[2a,27] between C8 and
C9. The diene precursor for the cyclization reaction was in
turn to be formed by the esterification of alcohol I-7 with
acid I-8. Although this RCM-based cyclization strategy
promised to offer efficient access to the core macrocycle, its
successful implementation was by no means ensured at the
time of its conception, since no examples had been reported
in the literature on RCM-based ring closure to a trisubstitut-
ed E-olefin.
Results and Discussion
Retrosynthesis: Given the overall objectives of our mycolac-
tone program, we were aiming for a synthesis that would be
modular in character, since this would 1) ensure a reasona-
ble potential for scale-up, should it be required; 2) enable
independent modifications of individual parts of the struc-
ture for SAR studies; and 3) allow the preparation of conju-
gates required for in vitro antibody selection. Within these
boundary conditions, the retrosynthetic analysis of 1a/b led
to two major disconnections, producing the polyunsaturated
fatty acid I-2, alkyl iodide I-3, and vinyl iodide I-4 as three
highly advanced intermediates (Scheme 2). The establish-
ment of the C13–C14 bond through connection of I-3 and I-
Synthesis of protected core macrocycle 3: According to the
above retrosynthesis of 1a/b, access to the protected, C13-
functionalized macrocycle I-3 required the synthesis of an
appropriately monoprotected diol I-7 and a carboxylic acid
building block I-8. As regards the protection of the primary
hydroxyl function in I-7, we felt that the use of a tosyl
group, although not commonly perceived as a protecting
Scheme 2. Retrosynthesis of mycolactones A/B (1a/b). PG=protecting group or H; protecting groups may vary independently.
Chem. Eur. J. 2011, 17, 13017 – 13031
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
13019

K.-H. Altmann et al.
group, would offer the advantage of simultaneously serving
as a protecting group (during ester formation) and as an ac-
tivating group for nucleophilic substitution reactions at C13
subsequent to macrocycle formation. The synthesis of the
corresponding alcohol 5 (Scheme 3) was initially pursued
Scheme 3. a) PMB-trichloroacetimidate, TfOH, Et2O, RT, 2 h, 58%.
b) DIBAL-H, CH2Cl2, ꢀ788C, 2 h, quant. c) Allyl-Sn-nBu3, SnCl4,
CH2Cl2, ꢀ908C, 1.5 h, 81%, de>20:1. d) (1) DDQ, H2O, CH2Cl2, RT,
1.75 h; (2) LiAlH4, Et2O, 0 8C!RT, 1 h, 76% (two steps). e) TsCl, Et3N,
DMAP, CH2Cl2, 358C, 3 h, 86%.
Scheme 4. a) PMB-Cl (0.15 equiv), NaH, THF, reflux, 18 h, 97%.
b) (COCl)2, DMSO, Et3N, CH2Cl2, ꢀ788C!RT, 1 h, 99%. c) (1) N-Pro-
pionyl-(2R)-bornane-(10,2)-sultam, Et2BOTf, CH2Cl2, ꢀ58C, 20 min;
(2) 15, ꢀ788C, 1.25 h, 83%, de>95%. d) LiAlH4, THF/Et2O, 0 8C!RT,
2.5 h, 78%. e) TsCl, Et3N, DMAP, CH2Cl2, 08C!RT, 16 h, 96%. f) NaH,
THF, RT!408C, 8 h, 98%. g) Isopropenyllithium, BF3·Et2O, Et2O,
ꢀ788C, 20 min, 90%. h) TESOTf, 2,6-lutidine, CH2Cl2, ꢀ788C!RT, 1 h,
98%. i) DDQ, CH2Cl2, buffer pH 7.2, RT, 45 min, 92%. j) (1) DMP,
CH2Cl2, 08C, 2 h; (2) NaClO2, 2-methyl-2-butene, tBuOH, NaH2PO4,
H2O, RT, 30 min, 91% (two steps). DMP=Dess–Martin periodinane.
through the allylation of tosylate 6, which was derived from
(S)-(+)-3-hydroxy-2-methylpropionic acid methyl ester ((S)-
Roche ester, 7; Scheme 3) by
tosylation and subsequent re-
duction with diisobutylalumi-
num hydride (DIBAL-H) in
90% yield.
Unfortunately, direct allylation of 6 was not a practically
viable option for the synthesis of 5. Although the use of al-
lyltributyltin gave the allylation product in good yield (79%,
including traces of residual tin derivatives), it was obtained
as an inseparable mixture of diastereoisomers (4:6 in favor
of the undesired syn isomer). Brown allylation[28] with
(+)-Ipc2B(CH2CH=CH2) (Ipc=isopinocampheyl), on the
other hand, yielded the desired diastereoisomer with good
selectivity (d.r.> 90%), but only in low yields (<20% after
extensive chromatographic purification).
In contrast to 6, p-methoxybenzyl (PMB)-protected alde-
hyde 9, obtained from 7 by reaction with PMB-trichloroace-
timidate followed by DIBAL-H reduction, reacted with al-
lyltributyltin under chelation control to provide the desired
homoallylic alcohol 10 in good yield and with excellent dia-
stereoselectivity (> 20:1). 2,3-Dichloro-5,6-dicyano-1,4-ben-
zoquinone (DDQ)-mediated cleavage of the PMB ether
moiety in 10 resulted in a mixture of diol 11 and the corre-
sponding p-methoxybenzoate. This mixture was directly
treated with LiAlH4, thus furnishing the desired diol 11 in
76% yield. Subsequent selective tosylation yielded building
block 5 with excellent efficiency (86% yield).
then submitted to an aldol reaction with the boron enolate
of N-propionyl-(2R)-bornane-(10,2)-sultam[29] to yield syn-
aldol product 16 in 83% yield as a single isomer; reductive
removal of the chiral auxiliary with LiAlH4 gave diol 17. In
our original strategy, we had
envisaged the elaboration of 17
into protected alcohol 21
(Scheme 4) through Cu-cata-
lyzed coupling of iodide 23 with
isopropenylmagnesium
mide.
bro-
However, all attempts to achieve this transformation were
unsuccessful,[30] thus forcing us to explore alternative routes
for the conversion of 17 into 21. Based on previous work by
Carreira and Du Bois[31] on a related transformation, we in-
vestigated the reaction of the primary tosylate 18, obtained
from 17 by reaction with TsCl in 96% yield, employing a
two-step protocol that involved treatment of 18 with
1.5 equiv of isopropenyllithium at ꢀ208C!RT for a total of
105 min, followed by an additional 3 equivalents of isopro-
penyllithium and 3 equiv of BF3·Et2O at ꢀ788C for 25 min.
This procedure gave the desired secondary alcohol 20 in
yields of up to 61%, but in general yields with this initial
protocol were moderate and the reaction lacked reproduci-
bility. However, further investigations revealed that 18 could
be converted into oxetane 19 by treatment with a slight
excess of either phenyllithium or sodium hydride. Oxetane
19 could be isolated in excellent yield (thereby also confirm-
The synthesis of carboxylic acid 12 (corresponding to I-8
in Scheme 2) started from 1,5-pentanediol (13), which was
converted into the PMB-protected hydroxyaldehyde 15 by
reaction with 0.15 equivalents of PMBCl (to produce mono-
PMB ether 14) and subsequent Swern oxidation of the re-
maining free hydroxyl group (Scheme 4). Aldehyde 15 was
13020
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 13017 – 13031

Total Synthesis of Mycolactones A/B Based on RCM
FULL PAPER
ing its intermediacy in the in situ transformation of 18 into
20); of the two bases, NaH was found to be more reliable
and give higher yields (98%, compared with 71% for PhLi).
Reaction of oxetane 19 with an excess of isopropenyllithium
resulted in regioselective opening of the oxetane ring to fur-
nish the desired alcohol 20, again in excellent yield. Impor-
tantly, a substantial excess of isopropenyllithium (>9 equiv)
was required to drive the reaction to completion. With alco-
hol 20 in hand, acid 12 could be accessed in a straightfor-
ward manner in four further steps. Thus, compound 20 was
converted into its corresponding TES ether 21, which was
followed by oxidative cleavage of the PMB ether to yield
the primary alcohol 22. Oxidation of 22 with Dess–Martin
periodinane[32] and subsequent Pinnick–Kraus oxidation[33]
of the resulting aldehyde finally gave the desired carboxylic
acid 12 in ten steps and 43% overall yield from 1,5-pentane-
diol (13). Acid 12 could be obtained on a multi-gram scale
by this route.
Similar results were obtained
for the RCM-based ring closure
of the C13-O TBDPS-protected
diene 26 (single experiment),
which produced exclusively the
E-configured macrocycle.2 This
reaction appeared to be slower
than the cyclization of 25, thus
pointing to a dependence of reaction rate on the nature of
the C13 substituent (see below).
The stereochemical outcomes of the RCM reactions with
24 and 26 are in line with the observations reported by Bur-
kart and co-workers[25b] for the cyclization of dienes closely
related to 24 and 26, which provided the corresponding E-
configured RCM products with high selectivity. Tosylate 25
could be converted into iodide 3 in excellent yield (95%) by
treatment with NaI in acetone at 608C for 9 h. (Reaction
times between 6.5 and 14 h all delivered excellent yields of
3. In a single initial experiment, heating of the reaction mix-
ture for 24 h led only to unidentified products, while none of
the desired 3 was obtained).
With both alcohol 5 and acid 12 in hand, we then turned
our attention towards the crucial ring-closure reaction and
the subsequent conversion of the tosyl group into the re-
quired iodo substituent. Esterification of 5 with 12 under
Hçfle–Steglich conditions[34] yielded diene 24 in 82% yield
(Scheme 5). Treatment of the latter with the second-genera-
Synthesis of vinyl iodide 27: The synthesis of 27 has been
described previously by Kishi and co-workers[19] and we
have followed the lead provided by this work with only
minor modifications (Scheme 6). Briefly, commercially avail-
able (R)-3-hydroxybutanoic acid methyl ester (28) was ela-
borated into aldehyde 30 by a sequence involving tert-butyl-
dimethylsilyl (TBS) protection, DIBAL-H reduction (to
give 29), asymmetric crotylboration of 29, TBS protection of
the newly formed hydroxyl group, and ozonolysis of the
double bond. The yield obtained in the crotylboration step
(39%) was significantly lower than that reported by Kishi
(78%)[19] and, more recently, by Negishi (88% for a 95:5
mixture of diastereomers)[24b] for the same transformation,
Scheme 5. a) DCC, DMAP, CH2Cl2, RT, 15 h, 82%. b) Grubbs II
(12 mol%), CH2Cl2 (3 mm), reflux, 6.5 h, 80%. c) NaI, acetone, 608C,
9 h, 95%.
tion Grubbs catalyst[35] in refluxing CH2Cl2 led to efficient
ring closure to produce the 12-membered lactone 25 (with
the desired E geometry of the C8–C9 double bond; myco-
lactone numbering) in isolated yields of 60–80% in most
cases; on some occasions, however, the yield from the RCM
reaction could be as low as 40% without any obvious
changes in the external reaction parameters. The isolated
yield of 25 appeared to depend on the time required for
complete consumption of the starting diene, which could
vary between 2.5 and 24 h under ostensibly identical condi-
tions. The best results were obtained when the diene had
completely disappeared after about 6 h, with both shorter
and longer time periods being associated with the formation
of significant amounts of unidentified side products that
were difficult to remove. The ring closure seems to be
highly E-selective, as 25 was the only isolable cyclization
product; however, as we did not rigorously pursue the isola-
tion of the corresponding Z isomer we cannot exclude the
possibility that the latter is formed in small amounts and
simply lost during chromatographic purification.
Scheme 6. a) TBS-Cl, imidazole, DMF, RT, 2.5 h, 88%. b) DIBAL-H,
hexane, ꢀ788C, 1.75 h, 70%. c) (Z)-2-Butene, nBuLi, KOtBu, (ꢀ)-
Ipc2BOMe, BF3·Et2O, THF/Et2O, ꢀ788C, 3 h, 39%. d) TBS-Cl, imida-
zole, DMF, RT, 16 h, 89%. e) (1) O3, CH2Cl2, ꢀ788C, 2 min; (2) PPh3,
ꢀ788C!RT, 1 h. f) CBr4, PPh3, Et3N, 0 8C, 15 min, 82% (two steps).
g) nBuLi, THF, ꢀ788C!08C!ꢀ788C, 1 h. h) MeI, ꢀ788C!RT, 1 h,
99% (two steps). i) (1) Cp2ZrHCl, THF, 408C, 3 h; (2) I2, 08C, 2 min,
67%.
2
Diene 26 and its derived RCM product were intermediates in the prep-
aration of immobilizable conjugates of mycolactone fragments for bio-
logical studies. Details of this work, including the synthesis of 26, will
be reported in a separate publication.
