An improved route to (+)-tedanolide and analysis of its subtle effects controlling conformation and biological behaviour

Article abstract of DOI:10.1002/chem.201103038

The improved synthesis of the antitumor compound (+)-tedanolide is described through an aldol coupling of bis-ketone 7. This modification shortens the synthetic steps in the endgame and provides rapid access to this biologically important natural product. Additionally, it serves as a probe in order to unravel the conformational effects that impede or enable its successful synthesis. Having this way access to des-epoxy-tedanolide, its biological characterization surprisingly unravelled the mode of action to resemble candidaspongiolide rather than deoxytedanolide. Copyright

Full text of DOI:10.1002/chem.201103038

DOI: 10.1002/chem.201103038
An Improved Route to (+)-Tedanolide and Analysis of Its Subtle Effects
Controlling Conformation and Biological Behaviour
Nina Diaz,[a] Mingzhao Zhu,[a] Gunnar Ehrlich,[a] Ulrike Eggert,[a] Yazh Muthukumar,[c]
Florenz Sasse,[c] and Markus Kalesse*[a, b]
Dedicated to Professor Ekkehard Winterfeldt on the occasion of his 80th birthday
Abstract: The improved synthesis of the antitumor compound (+)-tedanolide is
described through an aldol coupling of bis-ketone 7. This modification shortens
the synthetic steps in the endgame and provides rapid access to this biologically
important natural product. Additionally, it serves as a probe in order to unravel
the conformational effects that impede or enable its successful synthesis. Having
this way access to des-epoxy-tedanolide, its biological characterization surprisingly
unravelled the mode of action to resemble candidaspongiolide rather than deoxy-
tedanolide.
Keywords: aldol · antitumor agents ·
natural products
vinylogous aldol
· tedanolide ·
Introduction
studies. Thus far, several research groups have investigated
the construction of fragments and the total synthesis of
1 and 2 has been completed by Smith,[6] Roush[7] and Ka-
lesse.[8]
The marine macrolactone (+)-tedanolide (1 in Figure 1) was
isolated from the Caribbean sponge, Tedania ignis by
Schmitz et al. in 1984, and was found to be cytotoxic against
KB and PS tumor cell lines in picomolar range.[1] Of consid-
erable importance, 1 increases the lifespan of mice implant-
ed with lymphocytic leukaemia significantly.[2] A congener,
(+)-13-deoxytedanolide (2) was discovered later by Fusetani
et al. from the sponge Mycale adhaerens, and it shows simi-
lar biological activity in inhibiting the growth of P388 tu-
mours.[3] Furthermore, the isolation of additional metabo-
lites, namely tedanolide C (3), the candidaspongiolides (4,
a complex mixture of acyl esters) and candidaspongiolide B
(5) were reported by Ireland[4] and McKee.[5] The cytotoxic
activity of these compounds was shown to be due to inhibi-
tion of cellular protein synthesis.
As a consequence of the biological profiles and challeng-
ing structures 1 and 2 have intrigued a variety of synthetic
[a] Dr. N. Diaz, Dr. M. Zhu, Dr. G. Ehrlich, U. Eggert,
Prof. Dr. M. Kalesse
Leibniz Universitꢀt Hannover, Institut fꢁr Organische Chemie
Schneiderberg 1B, 30167 Hannover (Germany)
Fax : (+49)511-3011
E-mail: Markus.Kalesse@oci.uni-hannover.de
[b] Prof. Dr. M. Kalesse
Department of Medicinal Chemistry
Helmholtz Center for Infection Research
Inhoffenstr 7, 38124 Braunschweig (Germany)
[c] Y. Muthukumar, Dr. F. Sasse
Department of Chemical Biology
Helmholtz Center for Infection Research
Inhoffenstr 7, 38124 Braunschweig (Germany)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201103038.
Figure 1. Structures of tedanolides and candidaspongiolides.
4946
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 4946 – 4952

FULL PAPER
Results and Discussion
In our retrosynthetic analysis of
(+)-tedanolide (1) macrocyclic
intermediate 6 lead is further dis-
sected into methyl ketone 7 and
aldehyde 8 (Scheme 1). A signifi-
cant improvement would be to
install the carbonyl group at C5
already at an early stage, as em-
ployed in the synthesis of (+)-13-
deoxytedanolide (2) by Roush,[7]
thus shortening the number of
steps in the end-game of the syn-
thesis. Additionally, the incorpo-
ration of the Roush fragment
would provide additional data
that could help to rationalize the
Scheme 2. Deprotection of TES group.
different conformational behaviour of macrolactones of the
Roush route compared to those obtained by the Kalesse
route. Here we describe an improved route to (+)-tedano-
lide (1) as a combination of our previous synthesis of the
the reaction proceeded for 6–8 h, only a small amount (ca.
5–10%) of the starting material 12 remained. However,
after prolonged reaction times (up to 12 h) side product 14
began to accumulate and consequently, the reaction was
stopped as soon as 14 could be detected by TLC.
The crude product 13 was directly submitted to Dess–
Martin oxidation and provided 15 in 72% yield over two
steps. Next, the C11-PMB group was removed with DDQ
and subsequent oxidation with TPAP/NMO[12] provided
ketone 7 in good yield (Scheme 3).