Chem. Eur. J. 2011, 17, 13017 – 13031
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
13021

K.-H. Altmann et al.
but no efforts were made to optimize the reaction. Alde-
hyde 30 was then converted into dibromoolefin 31 by means
of Corey–Fuchs chemistry[36] in 82% yield; compound 31
was converted to the corresponding terminal acetylene,
which was then methylated to give 32. Subsequent hydrozir-
conation/iodination[37] furnished the desired building block
27, which was obtained from (R)-3-hydroxybutanoic acid
methyl ester (28) in nine steps and 12% overall yield.
methanolysis (to give 38), acetal hydrolysis, global TBS pro-
tection, and dithiane cleavage, in 64% overall yield for the
four-step sequence from 37. Finally, aldehyde 33 was ob-
tained as a single isomer through homologation of 39 ac-
cording to the procedures reported by Kishi[19] in 81% over-
all yield.
The synthesis of phosphonate 34 is depicted in Scheme 8
and it closely follows the route developed previously by
Gurjar and Cherian.[41] The major exception, apart from the
preparation of aldehyde 42,[42] is the direct, one-step conver-
sion of the latter into the homologated analogue 43 by reac-
tion with 2-(triphenylphosphoranylidene)-propionaldehyde,
which proceeded in high yield (82%). Unfortunately, subse-
quent two-carbon extensions were not possible with this
phosphorane and therefore had to be performed with the
more reactive ethoxycarbonyl derivative, as described by
Gurjar and Cherian. This necessitated an additional reduc-
tion/oxidation sequence after each elongation cycle to reach
the aldehyde oxidation state of the homologated product.
Phosphonate 34 was finally obtained from ethanediol (40)
as a single isomer in ten steps and 26% overall yield. Cou-
pling of 34 with aldehyde 33 proceeded in excellent yield
(90%), but the product 50 could only be obtained as an in-
separable mixture of E-D4’,5’ and Z-D4’,5’ isomers in a ratio of
84:16. Subsequent ester hydrolysis with LiOH in THF/
water/MeOH gave the desired penta-unsaturated acid 51 in
essentially quantitative yield as an approximately 70:30 mix-
ture of E-D4’,5’ and Z-D4’,5’ isomers, together with traces of
Synthesis of the polyene side chain: The synthesis of the
protected polyunsaturated fatty acid side chain(s) of myco-
lactones A/B (1a/b) was based on the same overall strategy
that was developed for this fragment by Kishi and co-work-
ers, and which involves final assembly of the full-length
carbon chain through HWE coupling of aldehyde 33
(Scheme 7) with phosphonate 34 (Scheme 8). Aldehyde 33
Scheme 7. a) (1) SO2Cl2, pyridine, CHCl3, ꢀ788C!508C, 5 h; (2) NaI,
MeOH/H2O, 0 8C, 10 min, 76% (two steps). b) H2, Raney Ni, KOH,
EtOH, 3 bar, RT, 4 h, 60%. c) (1) 1,3-Propanedithiol, conc. HCl, RT,
14 h; (2) cat. H2SO4, CuSO4, acetone, RT, 16 h, 80%. d) Ph3P, PhCOOH,
DIAD, THF, RT, 1 h, 75%. e) NaH, MeOH, RT, 2 d, 95%. f) cat. TFA,
AcOH/H2O, RT, 8 h, quant. g) TBS-Cl, imidazole, DMF, 708C, 24 h,
96%. h) MeI, 2,4,6-collidine, acetone/H2O, reflux, 10 h, 97%. i) Ph3P=C-
(CH3)COOEt, benzene, reflux, 22 h, 93%. j) DIBAL-H, CH2Cl2, ꢀ788C,
60 min, 91%. k) SO3·pyridine, DIEA, CH2Cl2/DMSO, RT, 60 min, 97%.
DIAD=diisopropyl azodicarboxylate.
in Kishiꢂs synthesis was derived from the shorter aldehyde
39, which in turn could be obtained in seven steps from (S)-
3-hydroxybutanoic acid methyl ester, albeit as a 3.8:1 mix-
ture of diastereoisomers.[19] We have thus explored an alter-
native approach to aldehyde 39 that utilizes d-glucose as an
inexpensive chiral starting material and that provides this in-
termediate as a single isomer. This approach was inspired by
previous work by Feringa, Minnaard, and van Summeren,[38]
who described the elaboration of methyl-a-d-glucopyrano-
side (35) into dithiane 36 (Scheme 7), building, in turn, on
extensive prior work by Jennings and Jones[39] on the prepa-
ration of chlorodeoxy sugars. Thus, compound 35 was con-
verted into methyl-4,6-dichloro-4,6-dideoxy-a-d-galactopyra-
noside by the method of Jennings and Jones in 76% yield;
this was followed by chloride reduction with H2 over Raney
Ni, acetal opening with a dithiane anion, and in situ aceto-
nide formation to give alcohol 36. Mitsunobu inversion[40]
then gave 37, which was elaborated into aldehyde 39 by
Scheme 8. a) TBS-Cl, pyridine, CH2Cl2, RT, 23 h, 87%. b) (COCl)2,
DMSO, pyridine, Et3N, CH2Cl2, ꢀ788C, 20 min, 84%. c) Ph3PC=C-
ACHUNTGRENN(UG CH3)CHO, benzene, RT, 2 d, 71%. d) Ph3PC=CCAHTUNGTRENN(UGN CH3)COOEt, benzene,
reflux, 3 h, 81%. e) DIBAL-H, CH2Cl2, ꢀ788C, 40 min, 87%. f) MnO2,
CHCl3, RT, 20 h, 98%. g) Ph3P=CHCOOEt, benzene, reflux, 4 h, 93%.
h) TBAF, THF, RT, 1 h, 92%. i) PBr3, Et2O, 0 8C, 4 h, 95% (crude). j) P-
AHCTUNGTRENNUNG
13022
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 13017 – 13031

Total Synthesis of Mycolactones A/B Based on RCM
FULL PAPER
other isomeric impurities (<2%) that were not separated at
this point.
change, and ZnCl2 in Et2O at ꢀ1008C, followed by reaction
of the presumed zincate with vinyl iodide 27 and PdACHTUNGTRENNUNG(PPh3)4
In an attempt to reduce the number of steps leading from
methyl-a-d-glucopyranoside (35) to a suitably protected ver-
sion of the mycolactone side chain acid, we also prepared al-
dehyde 53 (Scheme 9). The latter was obtained in three
in THF at RT, gave none of the desired coupling product.
Likewise, attempts to achieve the coupling of 27 and 3
under modified Suzuki conditions, that is, via the ate com-
plex formed from the organolithium species derived from 3
by reaction with 9-MeO-BBN, were unsuccessful. When
contemplating possible ways to address this situation, we
noted that difficulties in the sp2–sp3 coupling of a 1-iodo-2-
methyl-3-TBS-oxy-1-alkene had also been encountered by
Arefolov and Panek in the context of their work on disco-
dermolide;[45] as in our case, none of the desired coupling
product of the vinyl iodide with a primary alkyl iodide was
observed under either Negishi or modified Suzuki condi-
tions. Arefolov and Panek were able to overcome this prob-
lem by changing the hydroxyl protecting group from TBS to
methoxymethyl (MOM), with the modified substrate under-
going Negishi as well as modified Suzuki coupling in good
(62%) to very good (82%) yields, respectively. Although we
were mindful of the fact that the use of MOM protecting
groups might be sub-optimal in a 1,3-diol system and the
conditions required for the ultimate removal of these groups
would be of limited compatibility with the nature of the un-
saturated fatty acid side chain(s) in mycolactones A/B,[11b]
we still felt it important to establish whether similar effects
might be operative in our system as in the one described by
Arefolov and Panek. Bis-TBS-protected vinyl iodide 27 was
thus converted into the corresponding bis-MOM derivative
56 in 71% overall yield by TBS cleavage with HF·pyridine
and subsequent treatment of the free diol 55 with MOMCl
in the presence of diisopropylethylamine (DIEA)
(Scheme 10).
Scheme 9. a) TBS-Cl, imidazole, DMF, 708C, 14 h, 86%. b) MeI, 2,4,6-
collidine, acetone/H2O, reflux, 16 h, 90% (crude). c) Ph3P=CACHTNUGTRNEUNG(CH3)CHO,
benzene, RT, 2 h, 91%. d) 34, LDA, THF, ꢀ788C!RT, 3 h, 78%.
steps from dithiane 38 by silylation, dithiane cleavage, and
one-step homologation of the resulting aldehyde (52) with
2-(triphenylphosphoranylidene)-propionaldehyde (as com-
pared with six steps for the elaboration of 38 into aldehyde
33). The HWE reaction of 53 and 34 proceeded smoothly
and gave 54 in yields between 78 and 90% as an approxi-
mately 3:1 mixture of E-D4’,5’ and Z-D4’,5’ isomers, together
with some uncharacterized minor isomers. However, explor-
atory preliminary experiments with 54 on the simultaneous
removal of the acetonide and TBS protecting groups under
a number of conditions, including AcOH/THF/water, mont-
morillonite K10, pyridinium p-toluenesulfonate (PPTS)/
MeOH, trimethylsilyl bromide (TMSBr), or HF·pyridine at
different temperatures, returned only decomposition prod-
ucts. Selective cleavage of the TBS ether could be accom-
plished in excellent yield (95%) with TBAF and the desily-
lated product was also observed in the reaction with HF·pyr-
idine after short reaction times, but attempts to convert the
mono-deprotected product into the desired free triol were
again unsuccessful. In light of these findings, compound 54
was not further processed for use in the synthesis of myco-
lactones A/B (1a/b), due to concerns about difficulties in
the final deprotection steps.
Scheme 10. a) HF·pyridine, THF, RT, 3 h, 90%. b) MOMCl, DIEA/
CH2Cl2 (1:1), RT, 15 h, 79%. c) tBu2SiACHTUNTRGNEUNG(OTf)2, pyridine, CH2Cl2, 08C, 1 h,
90%.
Gratifyingly, modified Suzuki coupling of 56 with 3 pro-
duced the desired coupling product 58 in 40% yield
(Scheme 11). Although the efficiency of the reaction still
left room for improvement, we had now established a gener-
al proof-of-principle for the construction of the C13–C14
bond (mycolactone numbering) by means of Suzuki cross-
coupling, which provided the foundation for subsequent op-
timization efforts.
Assembly of fragments into mycolactones A/B: As outlined
in the section on retrosynthesis, the assembly of building
blocks 3, 27, and 51 into mycolactones A/B in a first step
was to involve the establishment of the C13–C14 bond
through a Pd0-catalyzed cross-coupling reaction between
vinyl iodide 27 and alkyl iodide 3, either under Negishi[19,43]
or modified Suzuki[44] conditions. Unfortunately, neither of
these concepts could be successfully implemented. Thus,
treatment of 3 with tBuLi, to affect halogen–lithium ex-
Preliminary experiments on the removal of the MOM
protecting groups from 58, after removal of the triethylsilyl
Chem. Eur. J. 2011, 17, 13017 – 13031
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
13023

K.-H. Altmann et al.
Scheme 11. a) (1) 3, 9-MeO-BBN, tBuLi, Et2O, THF, ꢀ788C!RT, 1.5 h;
(2) 56, [Pd(dppf)Cl2], AsPh3, Cs2CO3, DMF, RT, 4 h, 40%.
ACHTUNGTRENNUNG
(TES) group from C5-O, gave only low yields of the free
diol. For example, treatment of 58 with Me2SiCl2/nBu4N+
Brꢀ in CH2Cl2 at 08C, conditions that have been shown to
prevent methylene acetal formation upon cleavage of MOM
groups from a 1,3-diol in another case,[46] provided only a
19% yield of a 1:2 mixture of the free triol and the methyl-
ene acetal of the 1,3-diol moiety. No attempts were made to
optimize the deprotection conditions; instead, we investigat-
ed the modified Suzuki coupling of alkyl iodide 3 with vinyl
iodide 57 (Scheme 10) as an alternatively protected coupling
partner. Vinyl iodide 57 incorporates a cyclic silyl protecting
group, which we expected to be removable under conditions
that would be well compatible with the structural features of
the mycolactone framework, and which would also maintain
the possibility of final removal of all hydroxyl protecting
groups in a single step. At the same time, we reasoned that
the inclusion of the silyl protecting group on C17-O (myco-
lactone numbering) in a ring structure (rather than being
part of a separate TBS ether moiety) would reduce steric
hindrance in the oxidative addition step, which is what Are-
folov and Panek had hypothesized to be the main reason for
the failure of the cross-coupling reactions with their TBS-
protected substrate.[45] Vinyl iodide 57 was prepared from
the bis-TBS derivative 27 by desilylation and subsequent re-
Scheme 12. a) (1) 3, 9-MeO-BBN, tBuLi, Et2O, THF, ꢀ788C!RT, 1 h;
(2) 57, [PdACTHUNTRGNE(UNG dppf)Cl2], AsPh3, Cs2CO3, DMF, RT, 21 h, 80%. b) THF/
H2O/AcOH (2:1:1), RT, 2.5 h, 90%. c) 51, Cl3C6H2COCl, DIEA, DMAP,
THF, RT, 2 h, 89%. d) HF·pyridine, THF/pyridine (4:1), RT, 1 h, 84%.
e) TBAF, THF, RT, 75 min, 85%.
the observed differences in the outcomes of the (attempted)
coupling reactions.
With the protected extended core of mycolactones in
hand, completion of the total synthesis of mycolactones A/B
in the next step required the selective removal of the TES
protecting group from C5-O. The reaction was best carried
out in a mixture of THF/H2O/AcOH in a ratio of 2:1:1,
which furnished 60 in 90% yield (Scheme 12); treatment of
59 with HF·pyridine in THF for 3 h at RT proved to be less
efficient and provided 60 in only 55% yield. Esterification
of 60 with the unsaturated fatty acid side chain mixture 51
under Yamaguchi conditions proceeded smoothly and gave
fully protected mycolactones A/B (61) as an approximately
1:1 mixture of E-D4’,5’ and Z-D4’,5’ isomers in 89% yield
(other double-bond isomers were present at <1%). When
61 was treated with TBAF in THF, only the TBS protecting
groups were removed. On the other hand, when treated
with concentrated HF·pyridine in THF/pyridine (3:2) for
28 h, mycolactones A/B (1a/b) were obtained in 27% yield
after reverse phase (RP)-HPLC purification.3 We ascribe
the moderate yield to the instability of the pentaene side
chain under acidic reaction conditions. As a consequence,
we resorted to stepwise removal of the protecting groups
action of the free diol 55 with tBu2SiACTHNUTRGNE(UNG OTf)2 in the presence
of pyridine (Scheme 10; 81% yield for the two steps). Intri-
guingly, the modified Suzuki cross-coupling reaction of 57
with 3 proceeded smoothly (Scheme 12), with yields of the
desired coupling product 59 in initial experiments being
around 50%; under optimized conditions, compound 59
could ultimately be obtained in 80% yield, thus making the
modified Suzuki cross-coupling between 57 and 3 a viable
approach for the construction of the extended mycolactone
core structure. It should be noted at this point that subse-
quent to our studies on Negishi and modified Suzuki cross-
couplings involving alkyl iodide 3, the Kishi group reported
the successful and high-yielding (88%) Negishi coupling of
iodide 2 with vinyl iodide 27. (Iodide 2 differs from 3 in that
it has a PMB rather than a TES protecting group on C5-O).
In contrast to our own experiments with 3, the alkylzinc
iodide derived from 2 was generated by Kishi and co-work-
ers with an active Zn-Cu couple rather than by Li to Zn ex-
change. It remains to be investigated to what extent the dif-
ferent modes of preparation of the alkylzinc iodide and/or
the different protecting groups on C5-O are responsible for
3
The experiment was conducted with 5.3 mg of 61 and gave 1.6 mg of
deprotected material (50%, ca. 59% purity), which was submitted to
RP-HPLC purification.
13024
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 13017 – 13031

Total Synthesis of Mycolactones A/B Based on RCM
FULL PAPER
from the individual side chains, which delivered mycolacto-
nes A/B in significantly improved overall yield. Thus, treat-
ment of 61 with HF·pyridine in the presence of an excess of
pyridine resulted in clean cleavage of the cyclic bis-silyl
ether and furnished diol 62 in 84% yield; subsequent re-
moval of the three TBS protecting groups on the unsaturat-
ed acyl side chain was then accomplished with TBAF in
THF to finally provide mycolactones A/B (1a/b) in 71%
overall yield from the fully protected precursor 61 as an ap-
proximately 1:1 mixture of E-D4’,5’ and Z-D4’,5’, along with
some minor double-bond isomers (ca. 10%). RP-HPLC pu-
rification of the mixture gave mycolactones A/B (1a/b) of
95% HPLC purity in a final yield of 45%. All spectroscopic
data for 1a/b were in full agreement with literature data,
except that the ratio of 1a:1b was not fully identical with
that from previous preparations. Products 1a and 1b could
be separated by preparative RP-HPLC and were also char-
acterized individually.
When protected from light, mycolactone A (1a) under-
goes very slow isomerization in [D6]acetone solution (1a/1b
ratio of 100:7 after 7 days), while mycolactone B (1b) does
not isomerize over the same time period. Finally, it should
be noted that thorough examination of the 13C NMR spec-
trum of diol 62 in [D6]acetone revealed an additional set of
downfield-shifted signals in the C16–C20 region, which was
not observed in the spectra of either 61, 1a, or 1b. Shift dif-
ferences were most pronounced for satellite peaks of the hy-
droxyl-bearing carbons C17 and C19, but were also detecta-
ble for carbons C16, C18, and C20. Although we cannot pro-
vide a fully satisfactory explanation for this phenomenon,
we have also observed the appearance of additional down-
field-shifted signals in the 13C NMR spectra of simple 1,3-
diols in [D6]acetone, but not in CDCl3.
Scheme 13. a) Allyl-Sn-nBu3,
Ti
(O-iPr)4,
(S)-(ꢀ)-1,1’-bi-2-naphthol,
CH2Cl2, MS, ꢀ788C!ꢀ188C, 16 h, 47%, single isomer. b) 12, EDCl,
DMAP, CH2Cl2, RT, 4 h, 71%. c) Grubbs II, Grubbs I, or Hoveyda–
Grubbs II; CH2Cl2 or toluene. d) NaBH4, DMSO, 1008C, 30 min, 76%.
e) AcOH/THF/H2O (2:2:1), RT, 3 h, 98%. f) Sorbic acid, EDCl, DMAP,
CH2Cl2, RT, 15 h, 99%.
As an alternative to the RCM-based cyclization of 65, we
then examined whether tosylate 25 would represent a suita-
ble precursor for lactone 66. Gratifyingly, reduction of 25
proved to be possible with excess NaBH4 in DMSO at
1008C, thus providing access to 66 in good yields
(Scheme 13). In contrast, tosylate 25 could not be reduced
with CuCl2·2H2O/Li-DTBB (DTBB=di-tert-butylbiphenyl);
likewise, attempts at the reduction of iodide 3 with Na-
Synthesis of analogue 4: To assess the inherent biological ac-
tivity of the acylated mycolactone core macrocycle, we have
also pursued the synthesis of the simple model compound 4,
which represents a significantly truncated version of the nat-
ural toxins. Initially, the synthesis of 4 was planned to in-
volve RCM-based ring closure of diene 65 (Scheme 13). To
this end, chiral alcohol 64 was prepared from isobutyralde-
hyde as a single isomer in yields of up to 47% by catalytic
asymmetric allylation with allyl-tri-n-butylstannane[47]
(Scheme 13). The modest isolated yield obtained from the
allylation reaction can be attributed to the high volatility of
64, which hampered complete recovery of the product.
Esterification of 64 with acid 12 under Hçfle–Steglich condi-
tions[34] then provided diene 65. Very much to our surprise,
however, 65 failed to undergo RCM under the conditions
that had proven effective for the cyclization of 24 as well as
of 26. In fact, 65 could not be induced to cyclize under any
of the conditions investigated, which included the use of dif-
ferent catalysts (Grubbs II,[35] Grubbs I,[48] or Hoveyda–
Grubbs II[49]) and solvents (CH2Cl2 or toluene). We have no
coherent explanation for this outcome; apparently, the re-
duced size of the C11 substituent group in 65 affects the
RCM in an unfavorable way.[50]
ACHTUNGTREN(UNNG BH3CN) were unsuccessful. The silyl protecting group of 66
was then removed under acidic conditions to give the free
alcohol 67. Finally, analogue 4 was obtained in almost quan-
titative yield by Hçfle–Steglich esterification[34] of 67 with
sorbic acid.
Biological evaluation: As discussed in the Introduction, my-
colactones A/B (1a/b) have both apoptosis-inducing and cy-
topathic effects; they induce cytoskeletal rearrangements
and cause cell-cycle arrest.[13a] In the present study, we have
tested whether the cytotoxic activity of synthetic mycolacto-
nes A/B (1a/b) is comparable to that described for mycolac-
tones A/B extracted from M. ulcerans. To this end, morpho-
logical changes of mycolactone-treated murine L929 fibro-
blasts were analyzed by microscopy. Fibroblasts are the
most common cells in connective tissue, play a critical role
in wound healing, and are sensitive to extracted mycolac-
tones.[1,14,51–53] When treated for 24 h with synthetic mycolac-
tones A/B (1a/b) at concentrations ꢁ15 ngmLꢀ1 (ca. 20 nm),
L929 fibroblasts lost their capacity to form fiber-like appen-
dices and they displayed a rounded cell shape (Figure 1(A)).
After 48 h, the cytopathic effects were more pronounced,
leading to complete detachment of cells, blebbing of the
Chem. Eur. J. 2011, 17, 13017 – 13031
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
13025

K.-H. Altmann et al.
the identification of necrotic cell populations displaying loss
of nuclear membrane integrity. Quadrant analysis of the an-
nexin-V-FITC- and PI-treated cells showed that at mycolac-
tone concentrations ꢁ15 ngmLꢀ1, an annexin-V and PI
double-positive (A+/PI+) subpopulation had emerged
within 24 h (Figure 2(A)). After 48 h, >90% of the cells
treated with mycolactones A/B (1a/b) at concentrations
ꢁ15 ngmLꢀ1 displayed either apoptotic (A+/PIꢀ) or necrotic
(A+/PI+) properties (Figure 2(B)). The cytopathic effect
was slightly delayed when cells were treated with
7.5 ngmLꢀ1, and absent when treated with 3.75 ngmLꢀ1 of
mycolactones A/B (1a/b), respectively (Figure 2ACTHNUTRGENUG(N B, C)). The
minimum cytotoxic concentration of synthetic mycolactone
was comparable to that of the extracted material.[14,51] In
contrast, neither analogue 4 nor its non-acylated precursor
Figure 1. Synthetic mycolactones A/B (1a/b) cause strong dose-depen-
dent morphological changes in L929 fibroblasts. Cells were incubated for
24 (A) or 48 h (B) with different concentrations of 1a/b and stained with
a fluorescent phalloidin derivative to label actin and with DAPI to label
DNA. Staining was analyzed by fluorescence microcopy using a magnifi-
cation of 960ꢃ. (DIC=differential interference contrast).
plasma membrane, a decrease in cell size, and the formation
of apoptotic bodies (Figure 1(B)). These morphological
changes were accompanied by actin cytoskeleton depolyme-
rization and DNA fragmentation, as demonstrated by the
loss of staining of the actin cytoskeleton with a fluorescent
derivative of the F-actin-binding bicyclic peptide phalloidin,
and of nuclei staining with the DNA-binding dye 4’,6-diami-
dino-2-phenylindole (DAPI), respectively (Figure 1(B)).
The effects on L929 fibroblasts were analyzed quantita-
tively by flow cytometry. Cells treated for 24, 48, or 72 h
with different concentrations of synthetic mycolactones A/B
(1a/b) were stained with the fluorescein isothiocyanate
(FITC)-labeled phospholipid-binding protein annexin-V,
which allows the detection of exposed phosphatidylserine
residues flipped across the membrane of apoptotic cells. In
addition, cells were stained with the DNA-intercalating
agent propidium iodide (PI), a dye commonly used to quan-
titatively assess cell viability. PI is membrane-impermeant
and is generally excluded from viable cells, thus allowing for
Figure 2. Synthetic mycolactones A/B (1a/b) induce necrosis and apopto-
sis in L929 fibroblasts. Cells were incubated for 24 (A), 48 (B), or 72 h
(C) with different concentrations of mycolactones A/B and stained with
annexin-V-FITC and PI. The proportions of annexin-positive (A+) and
PI-positive (PI+) cells were determined by flow cytometry. Mean values,
including standard deviations calculated for triplicate experiments, are
shown.
13026
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 13017 – 13031

Total Synthesis of Mycolactones A/B Based on RCM
FULL PAPER
MgSO4, and concentrated under reduced pressure. Flash column chroma-
tography (n-hexane/EtOAc, 1:1!EtOAc) yielded 2.68 g (78%) of the
monoprotected triol 17 as a colorless oil and 2.14 g (82%) of the Oppolz-
67 showed any measurable cytopathic or apoptosis-inducing
effects.
er auxiliary as white crystals. Rf (n-hexane/EtOAc, 1:1)=0.08; [a]D25
=
ꢀ5.48 (c=1.02, CHCl3); 1H NMR (400 MHz, CDCl3): d=7.28 (d, J=
8.6 Hz, 2H), 6.89 (d, J=8.7 Hz, 2H), 4.43 (s, 2H), 3.85–3.78 (m, 1H),
3.80 (s, 3H), 3.70–3.65 (m, 2H), 3.45 (dt, J=6.3, 1.0 Hz, 2H), 2.38 (s,
2H), 1.82–1.72 (m, 1H), 1.70–1.35 (m, 6H), 0.90 ppm (d, J=7.1 Hz, 3H);
13C NMR (100 MHz, CDCl3): d=159.1, 130.6, 129.3, 113.7, 74.4, 72.6,
70.0, 67.2, 55.3, 39.1, 33.7, 29.6, 22.9, 10.1 ppm; IR (film): n˜ =1188 (br),
2937, 2862, 1609, 1502, 1463, 1253, 822 cmꢀ1; HRMS (ESI): calcd for
C16H26NaO4 [M+Na]+: 305.1723; found: 305.1725.