ꢀ
C1 C12 segment in combination with Roushꢃs strategy to
implement the C5-ketone at an early stage. Ketone 7 can be
prepared from ester 11 which in turn is obtained through
a highly stereoselective vinylogous Mukaiyama aldol reac-
tion (VMAR) of silyl ketene acetal 10 and aldehyde 9.[9]
Further transformations including a Sharpless dihydroxyla-
tion,[10] methylation and TBS protection furnished com-
pound 12 (Scheme 2). However, we encountered unexpected
problems when we tried to deprotect the C5-TES group of
12 in order to establish the corresponding ketone
(Scheme 2). Conditions employing fluoride sources such as
TBAF, TBAF buffered with carboxylic acids or HF/pyridine
resulted in decomposition of the starting material. On the
other hand, using acetic acid in methanol/dichloromethane
gave clean conversion to six-membered lactone 14.
Finally, removal of the TES group was achieved when ex-
cessive PPTS was employed in methanol/THF (1:3). After
Scheme 3. Synthesis of diketone 7 and aldol coupling with aldehyde 8.
The viability of 7 as a segment in our route was confirmed
by performing an aldol reaction between 7 and 8. Using
LiHMDS as base provided the aldol product 17 as a single
stereoisomer (Felkin product), albeit in low yields (21%,
80% brsm). At this stage the obvious modifications such as
additives (e.g., HMPA) different counter ions or solvents
did not significantly change the yields and we therefore ab-
stained from optimizing the yield of this reaction and in-
stead resubmitted aldehyde 8 and ketone 7 to repeated
aldol reactions (90% of unreacted 7 and 80% of unreacted
8 were recovered). The desired (13S)-configuration was
later confirmed by comparison of the spectroscopical data
to intermediate 21 also described in our previous synthesis.
The synthesis continued by protecting the newly formed sec-
Scheme 1. Retrosynthetic analysis of 1. MMTr=monomethoxy trityl.
Chem. Eur. J. 2012, 18, 4946 – 4952
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4947

M. Kalesse et al.
ondary alcohol as the TBS ether 18. Here, the retro aldol re-
action is responsible for the low yield observed. The MMTr
group was removed using hexafluoroisopropanol in
MeOH.[13] Subsequently, hydroxy ester 19 was treated with
[PdACHTUNGTRENNUNG(PPh3)2Cl2] and Bu3SnH to liberate the corresponding
seco-acid in quantitative yield. The macrolactonization was
successfully performed when the seco-acid was subjected to
the Mitsunobu protocol directly after Pd-catalyzed deprotec-
tion of the allyl ester.
In order to establish the C15 ketone, the PMB group was
removed with DDQ followed by Dess–Martin oxidation.[11]
This transformation proved to be problematic in the Roush
approach due to the formation of the stable hemiacetal that
prevented oxidation.[7a] However, in our previous route no
hemiacetal formation was observed at all. We originally con-
tributed this different behavior to the opposite (S)-configu-
ration at C15 compared to the Roush synthesis. In the
Roush synthesis due to the (15R) epimer, a stable hemiace-
tal with all substituents equatorial was generated whereas in
our route at least one substituent would be axial (see
Scheme 5). However, here using Roushꢃs northern segment,
a mixture of the open-chain and hemiacetal form was ob-
served telling that also the hybridization at C5 adds to the
undesired hemiacetal formation. The subsequent oxidation
with the Dess–Martin periodinane however generated
ketone 21 (Scheme 4). At this stage our improved route
merges with our first total synthesis of tedanolide (1) and
the spectroscopical data of compound 21 derived through
the two different routes were identical in all respects.
Scheme 5. Macrolactonization and end game strategy.
the superior combination for the TBS deprotection, as de-
scribed by Roush. Under these conditions, the deprotection
of four silyl ethers only proceeded sluggishly to half-conver-
sion after four days. By separation of the partial deprotected
material through column chromatography and re-subjecting
to the deprotection conditions, full conversion was achieved.
Finally, the last-step epoxidation remained to be accom-
plished. Since Smith and co-workers had shown that the di-
rected epoxidation of the D18 allylic alcohol exclusively oc-
curred with the correct stereoselectivity we envisaged to
perform the epoxidation with the completely deprotected
tetrahydroxy lactone.[8] The presence of the unprotected C7
allylic alcohol is tolerated and the D18 olefin is predomi-
nately epoxidized with the correct facial selectivity.
The final steps to be accomplished were the global TBS
deprotection and the critical final step epoxidation. Thor-
ough experimentation using combinations of fluoride and
proton sources revealed the combination of HF·Et3N to be
Cytotoxicity of des-epoxy-tedanolide: Besides the unexpect-
ed, yet important observations that the conformation is con-
trolled not only by the configuration at C15 but also by the
oxidation state/hybridization at C5, this route allowed ob-
taining sufficient material of des-epoxy-tedanolide (22) for
biological tests which were not accessible from the natural
product by removal of the epoxide. On the other hand, the
contribution of the epoxide to its biological performance
could so far only be deduced indirectly from co-crystalliza-
tion of deoxytedanolide to the ribosome.[14] The cytotoxic ef-
fects of des-epoxy-tedanolide (22) were evaluated with
mouse fibroblasts L-929, and human epidermoid cancer cells
A-431 using MTT assays. It was shown to be one order of
magnitude less active than tedanolide. The IC50 values were
0.53 ngmLꢀ1 and 0.3 ngmLꢀ1, respectively. FACS analysis
showed up to 33% apoptotic cells when incubated with
Scheme 4. Hemiacetal formation of (+)-tedanolide intermediates.
4948
www.chemeurj.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 4946 – 4952

An Improved Route to (+)-Tedanolide
FULL PAPER
50 ngmLꢀ1 des-epoxy-tedanolide after 24 h (of which 17.5%
were in late apoptosis) as compared to 6% in the control
(see Supporting Information). However, the compound
failed to show any activity against yeasts tested, Saccharo-
myces cerevisiae and Candida albicans. This is supposed to
be due to reduced cellular uptake. We also checked if the
cytotoxicity induced by des-epoxy-tedanolide was reversible
(see Supporting Information). Cultured A-431 cells were in-
cubated with various concentrations of des-epoxy-tedanolide
for either 6 or 24 h, washed with EBSS and grown with
fresh media for the cells to revive. After four days, an MTT
assay was performed to check cell viability. Control cells
were treated with methanol or des-epoxy-tedanolide (22)
without removing the drug for five days. The results showed
that the cytotoxicity was partially reversible when washed
out after 6 h. Comparing the IC50 values we calculated a re-
versibility factor of 10. After 24 h the effect was not reversi-
ble any more.
Translation inhibition assays: To determine the translation
inhibition potentials of des-epoxy-tedanolide an in vitro
translation inhibition assay was carried out using rabbit re-
ticulocyte lysate. The inhibition of translation was measured
by luciferase activity output. Cycloheximide, a known trans-
lation inhibitor was used as a positive control. Des-epoxy-te-
danolide inhibited translation with an IC50 of 70 ngmLꢀ1
whereas that of cycloheximide was 150 ngmLꢀ1 (Figure 2a).
Des-epoxy-tedanolide (22) also inhibited translation in the
wheat germ lysate system with an IC50 of 300 ngmLꢀ1 (data
not shown).
The translation inhibition was also checked in living cells
which were transfected with a pRLSV40 vector coding for
Renilla luciferase. 48 h after transfection, KB-3-1 cells were
treated with varying concentrations of des-epoxy-tedanolide
and cycloheximide for 4 h. Cyclohexi-
Figure 2. a) Inhibition of protein synthesis by des-epoxy-tedanolide (22).
in vitro translation inhibition using rabbit reticulocyte system. B) Inhibi-
tion of protein synthesis by des-epoxy-tedanolide (22). In vivo translation
inhibition of KB-3-1 cells transiently transfected with pRLSV40 vector
expressing Renilla luciferase.
with antibodies against eukaryotic initiation factor-2
a (eIF2a) and phosphorylated eIF2a (green). Control cells
were treated with the solvent only. As shown in Figure 3
des-epoxy-tedanolide induced phosphorylation of eIF2a ac-
counting for its inhibitory activities but it did not induce
mide inhibited in vivo translation with
an IC50 of 30 ngmLꢀ1 whereas des-
epoxy-tedanolide showed an IC50 of
0.1 ngmLꢀ1,
approximately
(Fig-
ure 2b). This value corresponds to the
IC50 values obtained from the cytotox-
icity assays. Des-epoxy-tedanolide in-
hibited translation 300-fold more than
cycloheximide.
Candidaspongiolide inhibits transla-
tion by induction of phosphorylation
of eIF2 alpha by PKR kinase.[15] In
contrast, the translation inhibitory
effect of 13-deoxytedanolide (2) is due
to the induction of ribotoxic stress.[3b,c]
To evaluate which of the above two
mechanisms led to inhibitory effects of
des-epoxy-tedanolide, cultured PtK2
cells were treated with 50 ngmLꢀ1 of
des-epoxy-tedanolide for 4 h. At the
end of incubation period cells were
Figure 3. Immunofluorescence investigations and comparison of the mode of action to the candidas-
pongiolides. Induction of eIF2a phosphorylation by des-epoxy-tedanolide (22) in cultured PtK2 cells.
Immunofluorescence stainings of eIF2a (red) and phosphorylated eIF2a (green) and overlay images
(from left to right) are shown for cells that were treated with des-epoxy-tedanolide (15 ngmLꢀ1
;
bottom row) in comparison with control cells treated with the vehicle (methanol) only (top row). In
des-epoxy-tedanolide treated cells there was a significant increase in p-eIF2a compared to the MeOH
fixed with formaldehyde and stained treated cells but no apparent change in the total amount of eIF2a.
Chem. Eur. J. 2012, 18, 4946 – 4952
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4949

M. Kalesse et al.
(20 mL). The combined organic layers were dried over Na2SO4, and con-
centrated in vacuo. The residue was purified by flash chromatography to
afford TBS ether 18 (23 mg, 58%) as a colorless oil together with un-
reacted 17 (5 mg). Rf =0.40 (hexane/ethyl acetate 20:1); [a]2D3 =31.1 (c =
stress granules, suggesting the mode of action is similar to
candidaspongiolide. Phosphorylation of eIF2a is one of the
mechanisms of translation regulation in eukaryotic cells and
prolonged inhibition of translation leads to apoptosis as
seen in the case of des-epoxy-tedanolide.