Conclusion
We have established a new synthetic route towards mycolac-
tones A/B (1a/b), which is based on efficient stereoselective
construction of the macrolactone core structure of these my-
cobacterial toxins by ring-closing metathesis. The approach
has allowed the preparation of multigram quantities of key
intermediate 3 and thus provides a sound basis not only for
the preparation of mycolactones A/B (1a/b), but also of
other natural mycolactone variants and non-natural myco-
lactone analogues for SAR studies. Surprisingly, diene 65
could not be induced to undergo RCM under any of the
conditions investigated, which highlights the sensitivity even
of this broadly applicable reaction to seemingly minor
changes in substrate structure. The minimum concentration
of synthetic mycolactones A/B (1a/b) required to induce
apoptosis and cytopathicity in L929 fibroblasts was compa-
rable to that reported for mycolactones A/B extracted from
M. ulcerans. In contrast, no such effects were found for sim-
plified analogue 4.
AHCTUNGTERG(NNUN 2S,3S)-3-Hydroxy-7-[(4-methoxybenzyl)oxy]-2-methylheptyl 4-methyl-
benzenesulfonate (18): Triethylamine (1.97 mL, 14.2 mmol), p-tosyl chlo-
ride (1.81 g, 9.49 mmol), and 4-dimethylaminopyridine (DMAP) (43 mg,
0.35 mmol) were added to a solution of alcohol 17 (2.00 g, 7.08 mmol) in
CH2Cl2 (70 mL) at 08C. After 30 min, the ice bath was removed and the
solution was stirred for 16 h at RT. Celite was then added, the mixture
was filtered, and the filtrate was concentrated under reduced pressure.
Flash column chromatography (n-hexane/EtOAc, 2:1!1:1!EtOAc)
yielded 2.98 g (96%) of 18 as a colorless oil. Rf (n-hexane/EtOAc, 1:1)=
0.55; [a]2D5 =ꢀ4.78 (c=0.75, CHCl3); 1H NMR (400 MHz, CDCl3): d=
7.80 (d, J=8.3 Hz, 2H), 7.36 (d, J=8.0 Hz, 2H), 7.27 (d, J=8.6 Hz, 2H),
6.89 (d, J=8.7 Hz, 2H), 4.43 (s, 2H), 4.11–4.05 (m, 1H), 3.92–3.86 (m,
1H), 3.81 (s, 3H), 3.71–3.65 (m, 1H), 3.43 (dt, J=6.5, 0.7 Hz, 2H), 2.45
(s, 3H), 2.06 (brs, 1H), 1.92–1.81 (m, 1H), 1.67–1.28 (m, 6H), 0.85 ppm
(d, J=7.1 Hz, 3H); 13C NMR (100 MHz, CDCl3): d=159.0, 144.6, 132.8,
130.4, 129.7, 129.1, 127.7, 113.6, 72.7, 72.4, 70.2, 69.7, 55.1, 37.6, 33.9, 29.3,
22.7, 21.4, 9.4 ppm; IR (film): n˜ =1427 (br), 2944, 2862, 1609, 1509, 1359,
1172, 1093, 1033, 961, 811, 668 cmꢀ1
C23H32NaO6S [M+Na]+: 459.1812; found: 459.1820.
AHCTUNGERTG(NNUN 2S,3S)-2-[4-(4-Methoxybenzyloxy)butyl]-3-methyloxetane (19): NaH
; HRMS (ESI): calcd for
Experimental Section
(60% dispersion in mineral oil, 1.02 g, 25.6 mmol) was added to a solu-
tion of tosylate 18 (6.22 g, 14.2 mmol) in THF (80 mL) at RT. The sus-
pension was stirred for 8 h at 428C. The reaction was then quenched by
adding saturated aqueous NH4Cl solution (50 mL). EtOAc was added
(100 mL), the layers were separated, and the aqueous layer was extracted
with EtOAc (3ꢃ50 mL). The combined organic phases were washed with
brine, dried over MgSO4, and concentrated under reduced pressure.
Flash column chromatography (n-hexane/EtOAc, 4:1) yielded 3.68 g
(98%) of oxetane 19 as a colorless oil. Rf (n-hexane/EtOAc, 2:1)=0.55;
[a]2D5 = +7.58 (c=0.70, CHCl3); 1H NMR (400 MHz, CDCl3): d=7.25 (d,
J=8.6 Hz, 2H), 6.87 (d, J=8.6 Hz, 2H), 4.83–4.77 (m, 1H), 4.75 (dd, J=
7.6, 5.8 Hz, 1H), 4.42 (s, 2H), 4.05 (t, J=5.6 Hz, 1H), 3.80 (s, 3H), 3.44
(td, J=6.5, 1.0 Hz, 2H), 3.05–2.94 (m, 1H), 1.81–1.71 (m, 1H), 1.68–1.54
(m, 3H), 1.48–1.23 (m, 2H), 1.15 ppm (d, J=7.0 Hz, 3H); 13C NMR
(100 MHz, CDCl3): d=159.2, 130.1, 129.2, 113.9, 84.9, 75.7, 72.6, 69.9,
55.3, 32.0, 31.5, 29.7, 21.5, 13.5 ppm; IR (neat): n˜ =2936, 2860, 2361, 2342,
1612, 1512, 1457, 1361, 1301, 1245, 1173, 1097, 1034, 971, 848, 819, 754,
ACHTUNGTRENNUNG(2R,3S)-1-[(6R,7aR)-8,8-Dimethyl-2,2-dioxidohexahydro-1H-3a,6-metha-
nobenzo[c]isothiazol-1-yl]-3-hydroxy-7-[(4-methoxybenzyl)oxy]-2-meth-
ylheptan-1-one (16): Triflic acid (2.61 mL, 29.5 mmol) was added drop-
wise to triethylborane (1.0 m in hexane, 29.5 mL, 29.5 mmol) at RT. The
solution was stirred for 15 min at RT until gas evolution ceased and it
had become homogeneous. After cooling to ꢀ58C, a solution of 1-
[(6R,7aR)-8,8-dimethyl-2,2-dioxidohexahydro-1H-3a,6-methanobenzo[c]i-
sothiazol-1-yl]propan-1-one in CH2Cl2 (40 mL) was added by means of a
syringe, followed by a solution of DIEA (5.65 mL, 32.4 mmol) in CH2Cl2
(13 mL). After stirring for 20 min at ꢀ58C, the solution was cooled to
ꢀ788C, whereupon aldehyde 15 in CH2Cl2 (13 mL) was added by means
of a syringe. The solution was stirred at ꢀ788C for 3 h and then aqueous
phosphate buffer (pH 7.2, 60 mL) was added. The mixture was allowed
to warm to RT, the layers were separated, and the organic layer was ex-
tracted with CH2Cl2 (3ꢃ50 mL). The combined organic layers were
washed with saturated aqueous NH4Cl solution (100 mL) and brine
(100 mL), dried over MgSO4, and concentrated under reduced pressure.
Flash column chromatography (n-hexane/EtOAc, 4:1!1:1) yielded 6.07 g
(83%) of the aldol product 16 as a yellowish oil. [a]2D5 =ꢀ67.38 (c=0.61,
CHCl3); 1H NMR (400 MHz, CDCl3): d=7.28 (d, J=8.5 Hz, 2H), 6.89
(d, J=8.5 Hz, 2H), 4.43 (s, 2H), 3.99–3.92 (m, 1H), 3.90 (t, J=6.3 Hz,
1H), 3.80 (s, 3H), 3.48–3.40 (m, 3H), 3.20–3.10 (brs, 1H), 3.08–3.00 (m,
1H), 2.09–2.02 (m, 2H), 1.95–1.85 (m, 3H), 1.68–1.50 (m, 5H), 1.48–1.30
(m, 4H), 1.30–1.25 (m, 3H), 1.18 (s, 3H), 0.98 ppm (s, 3H); 13C NMR
(100 MHz, CDCl3): d=176.6, 158.7, 130.5, 128.6, 113.4, 72.1, 70.3, 69.6,
64.6, 54.9, 52.7, 48.0, 47.4, 44.3, 43.9, 38.0, 33.5, 32.4, 29.2, 26.1, 22.3, 20.5,
19.5, 11.5 ppm; IR (film): n˜ =3523 (br), 2937, 2858, 1861, 1509, 1330,
1247, 1133, 1037 cmꢀ1; HRMS (ESI): calcd for C26H39NNaO6S [M+Na]+:
516.2390; found: 516.2409.
577 cmꢀ1
found: 287.1620.
(4S,5S)-9-[(4-Methoxybenzyl)oxy]-2,4-dimethylnon-1-en-5-ol (20): 2-Bro-
;
HRMS (ESI): calcd for C16H24NaO3 [M+Na]+: 287.1623;
AHCTUNGTRENNUNG
mopropene (3.03 mL, 34.0 mmol) in Et2O (10 mL) was slowly added to a
solution of tert-butyllithium (1.6 m in pentane, 21.3 mL, 34.0 mmol) in
Et2O (40 mL) at ꢀ788C. After stirring for 10 min at ꢀ788C, oxetane 19
(1.00 g, 3.78 mmol) in Et2O (10 mL) was added, followed by BF3·Et2O
(0.47 mL, 3.78 mmol). The clear brownish solution was stirred for 10 min
at ꢀ788C and then quenched by the addition of saturated aqueous
NH4Cl solution (20 mL). After warming to RT, Et2O (100 mL) and water
were added, the layers were separated, and the aqueous layer was ex-
tracted with Et2O (2ꢃ80 mL). The combined organic layers were succes-
sively washed with 2n aqueous NaOH (30 mL) and saturated aqueous
NaHCO3 (30 mL), dried over MgSO4, and the solvent was evaporated
under reduced pressure. Flash column chromatography (n-hexane/
EtOAc, 8:1!4:1) yielded 1.042 g (90%) of alcohol 20 as a colorless oil.
[a]2D5 =ꢀ3.48 (c=0.38, CHCl3); 1H NMR (400 MHz, CDCl3): d=7.27 (d,
J=8.6 Hz, 2H), 6.89 (d, J=8.6 Hz, 2H), 4.78 (s, 1H), 4.70 (s, 1H), 4.43
(s, 2H), 3.80 (s, 3H), 3.56–3.48 (m, 1H), 3.45 (dt, J=6.5, 0.8 Hz, 2H),
ACHTUNGTRENNUNG
a
a
30.3 mmol) in Et2O (30 mL) at 08C. After stirring at 08C for 4 h, saturat-
ed aqueous NH4Cl solution (100 mL) was added. When gas evolution
had ceased, the organic layer was decanted off, filtered through a plug of
Chem. Eur. J. 2011, 17, 13017 – 13031
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
13027

K.-H. Altmann et al.
2.19–2.11 (m, 1H), 1.93–1.85 (m, 1H), 1.79–1.30 (m, 8H), 1.71 (s, 3H),
0.83 ppm (d, J=6.8 Hz, 3H); 13C NMR (100 MHz, CDCl3): d=159.1,
144.4, 130.6, 129.2, 113.7, 111.8, 74.3, 72.5, 70.0, 55.2, 42.0, 35.7, 34.2, 29.7,
23.0, 22.1, 13.1 ppm; IR (film): n˜ =3430 (br), 2929, 2861, 1609, 1512, 1248,
sired acid 12 as a yellow oil. [a]2D5 =ꢀ7.08 (c=0.33, CHCl3); 1H NMR
(400 MHz, CDCl3): d=10.98 (brs, 1H), 4.74 (s, 1H), 4.66 (s, 1H), 3.61–
3.54 (m, 1H), 2.46 (t, J=7.4 Hz, 2H), 2.20 (d, J=10.1 Hz, 1H), 1.81–1.70
(m, 2H), 1.68 (s, 3H), 1.66–1.37 (m, 4H), 0.96 (t, J=8.0 Hz, 9H), 0.79 (d,
J=6.3 Hz, 3H), 0.60 ppm (q, J=7.8 Hz, 6H); 13C NMR (100 MHz,
CDCl3): d=180.1, 144.6, 111.4, 75.7, 40.5, 35.7, 34.2, 32.8, 22.2, 21.3, 14.0,
7.0, 5.2 ppm. IR (film): n˜ =2955, 2909, 2876, 1710, 1456, 1240, 1011, 886,
1094, 1037, 890, 822 cmꢀ1
306.2189: found: 306.2191.
;
HRMS (ESI): calcd for C19H30O3 [M]+:
TriethylACHTUNGTRENNUNG[(4S,5S)-9-(4-methoxybenzyloxy)-2,4-dimethylnon-1-en-5-yloxy]-
740 cmꢀ1
;
HRMS (EI): calcd for C17H34O3Si [MꢀC2H5]+: 285.1880;
silane (21): TESOTf (1.24 mL, 5.42 mmol) in CH2Cl2 (5 mL) was added
by means of a syringe to a solution of alcohol 20 (1.384 g, 4.52 mmol)
and 2,6-lutidine (1.32 mL, 11.3 mmol) in CH2Cl2 (25 mL) at ꢀ788C. The
mixture was stirred for 60 min at ꢀ788C and then the reaction was
quenched by the addition of MeOH (2.0 mL). The solution was allowed
to warm to RT, silica gel was added, and the solvent was evaporated
under reduced pressure. Purification by flash column chromatography (n-
hexane/EtOAc, 30:1) yielded 1.863 g (98%) of the protected diol 21 as a
found: 285.1881.