1
0.92, CHCl3); H NMR (400 MHz, C6D6): d=7.86–7.82 (m, 4H), 7.64 (d-
like, 2H, J=9.0 Hz, 2H), 7.33–7.38 (m, 4H), 7.23–7.18 (m, 4H), 6.94 (d-
like, J=8.7 Hz, 2H), 6.90 (d-like, J=9.0 Hz, 2H), 5.87 (ddt, J=17.2,
10.4 Hz, 5.8, 1H), 5.80 (dm, J=9.7 Hz, 1H), 5.39 (pent, J=6.4 Hz, 1H),
5.36 (tm, J=10.7 Hz, 1H), 5.31 (d, J=8.9 Hz, 1H), 5.28 (dm, J=17.2 Hz,
1H), 5.11 (dm, J=10.4 Hz, 1H), 4.92–4.88 (m, 1H), 4.76 (br, 1H), 4.73
(d, J=17.8 Hz, 1H), 4.72 (d, J=4.8 Hz, 1H), 4.63 (d, J=5.9 Hz, 1H),
4.63 (d, J=17.8 Hz, 1H), 4.64 (ddt, J=13.2, 5.8, 1.4 Hz, 1H), 4.58 (ddt,
J=13.2, 5.8, 1.4 Hz, 1H), 4.38 (dd, J=7.6, 3.0 Hz, 1H), 4.32 (dd, J=5.9,
4.4 Hz, 1H), 3.55 (s, 3H), 3.45 (s, 3H), 3.49–3.43 (m, 1H), 3.47–3.40 (m,
1H), 3.26 (dq, J=4.8, 7.1 Hz, 1H), 3.16 (dd, J=16.9, 6.8 Hz, 1H), 3.08
(dq, J=5.1, 7.1 Hz, 1H), 3.01 (dd, J=16.9, 5.6 Hz, 1H), 3.00–3.05 (m,
1H), 2.58–2.65 (m, 1H), 1.87 (d, J=1.0 Hz, 3H), 1.79 (d, J=1.1 Hz, 3H),
1.54 (dd, J=6.4, 1.2 Hz, 3H), 1.48 (d, J=7.1 Hz, 3H), 1.42 (d, J=7.1 Hz,
3H), 1.39 (d, J=7.1 Hz, 3H), 1.32 (d, J=6.9 Hz, 3H), 1.26 (s, 9H), 1.17
(s, 9H), 1.15 (s, 9H), 1.14 (s, 9H), 1.09 (d, J=6.7 Hz, 3H), 0.53 (s, 3H),
0.42 (s, 3H), 0.42 (s, 3H), 0.31 ((s, 2ꢄ3H), 0.28 (s, 3H), 0.27 (s, 3H),
0.25 ppm (s, 3H); 13C NMR (100 MHz, C6D6): d=211.91, 209.03, 171.35,
159.40, 159.19, 145.94, 145.30, 138.33, 136.30, 135.51, 132.19, 132.10,
131.83, 131.29, 129.47, 129.32, 129.20, 128.17, 127.97, 127.10, 127.06,
126.62, 121.78, 118.72, 113.73, 113.47, 87.37, 82.05, 79.11, 77.40, 76.77,
74.93, 74.83, 70.46, 65.54, 62.88, 60.16, 54.72, 49.31, 47.61, 47.27, 47.21,
45.85, 42.27, 30.65, 30.23, 26.57 (2C), 26.31, 26.09, 21.32, 18.69, 18.64,
18.60, 18.52, 16.88, 13.65, 13.55, 12.98, 12.12, 11.90, 11.48, ꢀ3.45 (2C),
ꢀ3.88, ꢀ3.96, ꢀ4.12, ꢀ4.69, ꢀ4.70, ꢀ4.80 ppm; HRMS (ESI): m/z: calcd
for C87H138O13Si4Na [M+Na]+: 1525.9112, found: 1525.9133.
Conclusion
In summary we have reported a new and improved route to
(+)-tedanolide (1) that shifts functional group manipula-
tions from the end game of the synthesis to the early assem-
bly of the building blocks. This strategic rearrangement pro-
vides valuable insight into the subtle difference that governs
the synthesis of this complex natural product. The formation
of the hemiacetal in our route is a very strong indication,
that not only the configuration at C15 but also the presence
of the C5 ketone stabilizes the undesired hemiacetal. Des-
epoxy-tedanolide was shown to have a similar mode of
action as candidaspongiolide. Obviously, the absence of the
epoxide which is known to contribute to binding to the ribo-
some[14] leads to a different mode of action of the otherwise
unchanged natural product.
Alcohol 19: Compound 18 (10 mg, 6.5 mmol) was dissolved in CH2Cl2
(0.5 mL) and (CF3)2CHOH (0.5 mL) was added dropwise at room tem-
perature. The colour of the solution turned yellow. After 10 min, metha-
nol was added dropwise until a slightly yellow colour remained and the
solution was stirred for 8 h. The solution was concentrated in vacuo and
the residue was directly submitted to flash chromatography (hexane/ethyl
acetate 25:1) without additional workup to give hydroxyl ester 19 as a col-
Experimental Section
General procedure and other experimental data available in the Support-
ing Information.
Ketone 15: A solution of TES ether 12 (110 mg, 0.13 mmol) and PPTS
(60 mg, 0.24 mmol) in THF (0.8 mL) and MeOH (3.2 mL) was stirred at
room temperature for 8 h. The solution was diluted with MTBE (50 mL),
washed with H2O (5 mL), NaHCO3 (5 mL), and brine (5 mL). The organ-
ic layer was dried over Na2SO4 and concentrated in vacuo. Without fur-
ther purification, the crude product 13 was dissolved in CH2Cl2 (1 mL).
The solution was cooled to 08C and pre-dried pyridine (50 mL) and a solu-
tion of DMP in CH2Cl2 (15%, w/w, 0.5 mL) was added dropwise. The re-
sulting yellow solution was stirred under N2 for 30 min, and was directly
submitted to flash chromatography (hexane/ethyl acetate 20:1 ! 12:1) to
afford recovered 12 (5 mg) and 15 (65 mg, 72%) as a colourless oil. Rf =
ourless oil (6.0 mg, 73%). Rf =0.40 (hexane/ethyl acetate 20:1); [a]D23
=
61.8 (c = 0.65, CHCl3); 1H NMR (400 MHz, CDCl3): d=7.24 (dm, J=
8.7 Hz, 2H), 6.85, (dm, J=8.7 Hz, 2H), 5.93 (ddt, J=17.2, 10.4, 6.0 Hz,
1H), 5.39 (dm, J=9.7 Hz, 1H), 5.36 (dm, J=17.2 Hz, 1H), 5.33–5.28 (m.
1H), 5.26 (dm, J=10.4 Hz, 1H), 5.18 (dm, J=9.2 Hz, 1H), 5.15 (dq, J=
9.1, 1.7 Hz, 1H), 4.65 (ddt, J=13.0, 6.1, 1.2 Hz, 1H), 4.60 (ddt, J=13.0,
6.1, 1.2 Hz, 1H), 4.47 (d, J=10.8 Hz, 2H), 4.42 (d, J=10.8 Hz, 2H), 4.38
(d, J=4.9 Hz, 1H), 4.29 (d, J=5.6 Hz, 1H), 4.29–4.25 (m, 1H), 4.25 (d,
J=7.8 Hz, 1H), 3.84 (dd, J=5.6, 4.4 Hz, 1H), 3.80 (s, 3H), 3.58–3.64 (m,
1H), 3.56 (dd, J=7.1, 2.6 Hz, 1H), 3.50–3.44 (m, 1H), 3.40 (s, 3H), 3.40–
3.31 (m, 1H), 3.26 (dq, J=9.7, 6.8 Hz, 1H), 2.90 (dq, J=4.4, 7.1 Hz, 1H),
2.79 (dq, J=5.2, 6.9 Hz, 1H), 2.76 (dd, J=17.0, 7.6 Hz, 1H), 2.60 (br,
1H), 2.51 (dd, J=17.0, 4.9 Hz, 1H), 2.24–2.18 (m, 1H), 2.10–2.02 (m,
1H), 1.67 (d, J=1.7 Hz, 3H), 1.61 (dd, J=6.7, 1.7 Hz, 3H), 1.61 (d, J=
1.3 Hz, 3H), 1.11 (d, J=7.1 Hz, 3H), 1.10 (d, J=6.9 Hz, 3H), 0.99 (d, J=
6.7 Hz, 3H), 0.98 (d, J=6.7 Hz, 3H), 0.90 (s, 9H), 0.88 (s, 9H), 0.87 (s,
9H), 0.85 (s, 9H), 0.10 (s, 3H), 0.08 (s, 3H), 0.06 (s, 3H), 0.02 (s, 2ꢄ3H),
0.01 (s, 3H), ꢀ0.04 (s, 3H), ꢀ0.05 ppm (s, 3H); 13C NMR (100 MHz,
CDCl3): d =212.62, 209.23, 171.34, 158.86, 138.16, 135.14, 134.75, 132.42,
131.63, 131.25, 129.01, 125.75, 122.08, 119.17, 113.51, 81.46, 81.35, 78.25,
77.20, 74.52, 73.63, 69.64, 65.62, 63.88, 60.14, 55.23, 48.86, 47.41, 46.88,
46.76, 45.29, 41.69, 30.28, 26.01 (2 CH3), 25.91, 25.77, 20.83, 18.35, 18.15,
18.14, 18.12 (2 C), 16.39, 13.56, 12.99, 12.39, 11.62, 11.28, 9.39, ꢀ3.95,
ꢀ4.28, ꢀ4.43, ꢀ4.45, ꢀ4.77, ꢀ4.97, ꢀ5.01, ꢀ5.14 ppm. Macrolactone 20:
1
0.51 (hexane/ethyl acetate 10:1); [a]2D3 = 9.3 (c=0.87 in CHCl3); H NMR
(400 MHz, CDCl3): d=7.28–7.24 (m, 2H), 6.89–6.85 (m, 2H), 5.92 (ddt,
J=17.1, 10.3, 6.0 Hz, 1H), 5.35 (dm, J=17.1 Hz, 1H), 5.25 (dm, J=
10.3 Hz, 1H), 5.15 (d, J=9.7 Hz, 1H), 4.64 (ddt, J=13.0, 6.0, 1.3 Hz,
1H), 4.58 (ddt, J=13.0, 6.0, 1.3 Hz, 1H), 4.52 (d, J=11.2 Hz, 1H), 4.34
(d, J=11.2 Hz, 1H), 4.20 (d, J=5.5 Hz, 1H), 4.17 (d, J=8.0 Hz, 1H),
3.95 (dd, J=5.4, 2.8 Hz, 1H), 3.80 (s, 3H), 3.38 (s, 3H), 3.18 (dq, J=7.9,
6.1 Hz, 1H), 2.97 (dq, J=7.8, 6.9 Hz, 1H), 2.65 (dq-like, J=2.7, 7.3 Hz,
1H), 2.41–2.30 (m, 1H), 1.56 (d, J=1.2 Hz, 3H), 1.21 (d, J=7.3 Hz, 3H),
1.14 (d, J=6.9 Hz, 3H), 1.04 (d, J=6.1 Hz, 3H), 0.98 (d, J=6.2 Hz, 3H),
0.91 (s, 9H), 0.87 (s, 9H), 0.08 (s, 3H), 0.05 (s, 3H), 0.03 (s, 3H),
ꢀ0.03 ppm (s, 3H); 13C NMR (100 MHz, CDCl3): d=213.81, 171.26,
159.01, 134.58, 131.67, 131.66, 131.13, 129.20, 119.13, 113.69, 113.68, 80.03,
79.67, 78.86, 75.37, 70.67, 65.60, 60.19, 55.25, 48.31, 47.58, 39.00, 26.96,
25.83, 25.78, 18.34, 18.16, 17.80, 16.68, 14.38, 11.87, 10.67, ꢀ4.42, ꢀ4.93,
ꢀ5.01, ꢀ5.23 ppm; HRMS (ESI): m/z: calcd for C40H74NO8Si2 [M+NH4]+:
752.4953, found: 752.4969.