AHCTUNGTRENGN(UN 5S,6S)-[(2S,3R)-2-Methyl-1-(tosyloxy)hex-5-en-3-yl]-6,8-dimethyl-5-(trie-
thylsilyloxy)non-8-enoate (24): DCC (430 mg, 2.21 mmol) in CH2Cl2
(2.5 mL) was added to a solution of alcohol 5 (525 mg, 1.85 mmol), acid
12 (500 mg, 1.60 mmol), and DMAP (60 mg, 0.50 mmol) in CH2Cl2
(10 mL) at 08C. After 30 min, the solution was allowed to warm to RT
and then stirred for 15 h. The solvent was then evaporated under reduced
pressure and the residue was purified by flash column chromatography
(n-hexane/EtOAc, 12:1) to yield 757 mg (82%) of ester 24 as a colorless
1
colorless oil. [a]2D5 =ꢀ7.98 (c=1.16, CHCl3); H NMR (400 MHz, CDCl3):
d=7.27 (d, J=8.6 Hz, 2H), 6.89 (d, J=8.6 Hz, 2H), 4.75 (s, 1H), 4.68 (s,
1H), 4.42 (s, 2H), 3.80 (s, 3H), 3.59–3.51 (m, 1H), 3.44 (t, J=6.6 Hz,
2H), 2.20 (d, J=10.3 Hz, 1H), 1.82–1.70 (m, 2H), 1.68 (s, 3H), 1.66–1.53
(m, 2H), 1.48–1.25 (m, 4H), 0.96 (t, J=7.9 Hz, 9H), 0.79 (d, J=6.4 Hz,
3H), 0.61 ppm (q, J=7.8 Hz, 6H); 13C NMR (100 MHz, CDCl3): d=
159.1, 144.8, 130.8, 129.2, 113.7, 111.2, 75.9, 72.5, 70.1, 55.3, 40.8, 35.6,
33.4, 30.0, 22.7, 22.3, 13.9, 7.0, 5.3 ppm; IR (film): n˜ =2955, 2940, 2876,
1609, 1509, 1459, 1244, 1094, 1037, 732 cmꢀ1; HRMS (ESI): calcd for
C25H48NO4Si [M+NH4]+: 438.3398; found: 438.3396.
1
oil. [a]2D5 =ꢀ23.78 (c=0.28, CHCl3); H NMR (400 MHz, CDCl3): d=7.79
(d, J=8.3 Hz, 2H), 7.35 (d, J=8.0 Hz, 2H), 5.72–5.60 (m, 1H), 5.09–5.00
(m, 2H), 4.90–4.83 (m, 1H), 4.74 (s, 1H), 4.68 (s, 1H), 4.08–4.02 (m, 1H),
3.92–3.85 (m, 1H), 3.60–3.53 (m, 1H), 2.46 (s, 3H), 2.39–2.31 (m, 1H),
2.29–2.18 (m, 4H), 2.14–2.03 (m, 1H), 1.80–1.60 (m, 3H), 1.69 (s, 3H),
1.57–1.28 (m, 3H), 0.99–0.92 (m, 12H), 0.79 (d, J=6.2 Hz, 3H), 0.61 ppm
(q, J=8.0 Hz, 6H); 13C NMR (100 MHz, CDCl3): d=172.7, 144.7, 144.6,
133.2, 132.9, 129.8, 127.9, 118.2, 111.3, 75.6, 73.2, 71.2, 40.4, 35.9 (2C),
35.8, 34.5, 32.7, 22.2, 21.6 (2C), 14.1, 13.4, 7.0, 5.2 ppm; IR (film): n˜ =
2969, 2933, 1735, 1362, 1180, 969, 815, 668 cmꢀ1; HRMS (ESI): calcd for
C31H56NO6SSi [M+NH4]+: 598.3592; found: 598.3590.
ACHTUNGTRENNUNG(5S,6S)-6,8-Dimethyl-5-(triethylsilyloxy)non-8-en-1-ol (22): DDQ (1.62 g,
7.13 mmol) was added to a solution of 21 (1.20 g, 2.85 mmol) in a mixture
of CH2Cl2 (40 mL) and aqueous phosphate buffer (pH 7.2, 4.0 mL) at RT.
The mixture was stirred at RT for 45 min and then the reaction was
quenched by the addition of saturated aqueous NaHCO3 solution
(20 mL) and CH2Cl2 (40 mL). The layers were separated, and the organic
layer was washed with saturated aqueous NaHCO3 (2ꢃ20 mL) and dried
over MgSO4. Evaporation of the solvent under reduced pressure and pu-
rification of the residue by flash column chromatography (fine silica gel;
n-hexane/CH2Cl2/MeOH, 20:2:1!20:3:1) yielded 785 mg (92%) of alco-
Toluene-4-sulfonic acid (S)-2-[(E)-(2R,7S,8S)-5,7-dimethyl-12-oxo-8-trie-
thylsilanyloxy-oxacyclododec-4-en-2-yl]propyl ester (25): Grubbs II cata-
lyst (50 mg, 0.059 mmol) in CH2Cl2 (1.5 mL) was added to a refluxing so-
lution of diene 24 (430 mg, 0.740 mmol) in CH2Cl2 (250 mL) and reflux-
ing of the brown solution was continued for 6.5 h, with additional por-
tions of Grubbs II catalyst (each of 13 mg, 0.015 mmol) being added after
2 h and 4 h, respectively. At this point (6.5 h), MS and TLC indicated
complete conversion of the starting diene and the reaction was quenched
by the addition of water (50 mL) and brine (50 mL). The layers were sep-
arated, the aqueous layer was extracted with CH2Cl2 (2ꢃ50 mL), and the
combined organic layers were dried over MgSO4 and concentrated under
reduced pressure. Silica gel and n-hexane were added when about half of
the CH2Cl2 had been evaporated. The residue was purified twice by flash
column chromatography (n-hexane/EtOAc, 20:1!12:1; n-hexane/EtOAc,
hol 22 as
a
yellow oil. [a]2D5 =ꢀ17.98 (c=1.58, CHCl3); 1H NMR
(400 MHz, CDCl3): d=4.74 (s, 1H), 4.67 (s, 1H), 3.65 (t, J=6.5 Hz, 2H),
3.60–3.53 (m, 1H), 2.20 (d, J=10.1 Hz, 1H), 1.82–1.71 (m, 2H), 1.69 (s,
3H), 1.63–1.50 (m, 2H), 1.49–1.25 (m, 5H), 0.96 (t, J=7.9 Hz, 9H), 0.78
(d, J=6.5 Hz, 3H), 0.60 ppm (q, J=7.8 Hz, 6H); 13C NMR (100 MHz,
CDCl3): d=144.4, 111.2, 75.8, 62.4, 40.6, 35.5, 33.1, 32.7, 22.1, 22.1, 13.8,
6.8, 5.4 ppm; IR (film): n˜ =3334 (br), 2934, 2869, 1648, 1456, 1373, 1054,
883 cmꢀ1; HRMS (ESI): calcd for C17H37O2Si [M+H]+: 301.2557; found:
301.2552.
16:1!9:1) to yield 329 mg (80%) of lactone 25 as a brown oil. [a]D25
=
ꢀ44.08 (c=0.61, CHCl3); 1H NMR (400 MHz, CDCl3): d=7.79 (d, J=
8.3 Hz, 2H), 7.35 (d, J=8.0 Hz, 2H), 4.97–4.85 (m, 2H), 4.08–4.02 (m,
1H), 3.86–3.80 (m, 1H), 3.37–3.42 (m, 1H), 2.46 (s, 3H), 2.43–2.35 (m,
2H), 2.11–2.00 (m, 2H), 1.94–1.58 (m, 6H), 1.65 (s, 3H), 1.45–1.30 (m,
2H), 0.99–0.90 (m, 15H), 0.60 ppm (q, J=8.0 Hz, 6H); 13C NMR
(100 MHz, CDCl3): d=173.2, 144.8, 138.1, 132.9, 129.8, 128.0, 120.4, 77.6,
72.5, 71.6, 45.3, 37.6, 35.6, 33.5, 33.3, 31.4, 21.7, 21.6, 18.7, 15.7, 13.4, 7.0,
5.1 ppm; IR (film): n˜ =2954, 2912, 2876, 1731, 1366, 1244, 1176, 1022,
969, 815, 672 cmꢀ1; HRMS (ESI): calcd for C29H48NaO6SiS [M+Na]+:
575.2833; found: 575.2827.
ACHTUNGTRENNUNG(5S,6S)-6,8-Dimethyl-5-(triethylsilyloxy)non-8-enoic acid (12): Dess–
Martin periodinane (15% in CH2Cl2, 44.2 mL, 20.8 mmol) was added
dropwise from a syringe to a solution of alcohol 22 (2.088 g, 6.95 mmol)
in CH2Cl2 (200 mL) at 08C. The solution was allowed to warm to RT and
stirred for 41= h. The reaction was then quenched by the addition of satu-
2
rated aqueous NaS2SO3 (200 mL), saturated aqueous NaHCO3 (200 mL),
and water (100 mL). The layers were separated and the aqueous layer
was extracted with CH2Cl2 (2ꢃ100 mL). The combined organic phases
were washed with brine and dried over MgSO4. Evaporation of the sol-
vent under reduced pressure yielded the crude aldehyde, which was used
immediately for the next reaction.
(6S,7S,12R,E)-12-[(R)-1-Iodopropan-2-yl]-7,9-dimethyl-6-(triethylsilylox-
y)oxacyclododec-9-en-2-one (3): Tosylate 25 (114 mg, 0.21 mmol) and
NaI (278 mg, 1.85 mmol) were dissolved in dry acetone (7.0 mL) and the
mixture was stirred at RT for 5 min. The clear yellowish solution was
then refluxed at 608C for 9 h. Silica gel was added and the solvent was
evaporated under reduced pressure. Flash column chromatography of the
residue (n-hexane/EtOAc, 50:1) yielded 99 mg (95%) of iodide 3 as a
slightly yellow oil. The alkyl iodide was used immediately for the next re-
action. Rf (n-hexane/EtOAc, 4:1)=0.91; 1H NMR (400 MHz, CDCl3):
d=4.98 (dd, J=9.5, 1.8 Hz, 1H), 4.90 (ddd, J=11.6, 7.2, 3.2 Hz, 1H),
3.43–3.33 (m, 1H), 3.29 (dd, J=9.4, 4.0 Hz, 1H), 2.97 (dd, J=9.8, 8.7 Hz,
1H), 2.55–2.46 (m, 1H), 2.40 (td, J=14.0, 11.4 Hz, 1H), 2.11 (d, J=
13.2 Hz, 1H), 2.06–1.56 (m, 7H), 1.65 (s, 3H), 1.50–1.31 (m, 2H), 1.09 (d,
J=6.8 Hz, 3H), 0.98 (d, J=6.8 Hz, 3H), 0.95 (t, J=7.8 Hz, 9H),
0.59 ppm (q, J=7.8 Hz, 6H); 13C NMR (100 MHz, CDCl3): d=173.3,
2-Methyl-2-butene (37 mL, 347 mmol), sodium dihydrogen phosphate
monohydrate (NaH2PO4·H2O, 2.40 g, 17.4 mmol), and sodium chlorite
(NaClO2, 3.14 g, 34.7 mmol) were added to a solution of the above crude
aldehyde (2.074 g, 6.95 mmol) in THF (240 mL), H2O (80 mL), and tert-
butanol (360 mL) at RT. In this process, the solution first turned yellow
and then became colorless once more. After stirring at RT for 20 min,
volatile solvents were evaporated under reduced pressure. EtOAc
(50 mL) and a small quantity of additional water were added to the re-
maining aqueous mixture and the layers were separated. The aqueous
layer was extracted with EtOAc (3ꢃ30 mL). The combined organic
layers were dried over MgSO4 and concentrated under reduced pressure.
Flash column chromatography of the residue (n-hexane/EtOAc, 9:1, con-
taining 0.1% HCOOH) yielded 1.989 g (91% over two steps) of the de-
13028
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 13017 – 13031

Total Synthesis of Mycolactones A/B Based on RCM
FULL PAPER
138.0, 120.5, 77.6, 74.3, 45.3, 40.5, 35.7, 33.5, 33.3, 31.3, 21.8, 18.7, 17.1,
15.7, 10.5, 7.0 (3C), 5.4 ppm (3C); IR (film): n˜ =2953, 2913, 2875, 1731,
1457, 1436, 1380, 1326, 1309, 1245, 1199, 1163, 1126, 1082, 1018, 985, 818,
761, 743, 725 cmꢀ1; HRMS (EI): calcd for C22H41IO3Si [M]+: 508.1865;
found: 508.1868.
(2.0 mL) at RT. Immediately, the bright-yellow solution became turbid.