1) Deprotection of allyl ester: To a stirred solution of 19 (7.0 mg,
5.7 mmol) and [PdACHTNUGTRENU(NG Ph3)2Cl2] (1 mg) in CH2Cl2 (2 mL) was added Bu3SnH
TBS-protected aldol product 18: 2,6-Lutidine (23 mL, 0.20 mmol) was
added to a solution of 17 (37 mg, 0.048 mmol) in CH2Cl2 (1.5 mL) at 08C.
Then TBSOTf (30 mL, 0.13 mmol) was added dropwise. After 5 h, the re-
action was quenched with methanol (10 mL). The solution was diluted
with MTBE (30 mL) and was washed with sat. aq. NaHCO3 (2 mL) and
brine (2 mL). The aqueous layer was extracted again with MTBE
(6.0 mL, 22.8 mmol) at room temperature. After 20 min, TLC showed
completed conversion. The solution was diluted with CH2Cl2 (20 mL),
and was washed with sat. aq. NaHCO3 (5 mL), sat. aq. NH4Cl (5 mL)
and brine. The aqueous layer was extracted again with CH2Cl2 (10 mL).
The combined organic layers were dried over MgSO4 and concentrated
4950
www.chemeurj.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 4946 – 4952

An Improved Route to (+)-Tedanolide
FULL PAPER
in vacuo to give the crude carboxylic acid which was used directly for
macrolactonization without purification.
In vivo translation inhibition assay: 100 mL of 106 KB-3-1 cells were
seeded in a 96 well plate with DMEM media. After 16 h, the media was
removed, cells were incubated with fresh media for one hour. Then, the
cells were transfected with pRLSV40 vector (Promega) using PEI as
a transfection agent (DNA/PEI 1:2). After 48 h cells were treated with
des-epoxy-tedanolide for 6 h. End of which, the medium was removed
and cells were lysed with 30 mL of 1X Renilla Luciferase Assay Lysis
Buffer (Promega). 10 mL of the lysate was mixed with 10 mL of 1X Renil-
la Luciferase Assay Substrate (Promega) and the luminescence was mea-
sured.
2) Macrolatonization: To a stirred solution of PPh3 (15 mg, 57 mmol) in
dry toluene (4 mL) was added DEAD (9 mL, 57 mmol). After 30 min, a so-
lution of the crude carboxylic acid in toluene (1 mL) was added dropwise.
No carboxylic acid was observed on TLC after 15 min. The reaction mix-
ture was directly submitted to flash chromatography (hexane/ethyl ace-
tate 60:1 ! 50:1) without additional workup to give macrolactone 20 as
a colourless oil (5.1 mg, 76%). Rf =0.31 (hexane/ethyl acetate 25:1);
[a]2D3 =47.3 (c=0.95 inn CHCl3); 1H NMR (400 MHz, CDCl3): d=7.24
(dm, J=8.7 Hz, 2H), 6.87 (dm, J=8.7 Hz, 2H), 5.29 (dq-like, J=10.6,
6.7 Hz, 1H), 5.19 (dd, J=9.5, 1.1 Hz, 1H), 5.18–5.12 (m, 1H), 5.08 (d, J=
8.8 Hz, 1H), 4.52 (d, J=10.7 Hz, 1H), 4.43–4.38 (m, 1H), 4.40 (d, J=
7.6 Hz, 1H), 4.23 (d, J=10.6 Hz, 1H), 4.09 (d, J=9.7 Hz, 1H), 4.03 (d,
J=4.5 Hz, 2H), 3.98 (d-like, J=6.7 Hz, 2H), 3.84 (t, J=4.6 Hz, 1H), 3.81
(s, 3H), 3.36 (s, 3H), 3.33–3.30 (m, 1H), 3.34–3.28 (m, 1H), 3.14 (dq, J=
10.0, 6.7 Hz, 1H), 3.06 (dq, J=9.7, 7.0 Hz, 1H), 2.72 (dq, J=7.3, 4.7 Hz,
1H), 2.57–2.48 (m, 2H), 4.54–2.25 (m, 1H), 2.49–2.43 (m, 1H), 1.61 (d,
J=1.1 Hz, 3H), 1.58 (dd, J=6.7, 1.7 Hz, 3H), 1.57 (br, 3H), 1.28 (d, J=
7.4 Hz, 3H), 1.20 (d, J=6.9 Hz, 3H), 1.08 (d, J=6.7 Hz, 3H), 0.94 (d, 3H