Alcohol 60 (9.7 mg, 17 mmol) in THF (0.74 mL) was added and the sus-
pension was stirred for 2 h at RT, protected from light. The reaction was
then quenched by the addition of saturated aqueous NaHCO3 solution
(5 mL) and EtOAc (20 mL). The layers were separated and the aqueous
layer was extracted with EtOAc (2ꢃ10 mL). The combined organic
phases were washed with brine (5 mL), dried over MgSO4, and concen-
trated under reduced pressure. Purification by flash column chromatogra-
phy (n-hexane/EtOAc, 100:1!33:1) yielded 18.7 mg (89%) of the ester
in the form of a yellow oil as a mixture of isomers that was not separated
(6S,7S,12R,E)-12-{(2S,6R,E)-6-[(4R,6R)-2,2-Di-tert-butyl-6-methyl-1,3,2-
dioxasilinan-4-yl]-4-methylhept-4-en-2-yl}-7,9-dimethyl-6-(triethylsilylox-
y)oxacyclododec-9-en-2-one (59): tert-Butyllithium (1.6 m in pentane,
0.22 mL, 0.35 mmol) in THF (1.0 mL) was slowly added by means of a
syringe to a solution of alkyl iodide 3 (80 mg, 0.16 mmol) and 9-MeO-
BBN (1.0 m in hexane, 0.38 mL, 0.38 mmol) in Et2O (1.0 mL) in a 10 mL
Schlenk flask at ꢀ788C. The resulting mixture was stirred for 10 min at
ꢀ788C and then allowed to warm to RT, whereupon it changed from a
milky, slightly yellow suspension into a colorless solution, which was
stirred at RT for an additional 45 min. This solution was then transferred
(E-D4’,5’/Z-D4’,5’
, =
1:1). Rf (n-hexane/EtOAc, 20:1)=0.34, 0.40; [a]D25
ꢀ32.78 (c=1.00, CHCl3). Z-D4’,5’ isomer: 1H NMR (500 MHz,
[D6]acetone): d=7.93 (d, J=15.7 Hz, 1H), 6.67 (dd, J=15.1, 11.1 Hz,
1H), 6.43 (d, J=15.1 Hz, 1H), 6.34 (s, 1H), 6.19 (d, J=11.1 Hz, 1H),
5.94 (d, J=15.7 Hz, 1H), 5.69–5.65 (m, 1H), 5.09 (d, J=11.1 Hz, 1H),
5.06 (d, J=10.0 Hz, 1H), 4.92–4.85 (m, 1H), 4.74–4.68 (m, 1H), 4.58 (dd,
J=9.2, 3.2 Hz, 1H), 4.25–4.17 (m, 1H), 4.05–3.97 (m, 1H), 3.85–3.75 (m,
2H), 2.55–2.36 (m, 3H), 2.16–1.78 (m, 10H), 2.01 (s, 3H), 1.98 (s, 3H),
1.73–1.54 (m, 6H), 1.71 (d, J=3.2 Hz, 3H), 1.66 (s, 3H), 1.47–1.31 (m,
3H), 1.20 (d, J=6.1 Hz, 3H), 1.17 (d, J=6.0 Hz), 1.04 (d, J=7.0 Hz,
3H), 1.03 (s, 9H), 1.00 (s, 9H), 0.93–0.88 (m, 33H), 0.12–0.07 (m, 15H),
0.03 ppm (s, 3H); 13C NMR (100 MHz, [D6]acetone): d=173.2, 166.9,
143.1, 141.8, 140.0, 137.3, 135.7, 135.0 (2C), 134.7, 134.0, 132.0, 130.4,
125.3, 123.9, 119.7, 79.2, 78.9, 76.2, 74.2, 72.1, 71.5, 66.8, 46.4, 44.4, 44.0,
42.6, 41.3, 36.0, 35.3, 32.9, 31.5, 29.7 (obscured by [D6]acetone signal),
28.0 (3C), 27.7 (3C), 26.4–26.2 (m, 9C), 25.3, 24.3, 23.3, 21.1, 20.8, 20.5,
20.1, 18.7, 18.6 (2C), 17.7, 17.1, 16.3, 16.0, 15.0, 13.9, ꢀ3.8, ꢀ3.9, ꢀ4.1,
ꢀ4.3, ꢀ4.4 ppm. E-D4’,5’ isomer: 1H NMR (500 MHz, [D6]acetone): d=
7.37 (d, J=15.6 Hz, 1H), 6.66 (dd, J=15.1, 11.1 Hz, 1H), 6.48 (d, J=
15.1 Hz, 1H), 6.47 (s, 1H), 6.39 (d, J=11.1 Hz, 1H), 5.90 (d, J=15.6 Hz,
1H), 5.69–5.65 (m, 1H), 5.09 (d, J=11.1 Hz, 1H), 5.06 (d, J=10.0 Hz,
1H), 4.92–4.85 (m, 1H), 4.74–4.68 (m, 1H), 4.58 (dd, J=9.2, 3.2 Hz, 1H),
4.25–4.17 (m, 1H), 4.05–3.97 (m, 1H), 3.85–3.75 (m, 2H), 2.55–2.36 (m,
3H), 2.16–1.78 (m, 10H), 2.09 (s, 3H), 2.06 (s, 3H), 1.73–1.54 (m, 6H),
1.71 (d, J=3.2 Hz, 3H), 1.66 (s, 3H), 1.47–1.31 (m, 3H), 1.20 (d, J=
6.1 Hz, 3H), 1.17 (d, J=6.0 Hz), 1.04 (d, J=7.0 Hz, 3H), 1.03 (s, 9H),
1.00 (s, 9H), 0.93–0.88 (m, 33H), 0.12–0.07 (m, 15H), 0.04 ppm (s, 3H);
13C NMR (100 MHz, [D6]acetone): d=173.2, 166.9, 151.2, 144.3, 140.3,
137.3, 136.2, 135.7, 135.4, 135.0, 134.0, 133.2, 130.4, 125.3, 123.9, 117.4,
79.2, 78.9, 76.2, 74.2, 72.1, 71.5, 66.8, 46.4, 44.4, 44.0, 42.6, 41.3, 36.0, 35.3,
32.9, 31.5, 29.7 (obscured by [D6]acetone signal), 28.0 (3C), 27.7 (3C),
26.4–26.2 (m, 9C), 25.3, 24.3, 23.3, 20.8, 20.5, 20.1, 18.7, 18.6 (2C), 17.2,
17.1, 16.3, 16.0, 15.0, 14.3, 13.9, ꢀ3.8, ꢀ3.9, ꢀ4.1, ꢀ4.3, ꢀ4.4 ppm; IR
(film): n˜ =2956, 2929, 2894, 2857, 1729, 1711, 1472, 1462, 1377, 1362,
into a solution of vinyl iodide 57 (73 mg, 0.17 mmol), [PdACTHUNRGTNEUNG(dppf)Cl2]
(11.5 mg, 16 mmol), AsPh3 (14.5 mg, 47 mmol), and aqueous Cs2CO3 solu-
tion (3.0 m in H2O, 178 mL, 0.53 mmol) in DMF (1.0 mL) by means of a
syringe. The yellow solution was stirred in the dark for 21 h at RT. H2O
(20 mL) and Et2O (20 mL) were then added, the layers were separated,
and the aqueous layer was extracted with Et2O (2ꢃ20 mL). The com-
bined organic extracts were washed with brine (5 mL), dried over
MgSO4, and the solvent was evaporated under reduced pressure. Purifica-
tion of the residue by flash column chromatography (n-hexane/EtOAc,
100:1) yielded 86 mg (80%) of the coupling product 59 as a slightly
yellow oil. Rf (n-hexane/EtOAc, 9:1)=0.84. [a]2D5 =ꢀ27.58 (c=1.45,
CHCl3); 1H NMR (400 MHz, CDCl3): d=4.99–4.92 (m, 2H), 4.86–4.78
(m, 1H), 4.18–4.09 (m, 1H), 3.68 (ddd, J=11.4, 7.7, 1.9 Hz, 1H), 3.41–
3.35 (m, 1H), 2.54–2.25 (m, 3H), 2.09 (dd, J=13.2, 5.3 Hz, 1H), 2.04–
1.58 (m, 18H), 1.66 (s, 3H), 1.61 (d, J=1.2 Hz, 3H), 1.45–1.25 (m, 3H),
1.18 (d, J=6.1 Hz, 3H), 1.03–0.92 (m, 34H), 1.01 (s, 9H), 0.98 (s, 9H),
0.95 (t, J=7.9 Hz, 9H), 0.60 ppm (q, J=7.9 Hz, 6H); 13C NMR
(100 MHz, CDCl3): d=173.5, 137.4, 132.8, 129.8, 121.5, 77.9, 77.8, 75.5,
70.5, 45.5, 43.5, 41.7, 40.5, 36.0, 34.8, 33.5, 33.5, 29.6, 27.6 (3C), 27.2 (3C),
24.9, 22.7, 21.7, 19.6, 18.7, 16.7, 16.1, 15.6, 14.5, 7.0 (3C), 5.1 ppm (3C);
IR (film): n˜ =2959, 2933, 2876, 2858, 2360, 2341, 1730, 1473, 1458, 1375,
1248, 1164, 1134, 1108, 1069, 1024, 983, 912, 886, 826, 792, 761, 742, 668,
+
651 cmꢀ1; HRMS (ESI): calcd for C39H74NaO5Si2 [M+Na]+: 701.4972;
found: 701.4955.
(6S,7S,12R,E)-12-{(2S,6R,E)-6-[(4R,6R)-2,2-Di-tert-butyl-6-methyl-1,3,2-
dioxasilinan-4-yl]-4-methylhept-4-en-2-yl}-6-hydroxy-7,9-dimethyloxacy-
clododec-9-en-2-one (60): H2O (0.5 mL) and HOAc (0.5 mL) were added
to a solution of 59 (19.7 mg, 29 mmol) in THF (1.0 mL) at RT and the re-
action mixture was stirred for 21= h. Saturated aqueous NaHCO3 solution
1251, 1156, 1132, 1096, 1068, 1038, 1005, 981, 834, 807, 773, 759, 651 cmꢀ1
;
2
+
HRMS (ESI): calcd for C70H128NaO9Si4 [M+Na]+: 1247.8528; found:
was then added until no more gas was evolved. EtOAc was added, the
layers were separated, and the aqueous layer was extracted with EtOAc
(2ꢃ10 mL). The combined organic phases were washed with brine, dried
over MgSO4, and concentrated under reduced pressure. Purification of
the residue by flash column chromatography (n-hexane/EtOAc, 9:1)
yielded 14.8 mg (90%) of 60 as a colorless oil. Rf (n-hexane/EtOAc,
1247.8584.
(2E,8E,10E,12S,13S,15S)-{(6S,7S,12R,E)-12-[(2S,6R,7R,9R,E)-7,9-Dihy-
droxy-4,6-dimethyldec-4-en-2-yl]-7,9-dimethyl-2-oxooxacyclododec-9-en-
6-yl}-12,13,15-tris(tert-butyldimethylsilyloxy)-4,6,10-trimethylhexadeca-
2,4,6,8,10-pentaenoate (62): HF·pyridine (ca. 70% HF, 50 mL) was added
to a solution of 61 (23.8 mg, 19.4 mmol) in THF (2.0 mL) and pyridine
(450 mL) in a Falcon tube (10 mL) and the yellow solution was stirred at
RT for 1 h. The reaction was then quenched by the addition of saturated
aqueous NaHCO3 until gas evolution ceased. EtOAc was added (20 mL),
the layers were separated, and the aqueous layer was extracted with
EtOAc (2ꢃ20 mL). The combined organic phases were dried over
MgSO4 and concentrated under reduced pressure. Purification of the resi-
due by flash column chromatography (n-hexane/EtOAc, 9:1!2:1, 0.2%
MeOH) yielded 17.8 mg (84%) of 62 as a mixture of isomers (trans/cis
5:4) as a yellow oil. Rf (n-hexane/EtOAc, 2:1)=0.28; [a]2D5 =ꢀ27.68 (c=
0.71, MeOH). Z-D4’,5’ isomer: 1H NMR (500 MHz, [D6]acetone): d=7.93
(d, J=15.6 Hz, 1H), 6.70–6.62 (m, 1H), 6.43 (d, J=15.2 Hz, 1H), 6.34 (s,
1H), 6.19 (d, J=11.1 Hz, 1H), 5.93 (d, J=15.6 Hz, 1H), 5.69–5.64 (m,
1H), 5.12 (d, J=10.4 Hz, 1H), 5.03 (d, J=9.7 Hz, 1H), 4.93–4.85 (m,
1H), 4.75–4.65 (m, 1H), 4.58 (dd, J=9.2, 3.2 Hz, 1H), 4.27–4.20 (m, 2H),
4.04–3.92 (m, 2H), 3.81–3.75 (m, 1H), 3.54–3.47 (m, 1H), 2.55–2.45 (m,
1H), 2.45–2.34 (m, 2H), 2.16–1.79 (m, 8H), 2.04 (s, 3H), 1.98 (s, 3H),
1.95 (s, 3H), 1.78–1.51 (m, 7H), 1.71 (s, 3H), 1.64 (s, 3H), 1.43–1.26 (m,
1
2:1)=0.53; [a]2D5 =ꢀ36.28 (c=0.47, CHCl3); H NMR (400 MHz, CDCl3):
d=5.05–4.93 (m, 3H), 4.18–4.08 (m, 1H), 3.69 (ddd, J=11.3, 7.5, 1.8 Hz,
1H), 3.54–3.47 (m, 1H), 2.46 (dt, J=14.2, 11.5 Hz, 1H), 3.48–2.20 (m,
3H), 2.14–2.05 (m, 2H), 2.02–1.83 (m, 4H), 1.79–1.57 (m, 5H), 1.62 (s,
3H), 1.60 (d, J=1.2 Hz, 3H), 1.38–1.27 (m, 3H), 1.18 (d, J=6.1 Hz, 3H),
1.04–0.92 (m, 6H), 1.00 (s, 9H), 0.98 (s, 9H), 0.85 ppm (d, J=6.6 Hz,
3H); 13C NMR (100 MHz, CDCl3): d=175.2, 137.0, 132.6, 129.9, 123.1,
77.8, 77.2, 71.0, 70.5, 45.4, 43.3, 41.7, 40.4, 35.0, 34.9, 34.8, 34.1, 30.0, 27.6,
27.2, 24.9, 22.7, 20.3, 19.6, 16.7, 16.5, 16.1, 14.7 ppm; IR (neat): n˜ =3672,
2967, 2931, 2905, 2361, 2339, 1713, 1470, 1454, 1379, 1251, 1161, 1136,
1101, 1053, 981, 909, 885, 824, 790, 734, 649, 441 cmꢀ1; HRMS (ESI):
calcd for C33H60NaO5Si+ [M+Na]+: 587.4108; found: 587.4107.
(2E,8E,10E,12S,13S,15S)-((6S,7S,12R,E)-12-{(2S,6R,E)-6-[(4R,6R)-2,2-
Di-tert-butyl-6-methyl-1,3,2-dioxasilinan-4-yl]-4-methylhept-4-en-2-yl}-
7,9-dimethyl-2-oxooxacyclododec-9-en-6-yl)-12,13,15-tris(tert-butyldime-
thylsilyloxy)-4,6,10-trimethylhexadeca-2,4,6,8,10-pentaenoate (61): DIEA
(24 mL, 138 mmol), Cl3C6H2COCl (11 mL, 244 mmol), and DMAP (21 mg,
172 mmol) were added to a solution of acid 51 (23.4 mg, 35 mmol) in THF
Chem. Eur. J. 2011, 17, 13017 – 13031
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
13029

K.-H. Altmann et al.
1H), 1.17 (d, J=6.0 Hz, 3H), 1.14 (d, J=6.1 Hz, 3H), 0.98 (d, J=6.7 Hz,
3H), 0.92–0.89 (m, 33H), 0.11–0.07 (m, 15H), 0.05–0.01 ppm (m, 3H);
13C NMR (125 MHz, [D6]acetone): d=173.3, 166.9, 143.1, 141.9, 140.0,
137.3, 135.7, 135.1, 135.1, 134.7, 133.5, 132.0, 131.2, 125.3, 123.9, 119.7,
79.3, 77.0, 76.3, 74.4, 72.1, 69.0, 66.8, 46.4, 44.4, 44.0, 43.8, 40.5, 36.0, 35.4,
32.9, 31.5, 29.7, 26.4, 26.3, 26.3, 24.7, 24.3, 21.1, 20.8, 20.6, 18.7, 18.6, 17.7,
17.4, 16.2, 15.9, 15.0, 13.9, ꢀ3.7, ꢀ3.9, ꢀ3.9, ꢀ4.1, ꢀ4.2, ꢀ4.3 ppm. E-D4’,5’
isomer: 1H NMR (500 MHz, [D6]acetone): d=7.37 (d, J=15.5 Hz, 1H),
6.70–6.62 (m, 1H), 6.48 (d, J=14.4 Hz, 1H), 6.47 (s, 1H), 6.39 (d, J=
11.5 Hz, 1H), 5.89 (d, J=15.5 Hz, 1H), 5.69–5.64 (m, 1H), 5.12 (d, J=
10.4 Hz, 1H), 5.03 (d, J=9.7 Hz, 1H), 4.93–4.85 (m, 1H), 4.75–4.65 (m,
1H), 4.58 (dd, J=9.2, 3.2 Hz, 1H), 4.27–4.20 (m, 2H), 4.04–3.92 (m, 2H),
3.81–3.75 (m, 1H), 3.54–3.47 (m, 1H), 2.55–2.45 (m, 1H), 2.45–2.34 (m,
2H), 2.16–1.79 (m, 8H), 2.09 (s, 3H), 2.06 (s, 3H), 1.95 (s, 3H), 1.78–1.51
(m, 7H), 1.71 (s, 3H), 1.64 (s, 3H), 1.43–1.26 (m, 1H), 1.17 (d, J=6.0 Hz,
3H), 1.14 (d, J=6.1 Hz, 3H), 0.98 (d, J=6.7 Hz, 3H), 0.92–0.89 (m,
33H), 0.11–0.07 (m, 15H), 0.05–0.01 ppm (m, 3H); 13C NMR (125 MHz,
[D6]acetone): d=173.3, 166.9, 151.2, 144.3, 140.3, 137.3, 136.2, 135.7,
135.4, 135.0, 133.5, 133.2, 131.2, 125.2, 123.9, 117.4, 79.3, 77.0, 76.3, 74.4,
72.1, 69.0, 66.8, 46.4, 44.4, 44.0, 43.8, 40.5, 36.0, 35.4, 32.9, 31.5, 29.7, 26.4,
26.3, 26.3, 24.7, 24.3, 20.8, 20.6, 18.7, 18.6, 17.4, 17.4, 16.2, 15.9, 15.0, 14.4,
13.9, ꢀ3.7, ꢀ3.9, ꢀ3.9, ꢀ4.1, ꢀ4.2, ꢀ4.3 ppm; IR (film, acetone): n˜ =2956,
2929, 2857, 2359, 1730, 1710, 1460, 1385, 1252, 1157, 1091, 1070, 836,
For microscopy analysis, cells were seeded in four-chamber glass slides
(BD Falcon) and left to adhere overnight. The medium was then aspirat-
ed and replaced by medium containing different concentrations of myco-
lactones A/B (1a/b) from a stock solution in DMSO. All chambers, in-
cluding a control without mycolactone, contained DMSO at a final con-
centration of 0.12% (v/v). After incubation for 24 or 48 h, the fibroblasts
were washed once with PBS and fixed with 4% formaldehyde (Medite)
for 20 min. After an additional washing step with PBS, the cells were per-
meabilized with Triton X-100 (0.1% in PBS) for 20 min, rinsed once with
PBS, and incubated in blocking solution (2% FBS in PBS) for an addi-
tional 20 min. The actin cytoskeleton was stained with Texas Red-X phal-
loidin (1% in blocking solution) and incubated for 1 h at room tempera-
ture. The cells were washed with blocking solution, then with PBS prior
to treatment with a solution of ProLong Gold antifade reagent (Invitro-
gen) containing DAPI to stain the nuclei. The slides were fitted with
cover slips and were kept in the dark until further usage. The morpholo-
gy of the fibroblasts was analyzed by fluorescence microscopy (Leica
DM 5000B) and representative images at a magnification of 960ꢃ were
acquired.
For flow cytometry analysis, 50000 L929 murine fibroblasts were seeded
into 24-well plates (Falcon) and allowed to adhere overnight. The
medium was then aspirated and replaced by 500 mL of medium contain-
ing different concentrations of mycolactones A/B (1a/b) and 0.12%
DMSO (v/v). After incubation for 24, 48, or 72 h, cells were harvested
and stained with annexin V-FITC and PI according to the manufacturerꢂs
protocol (annexin V kit, Calbiochem PF032). Staining of cells was ana-
lyzed by flow cytometry using an FACS Calibur device (Becton Dickin-
son). The assay was performed in triplicate and time-course experiments
were performed in duplicate.
+
774 cmꢀ1; HRMS (ESI): calcd for C62H112NaO9Si3 [M+Na]+: 1107.7506;
found: 1107.7499.
Mycolactones A/B (1a/1b): TBAF (1 m in THF, 100 mL, 100 mmol) was
added to a solution of 62 (7.2 mg, 6.6 mmol) in THF (1 mL) at RT. The
slightly orange solution was stirred at RT for 75 min. n-Hexane (1.5 mL)
was then added and the solution was directly applied to a flash column.
Purification by flash column chromatography (EtOAc!EtOAc contain-
ing 3% MeOH) yielded 4.2 mg (85%, HPLC purity >73%) of mycolac-
tones A/B as an inseparable mixture of isomers (E-D4’,5’/Z-D4’,5’, 1:1,
minor isomers ca. 10%) as a yellow oil. Rf (EtOAc/MeOH, 97:3)=0.27;
[a]2D5 =ꢀ40.428 (c=0.63, MeOH). Z-D4’,5’ isomer (mycolactone A):
1H NMR (500 MHz, [D6]acetone): d=7.92 (d, J=15.7 Hz, 1H), 6.64 (dd,
J=15.2, 11.1 Hz, 1H), 6.40 (d, J=15.2 Hz, 1H), 6.34 (s, 1H), 6.16 (d, J=
11.1 Hz, 1H), 5.94 (d, J=15.7 Hz, 1H), 5.59 (d, J=8.6 Hz, 1H), 5.14–5.09
(m, 1H), 5.04 (d, J=9.8 Hz, 1H), 4.91–4.86 (m, 1H), 4.72–4.68 (m, 1H),
4.30–4.25 (m, 3H), 4.04 (s, 1H), 4.05–3.90 (m, 3H), 3.70–3.65 (m, 1H),
3.53–3.48 (m, 1H), 2.53–2.45 (m, 1H), 2.43–2.35 (m, 2H), 2.15–1.92 (m,
7H), 2.02 (s, 3H), 1.97 (s, 3H), 1.91 (s, 3H), 1.83 (dd, J=13.1, 8.5 Hz,
1H), 1.72–1.63 (m, 3H), 1.70 (s, 3H), 1.64 (s, 3H), 1.60–1.46 (m, 4H),
1.42–1.35 (m, 1H), 1.14 (d, J=6.2 Hz, 3H), 1.13 (d, J=6.3 Hz, 3H), 0.98
(d, J=6.7 Hz, 3H), 0.89 (d, J=6.8 Hz, 3H), 0.89 ppm (d, J=7.0 Hz, 3H);
13C NMR (125 MHz, [D6]acetone): d=173.3, 166.9, 143.1, 141.8, 140.0,
137.3, 137.3, 134.9, 134.9, 134.7, 133.5, 132.1, 131.2, 125.2, 123.9, 119.7,
79.3, 77.0, 76.3, 75.8, 72.4, 69.0, 67.7, 46.4, 44.4, 43.8, 41.9, 40.5, 36.0, 35.4,
32.8, 31.5, 29.7, 24.6, 24.3, 21.0, 20.8, 20.6, 17.6, 17.2, 16.2, 15.9, 15.0,
13.4 ppm. E-D4’,5’ isomer (mycolactone B): 1H NMR (500 MHz,
[D6]acetone): d=7.37 (d, J=15.6 Hz, 1H), 6.65 (dd, J=15.1, 11.2 Hz,
1H), 6.47 (s, 1H), 6.44 (d, J=15.1 Hz, 1H), 6.37 (d, J=11.2 Hz, 1H),
5.89 (d, J=15.6 Hz, 1H), 5.60 (d, J=8.8 Hz, 1H), 5.14–5.10 (m, 1H), 5.03
(d, J=9.7 Hz, 1H), 4.92–4.87 (m, 1H), 4.73–4.69 (m, 1H), 4.30–4.20 (m,
4H), 4.04 (s, 1H), 4.01–3.91 (m, 3H), 3.70–3.65 (m, 1H), 3.54–3.48 (m,
1H), 2.54–2.46 (m, 1H), 2.44–2.35 (m, 2H), 2.15–1.94 (m, 7H), 2.08 (s,
3H), 2.07 (s, 3H), 1.90 (s, 3H), 1.88–1.80 (m, 1H), 1.74–1.62 (m, 3H),
1.71 (s, 3H), 1.64 (s, 3H), 1.62–1.46 (m, 4H), 1.42–1.35 (m, 1H), 1.13 (d,
J=6.1 Hz, 3H), 1.11 (d, J=6.2 Hz, 3H), 0.98 (d, J=6.7 Hz, 3H), 0.90 (d,
J=6.4 Hz, 3H), 0.89 ppm (d, J=6.4 Hz, 3H); 13C NMR (125 MHz,
[D6]acetone): d=173.3, 166.9, 151.2, 144.3, 140.4, 137.3, 136.1, 135.3,
134.7, 133.5, 133.2, 131.2, 125.2, 123.9, 117.4, 79.3, 77.0, 76.3, 75.8, 72.4,
69.0, 67.7, 46.4, 44.4, 43.8, 41.9, 40.5, 36.0, 35.4, 32.8, 31.5, 29.7, 24.6, 24.3,
20.8, 20.6, 17.2, 17.2, 16.2, 15.9, 15.0, 14.3, 13.4 ppm; IR (film, acetone):
n˜ =3391, 2966, 2934, 2369, 2359, 2342, 2331, 1706, 1456, 1437, 1386, 1304,
Acknowledgements
Financial support of this work by the Novartis Institutes for Developing
World Medical Research (Ph.D. fellowship to P.G.) is gratefully acknowl-
edged. We are indebted to Kurt Hauenstein for practical advice, to Bern-
hard Pfeiffer and Fabienne Gaugaz for NMR support, and to the ETHZ-
LOC MS-Service for high-resolution mass spectral acquisition. This proj-
ect was carried out within the framework of the COST Action 804
(“Chemical Biology with Natural Products”). Part of this work was sup-
ported by the Volkswagen Foundation (grant to G.P.).