overlapped with TBS), 0.88 (d, J=7.3 Hz, 3H), 0.94 (s, 9H), 0.93 (s, 9H),
0.87 (s, 9H), 0.83 (s, 9H), 0.13 (s, 3H), 0.12 (s, 3H), 0.10 (s, 3H), 0.09 (s,
3H), 0.07 (s, 3H), 0.05 (s, 3H), ꢀ0.03 (s, 3H), ꢀ0.04 (s, 3H), ꢀ0.08 ppm
(s, 3H); 13C NMR (100 MHz, CDCl3): d =213.69, 209.07, 171.41, 158.85,
137.97, 134.82, 133.30, 131.11, 128.68, 128.10, 121.92, 113.56, 80.94, 80.58,
78.74, 77.21, 75.92, 73.90, 71.04, 64.50, 60.75, 55.24, 49.00, 47.27, 46.76,
46.50, 42.53, 30.28, 29.69, 26.30, 26.01, 25.95, 25.74, 18.59, 18.32, 18.18,
18.11, 15.36, 14.96, 12.98, 12.10 (2 ꢄ CH3), 11.99, 10.62, ꢀ4.12, ꢀ4.22,
ꢀ4.39, ꢀ4.40, ꢀ4.49, ꢀ4.71, ꢀ4.83, ꢀ5.04 ppm; HRMS (ESI): m/z: calcd
for C64H116O11Si4Na [M+Na]+: 1195.7492, found: 1195.7504.
Immunofluorescence: PtK2 (ATCC CCL-56) cells were grown on glass
coverslips (13 mm diameter) in 4-well plates O/N, and incubated with
50 ngmLꢀ1 des-epoxy-tedanolide for 4 fours. Then the cells were fixed
with formaldehyde (3.7%) for 10 min, labelled with mouse anti-eIF2a
antibody (Santa Cruz Biotech) and rabbit anti-p-eIF2a antibody (Santa
Cruz Biotech) for one hour, and stained with anti-mouse Alexa fluor
A594 and anti-rabbit Alexa fluor A488 antibodies. Cells were rinsed with
PBS, mounted in Prolong Antifade containing DAPI (Molecular Probes),
and viewed with a Zeiss Axiophot fluorescence microscope using appro-
priate filter sets.
Acknowledgements
We are grateful to the German Research Foundation (DFG) for financial
support (Ka 913/9-2).
[1] F. J. Schmitz, S. P. Gunasekera, G. Yalamanchili, M. B. Hossain, D.
Vanderhelm, J. Am. Chem. Soc. 1984, 106, 7251–7252.
[2] F. J. Schmitz, S. P. Gunasekera, M. B. Hossain, D. van der Helm, G.
Yalamanchili, US Patent Application, 1988, 87-7347.
[3] a) N. Fusetani, T. Sugawara, S. Matsunaga, H. Hirota, J. Org. Chem.
1991, 56, 4971–4974; b) K. H. Lee, S. Nishimura, S. Matsunaga, N.
Fusetani, H. Ichijo, S. Horinouchi, M. Yoshida, Biosci. Biotechnol.
Biochem. 2006, 70, 161–171; c) S. Nishimura, S. Matsunaga, M.
Yoshida, H. Hirota, S. Yokoyama, N. Fusetani, Bioorg. Med. Chem.
2005, 13, 449–454; d) S. Nishimura, S. Matsunaga, S. Yoshida, Y.
Nakao, H. Hirota, N. Fusetani, Bioorg. Med. Chem. 2005, 13, 455–
462.
15-Hydroxy macrolactone and the hemiketal: To a stirred solution of
PMB-protected macrolactone 20 (7.0 mg, 6.0 mmol) in CH2Cl2 (0.8 mL)
and phosphate buffer (pH 7.0, 0.1 mL) at 08C was added a DDQ solution
in CH2Cl2 (7.2ꢄ10ꢀ3 m, 0.1 mL). After 5 h the mixture was diluted with
MTBE (40 mL), and was washed with sat. aq. NaHCO3 (5 mL) and brine
(5 mL). The aqueous layer was extracted with MTBE (20 mL). The com-
bined organic layers were dried over MgSO4 and concentrated under
vacuo. The residue was purified by flash chromatography (hexane/ethyl
acetate 60:1) to give the desired 15-hydroxy macrolactone together with
the hemiketal as inseparable mixtures (4.5 mg, 71%) as a colourless oil.
Rf =0.30 (hexane/ethyl acetate 30: 1); HRMS (ESI): m/z: calcd for
C56H108O10Si4Na [M+Na]+: 1075.6917, found: 1075.7007.
[4] C. Chevallier, T. S. Bugni, X. D. Feng, M. K. Harper, A. M. Orendt,
C. M. Ireland, J. Org. Chem. 2006, 71, 2510–2513.
Cell cultures and inhibition assays: Cell lines were from the German Col-
lection of Microorganisms and Cell Cultures (DSMZ) or the American
Type Culture Collection (ATCC). Antiproliferative activity was mea-
sured in 96-well plates. 60 mL of serial dilutions of the inhibitor were in-
cubated with 120 mL of suspended cells (50000 cells per mL; two repli-
cates). After five days, the metabolic activity in each well was determined
[5] a) T. L. Meragelman, R. H. Willis, G. M. Woldemichael, A. Heaton,
P. T. Murphy, K. M. Snader, D. J. Newman, R. van Soest, M. R.
Boyd, J. H. Cardellina, T. C. McKee, J. Nat. Prod. 2007, 70, 1133–
1138; b) E. L. Whitson, K. M. Pluchino, M. D. Hall, J. B. McMahon,
T. C. McKee, Org. Lett. 2011, 13, 3518–3521.
[6] a) A. B. Smith, C. M. Adams, S. A. L. Barbosa, A. P. Degnan, Proc.
Natl. Acad. Sci. USA 2004, 101, 12042–12047; b) A. B. Smith, C. M.
Adams, S. A. L. Barbosa, A. P. Degnan, J. Am. Chem. Soc. 2003,
125, 350–351; c) A. B. Smith, D. J. Lee, J. Am. Chem. Soc. 2007, 129,
10957–10962.
[7] a) J. R. Dunetz, L. D. Julian, J. S. Newcom, W. R. Roush, J. Am.
Chem. Soc. 2008, 130, 16407–16416; b) L. D. Julian, J. S. Newcom,
W. R. Roush, J. Am. Chem. Soc. 2005, 127, 6186–6187.
[8] a) G. Ehrlich, J. Hassfeld, U. Eggert, M. Kalesse, J. Am. Chem. Soc.
2006, 128, 14038–14039; b) G. Ehrlich, J. Hassfeld, U. Eggert, M.
Kalesse, Chem. Eur. J. 2008, 14, 2232–2247.
[9] a) M. Christmann, U. Bhatt, M. Quitschalle, E. Claus, M. Kalesse,
Angew. Chem. 2000, 112, 4535–4538; Angew. Chem. Int. Ed. 2000,
39, 4364–4366; b) M. Christmann, M. Kalesse, Tetrahedron Lett.
2001, 42, 1269–1271; c) J. Hassfeld, M. Christmann, M. Kalesse,
Org. Lett. 2001, 3, 3561–3564; d) J. Hassfeld, M. Kalesse, Synlett
2002, 2007–2010; e) J. Hassfeld, M. Kalesse, Tetrahedron Lett. 2002,
43, 5093–5095.
using
(MTT) assay.
a
3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide
FACS analysis: BD Biosciences apoptosis detection kit and protocol was
used. In short, cultured A-431 cells were treated with 50 ngmLꢀ1 of des-
epoxy-tedanolide for 24 h. The cells were then trypsinized, centrifuged at
1500 rpm for 5 min. The cell pellet was washed twice with ice cold PBS
and resuspended in 1X Binding buffer to a concentration of 1ꢄ106
cellsmLꢀ1. 100 mL of this mix was incubated with 5 mL annexin V and
5 mL of PI, mixed well and incubated in the dark for 15 min. Then the
volume was made up to 500 mL with 1X Binding buffer and analysed by
flow cytometry within 1 h. MeOH treated cells served as control.
Translation inhibition assays
In vitro translation inhibition: The assay was performed using Flexi
Rabbit Reticulocyte Lysate System (Promega). In brief, 1 mL of des-
epoxy-tedanolide solution was mixed with a 25 mL reaction mixture con-
taining 100 ng of luciferase mRNA, 20 mL of rabbit reticulocyte lysate,
0.2 mL of each Leu and Met amino acid mixture (1 mm), 0.5 mL of 2.5m
KCl, 10 U of RNasin inhibitor and nuclease free water. The reaction mix
was incubated at 308C for 90 min. Then 2 mL of the lysate was mixed
with 10 mL of the assay buffer and the luminescence was measured.
[10] K. B. Sharpless, W. Amberg, Y. L. Bennani, G. A. Crispino, J. Har-
tung, K. S. Jeong, H. L. Kwong, K. Morikawa, Z. M. Wang, D. Q.
Xu, X. L. Zhang, J. Org. Chem. 1992, 57, 2768–2771.
Chem. Eur. J. 2012, 18, 4946 – 4952
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4951

M. Kalesse et al.
[11] D. B. Dess, J. C. Martin, J. Org. Chem. 1983, 48, 4155–4156.
[12] S. V. Ley, J. Norman, W. P. Griffith, S. P. Marsden, Synthesis 1994,
639–666.
[15] D. Trisciuoglio, B. Uranchimeg, J. H. Cardellina, T. L. Meragelman,
S. Matsunaga, N. Fusetani, D. Del Bufalo, R. H. Shoemaker, G. Me-
lillo, J. Natl. Med. Assoc. J. Nat. Cancer Inst. 2008, 100, 1233–1246.
[13] N. J. Leonard, Neelima, Tetrahedron Lett. 1995, 36, 7833–7836.
[14] S. J. Schroeder, G. Blaha, J. Tirado-Rives, T. A. Steitz, P. B. Moore,
J. Mol. Biol. 2007, 367, 1471–1479.
Received: September 28, 2011
Published online: March 5, 2012
4952
www.chemeurj.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 4946 – 4952