[1] K. M. George, D. Chatterjee, G. Gunawardana, D. Welty, J.
Hayman, R. Lee, P. L. Small, Science 1999, 283, 854–857.
[2] For recent general reviews on Buruli ulcer, see: a) C. Demangel,
T. P. Stinear, S. T. Cole, Nat. Rev. Microbiol. 2009, 7, 50–60; c) D. S.
Walsh, F. Portaels, W. M. Meyers, Dermatol. Clin. 2011, 29, 1–8.
[3] M. Wansbrough-Jones, R. Phillips, Lancet 2006, 367, 1849–1858.
[4] J. Guarner, J. Bartlett, E. A. S. Whitney, P. L. Raghunathan, Y.
Stienstra, K. Asamoa, S. Etuaful, E. Klutse, E. Quarshie, T. S. van
der Werf, W. T. A. van der Graaf, C. H. King, D. A. Ashford, Emerg-
ing Infect. Dis. 2003, 9, 651–656.
[5] D. Schꢀtte, G. Pluschke, Expert Opin. Biol. Ther. 2009, 9, 187–200.
[6] a) J. Hayman, Int. J. Epidemiol. 1991, 20, 1093–1098; b) H. Aiga, T.
Amano, S. Cairncross, J. Adomako, O. K. Nanas, S. Coleman, Am. J.
Trop. Med. Hyg. 2004, 71, 387–392; c) P. L. Raghunathan, E. A. S.
Whitney, K. Asamoa, Y. Stienstra, T. H. Taylor, G. K. Amofah, D.
Ofori-Adjei, K. Dobos, J. Guarner, S. Martin, S. Pathak, E. Klutse,
S. Etuaful, W. I. A. van der Graaf, T. S. van der Werf, C. H. King,
J. W. Tappero, D. A. Ashford, Clin. Infect. Dis. 2005, 40, 1445–1453.
[7] WHO, Wkly. Epidemiol. Rec. 2008, 83, 145–154.
[8] L. Marsollier, R. Robert, J. Aubry, J. P. Saint Andre, H. Kouakou, P.
Legras, A. L. Manceau, C. Mahaza, B. Carbonnelle, Appl. Environ.
Microbiol. 2002, 68, 4623–4628.
+
1273, 1254, 1160, 1012, 983 cmꢀ1; HRMS (ESI): calcd for C44H70NaO9
[M+Na]+: 765.4912; found: 765.4909.
Biological testing: L929 murine fibroblasts (passage 3–6) were cultivated
at 378C in RPMI medium supplemented with 5% fetal calf serum and
2 mm l-glutamine in a humid atmosphere with 5% CO2.
[9] P. D. Johnson, J. Azuolas, C. J. Lavender, E. Wishart, T. P. Stinear,
J. A. Hayman, L. Brown, G. A. Jenkin, J. A. Fyfe, Emerging Infect.
Dis. 2007, 13, 1653–1660.
13030
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 13017 – 13031

Total Synthesis of Mycolactones A/B Based on RCM
FULL PAPER
[10] a) WHO, Wkly. Epidemiol. Rec. 2004, 79, 145–152; b) S. Etuaful, B.
Carbonnelle, J. Grosset, S. Lucas, C. Horsfield, R. Phillips, M.
Evans, D. Ofori-Adjei, E. Klustse, J. Owusu-Boateng, G. K. Amedo-
fu, P. Awuah, E. Ampadu, G. Amofah, K. Asiedu, M. Wansbrough-
Jones, Antimicrob. Agents Chemother. 2005, 49, 3182–3186.
[11] WHO, Provisional guidance on the role of specific antibiotics in the
management of Mycobacterium ulcerans disease (Buruli ulcer).
(WHO/CDS/CPE/GBUI/2004), WHO 2004.
[30] Treatment of 23 with 2 equiv of isopropenylmagnesium bromide in
the presence of 20 mol% of CuI for 24 h (ꢀ788C to RT, THF) did
not give any conversion; the reaction of 23 with 5 equiv of isoprope-
nylmagnesium bromide and 1 mol% of Li2CuCl4 for 24 h (08C to
RT, THF) resulted in about 30% conversion into unidentified prod-
uct(s).
[31] E. M. Carreira, J. Dubois, J. Am. Chem. Soc. 1994, 116, 10825–
10826.
[12] a) A. Tanghe, P. Y. Adnet, T. Gartner, K. Huygen, Infect. Immun.
2007, 75, 2642–2644; b) K. Huygen, O. Adjei, D. Affolabi, G. Bret-
zel, C. Demangel, B. Fleischer, R. C. Johnson, J. Pedrosa, D. M.
Phanzu, R. O. Phillips, G. Pluschke, V. Siegmund, M. Singh, T. S. van
der Werf, M. Wansbrough-Jones, F. Portaels, Med. Microbiol. Immu-
nol. 2009, 198, 69–77.
[13] For reviews, see: a) H. Hong, C. Demangel, S. J. Pidot, P. F. Leadlay,
T. Stinear, Nat. Prod. Rep. 2008, 25, 447–454; b) S. J. Pidot, H.
Hong, T. Seemann, J. L. Porter, M. J. Yip, A. Men, M. Johnson, P.
Wilson, J. K. Davies, P. F. Leadlay, T. O. Stinear, BMC Genomics
2008, 9, 462.
[32] D. B. Dess, J. C. Martin, J. Org. Chem. 1983, 48, 4155–4156.
[33] a) B. O. Lindgren, H. Nilsson, Acta Chem. Scand. 1973, 27, 888–890;
b) G. A. Kraus, B. Roth, J. Org. Chem. 1980, 45, 4825–4830; c) B. S.
Bal, W. E. Childers, H. W. Pinnick, Tetrahedron 1981, 37, 2091–
2096.
[34] a) G. Hçfle, W. Steglich, H. Vorbrꢀggen, Angew. Chem. 1978, 90,
602–615; Angew. Chem. Int. Ed. Engl. 1978, 17, 569–583.
[35] a) M. Scholl, T. M. Trnka, J. P. Morgan, R. H. Grubbs, Tetrahedron
Lett. 1999, 40, 2247–2250; b) R. H. Grubbs, Tetrahedron 2004, 60,
7117–7140; c) T. M. Trnka, R. H. Grubbs, Acc. Chem. Res. 2000, 33,
18–29.
[14] a) K. M. George, L. Pascopella, D. M. Welty, P. L. C. Small, Infect.
Immun. 2000, 68, 877–883; b) C. Bozzo, R. Tiberio, F. Graziola, G.
Pertusi, G. Valente, E. Colombo, P. L. Small, G. Leigheb, Microbes
Infect. 2010, 12, 1258–1263.
[36] E. J. Corey, P. L. Fuchs, Tetrahedron Lett. 1972, 13, 3769–3772.
[37] D. W. Hart, J. Schwartz, J. Am. Chem. Soc. 1974, 96, 8115–8116.
[38] R. P. van Summeren, B. L. Feringa, A. J. Minnaard, Org. Biomol.
Chem. 2005, 3, 2524–2533.
[15] H. Hong, T. Stinear, J. Porter, C. Demangel, P. F. Leadlay, ChemBio-
Chem 2007, 8, 2043–2047.
[16] A. Mve-Obiang, R. E. Lee, F. Portaels, P. L. Small, Infect. Immun.
2003, 71, 774–783.
[39] a) H. J. Jennings, J. K. N. Jones, Can. J. Chem. 1962, 40, 1408–1414;
b) H. J. Jennings, J. K. N. Jones, Can. J. Chem. 1963, 41, 1151.
[40] a) O. Mitsunobu, M. Yamada, Bull. Chem. Soc. Jpn. 1967, 40, 2380–
2382; b) O. Mitsunobu, Synthesis 1981, 1–28.
[17] R. E. Simmonds, F. V. Lali, T. Smallie, P. L. Small, B. M. Foxwell, J.
Immunol. 2009, 182, 2194–2202.
[18] a) A. B. Benowitz, S. Fidanze, P. L. C. Small, Y. Kishi, J. Am. Chem.
Soc. 2001, 123, 5128–5129; b) F. Song, S. Fidanze, A. B. Benowitz,
Y. Kishi, Org. Lett. 2002, 4, 647–650; c) Y. Kishi, Proc. Natl. Acad.
Sci. USA 2011, 108, 6703–6708.
[41] M. K. Gurjar, J. Cherian, Heterocycles 2001, 55, 1095–1103.
[42] W. Qu, M.-P. Kung, C. Hou, T. E. Benedum, H. F. Kung, J. Med.
Chem. 2007, 50, 2157–2165.
[43] For a review, see: E. Negishi (Ed.), Handbook of Organopalladium
Chemistry for Organic Synthesis, Part III, Wiley, New York, 2002,
pp. 215–1119.
[19] F. Song, S. Fidanze, A. B. Benowitz, Y. Kishi, Tetrahedron 2007, 63,
5739–5753.
[20] K. L. Jackson, W. Li, C.-L. Chen, Y. Kishi, Tetrahedron 2010, 66,
2263–2272.
[21] T. C. Judd, A. Bischoff, Y. Kishi, S. Adusumilli, P. L. C. Small, Org.
Lett. 2004, 6, 4901–4904.
[22] S. Aubry, R. E. Lee, E. A. Mahrous, P. L. Small, D. Beachboard, Y.
Kishi, Org. Lett. 2008, 10, 5385–5388.
[23] H. J. Kim, Y. Kishi, J. Am. Chem. Soc. 2008, 130, 1842–1844.
[24] a) N. Yin, G. Wang, M. Qian, E. Negishi, Angew. Chem. 2006, 118,
2982–2986; Angew. Chem. Int. Ed. 2006, 45, 2916–2920; b) G.
Wang, N. Yin, E.-i. Negishi, Chem. Eur. J. 2011, 17, 4118–4130.
[25] a) M. D. Alexander, S. D. Fontaine, J. J. La Clair, A. G. Dipasquale,
A. L. Rheingold, M. D. Burkart, Chem. Commun. 2006, 4602–4604;
b) K. S. Ko, M. D. Alexander, S. D. Fontaine, J. E. Biggs-Houck, J. J.
La Clair, M. D. Burkart, Org. Biomol. Chem. 2010, 8, 5159–5165.
[26] F. Feyen, A. Jantsch, K.-H. Altmann, Synlett 2007, 415–418.
[27] a) A. Deiters, S. F. Martin, Chem. Rev. 2004, 104, 2199–2238;
b) K. C. Nicolaou, P. G. Bulger, D. Sarlah, Angew. Chem. 2005, 117,
4564–4601; Angew. Chem. Int. Ed. 2005, 44, 4490–4527.
[28] P. K. Jadhav, K. S. Bhat, P. T. Perumal, H. C. Brown, J. Org. Chem.
1986, 51, 432–439.
[44] For reviews, see: a) N. Miyaura, A. Suzuki, Chem. Rev. 1995, 95,
2457–2483; b) S. Kotha, K. Lahiri, D. Kashinath, Tetrahedron 2002,
58, 9633–9695.
[45] A. Arefolov, J. S. Panek, J. Am. Chem. Soc. 2005, 127, 5596–5603.
[46] A. K. Ghosh, W. Liu, J. Org. Chem. 1997, 62, 7908–7909.
[47] G. E. Keck, K. H. Tarbet, L. S. Geraci, J. Am. Chem. Soc. 1993, 115,
8467–8468.
[48] P. Schwab, R. H. Grubbs, J. W. Ziller, J. Am. Chem. Soc. 1996, 118,
100–110.
[49] J. S. Kingsbury, J. P. A. Harrity, P. J. Bonitatebus, A. H. Hoveyda, J.
Am. Chem. Soc. 1999, 121, 791–799.
[50] As suggested by one of the reviewers, one might speculate that the
reason for the fact that 25, but not 65, undergoes RCM is due to a
stereoelectronic effect of the tosyl group that either helps to activate
the metal carbene intermediate by complexing to the metal or helps
to place the two double bonds in close proximity.
[51] S. Adusumilli, A. Mve-Obiang, T. Sparer, W. Meyers, J. Hayman,
P. L. Small, Cell. Microbiol. 2005, 7, 1295–1304.
[52] D. S. Snyder, P. L. Small, Microb. Pathog. 2003, 34, 91–101.
[53] E. Coutanceau, L. Marsollier, R. Brosch, E. Perret, P. Goossens, M.
Tanguy, S. T. Cole, P. L. Small, C. Demangel, Cell. Microbiol. 2005,
7, 1187–1196.
[29] W. Oppolzer, J. Blagg, I. Rodriguez, E. Walther, J. Am. Chem. Soc.
1990, 112, 2767–2772.
Received: June 12, 2011
Published online: October 4, 2011
Chem. Eur. J. 2011, 17, 13017 – 13031
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
13031