Epothilone analogues with benzimidazole and quinoline side chains: Chemical synthesis, antiproliferative activity, and interactions with tubulin

Article abstract of DOI:10.1002/chem.200901376

A series of epothilone B and D analogues bearing isomeric quinoline or functionalized benzimidazole side chains has been prepared by chemical synthesis in a highly convergent manner. All analogues have been found to interact with the tubulin/microtubule s

Full text of DOI:10.1002/chem.200901376

DOI: 10.1002/chem.200901376
Epothilone Analogues with Benzimidazole and Quinoline Side Chains:
Chemical Synthesis, Antiproliferative Activity, and Interactions with Tubulin
Silvia Anthoine Dietrich,[a] Renate Lindauer,[a] Claire Stierlin,[a] Jꢀrg Gertsch,[a]
Ruth Matesanz,[b] Sara Notararigo,[b] Josꢁ Fernando Dꢂaz,[b] and Karl-Heinz Altmann*[a]
Abstract: A series of epothilone B and
D analogues bearing isomeric quino-
line or functionalized benzimidazole
side chains has been prepared by
chemical synthesis in a highly conver-
gent manner. All analogues have been
found to interact with the tubulin/mi-
crotubule system and to inhibit human
cancer cell proliferation in vitro, albeit
with different potencies (IC50 values
between 1 and 150 nm). The affinity of
quinoline-based epothilone B and D
analogues for stabilized microtubules
clearly depends on the position of the
N-atom in the quinoline system, while
the induction of tubulin polymerization
in vitro appears to be less sensitive to
N-positioning. The potent inhibition of
human cancer cell growth by epothi-
lone analogues bearing functionalized
benzimidazole side chains suggests that
these systems might be conjugated with
tumor-targeting moieties to form
tumor-targeted prodrugs.
Keywords:
epothilone
antitumor agents
microtubules
·
·
·
natural products · structure–activity
relationships · total synthesis
Introduction
through the same mechanism of action as the established
clinical anticancer drugs taxol (paclitaxel; Taxol) and doce-
taxel (Taxotere), the interference of which with MT func-
tionality is associated with cell cycle arrest in mitosis and
the induction of apoptosis.[4] In contrast to taxol, however,
epothilones are also active against various types of multi-
drug-resistant cell lines that overexpress the P-gp170 efflux
pump (for which Epo A and B are very poor substrates)[3,5]
or that have acquired taxol-resistant tubulin mutations.[6]
The discovery of their “taxol-like” mechanism of action
led to an instantaneous surge of interest in the epothilones
as important new lead structures for anticancer drug discov-
ery[7] and as such they also became highly relevant targets
for total synthesis.[8] Over the last 15 years, research into the
structure–activity relationships (SARs) for epothilones has
led to an exceptionally comprehensive knowledge base on
the activity of structurally modified epothilone analogues
(fully synthetic variants as well as semi-synthetic deriva-
tives), ranging from modifications that lead to improved an-
tiproliferative activity over the natural product leads to
those that are completely detrimental to biological activity.[7]
Most importantly, these efforts have also produced several
compounds that have entered clinical evaluation in humans
and are currently at different stages of clinical develop-
ment.[9] In fact, the most advanced of these compounds, the
Epo B lactam BMS-247550 (ixabepilone) was approved in
2007 by the US FDA for clinical use in breast cancer pa-
tients (under the trade name Ixempra).[10]
Epothilones comprise a family of bacterial natural products,
the major representatives of which, epothilones A and B
(Epo A and Epo B; Figure 1), were first isolated in 1986 by
Hçfle, Reichenbach, and co-workers from the cellulose-de-
grading myxobacterium Sorangium cellulosum.[1,2] These
compounds were quickly recognized to exhibit potent anti-
proliferative activity against human cancer cells in vitro,[1]
but it was not before 1995 that these growth inhibitory ef-
fects were demonstrated by Bollag et al. to arise from the
ability of the epothilones to stabilize cellular microtubules
(MTs).[3] Thus, epothilones inhibit human cancer cell growth
[a] S. A. Dietrich, R. Lindauer, C. Stierlin, Dr. J. Gertsch,
Prof. Dr. K.-H. Altmann
Swiss Federal Institute of Technology (ETH) Zꢀrich
Department of Chemistry and Applied Biosciences
Institute of Pharmaceutical Sciences, HCI H405
Wolfgang-Pauli-Str. 10, 8093 Zꢀrich (Switzerland)
Fax : (+41)44-6331369
E-mail: karl-heinz.altmann@pharma.ethz.ch
Homepage: http://www.pharma.ethz.ch/institute_groups/
institute_groups/pharmaceutical_biology/
[b] Dr. R. Matesanz, S. Notararigo, Dr. J. F. Dꢁaz
Centro de Investigaciones Biolꢂgicas
Consejo Superior de Investigaciones Cientꢁficas
Ramiro de Maeztu 9, 28040 Madrid (Spain)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200901376.
10144
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 10144 – 10157

FULL PAPER
cle has a significant influence on tubulin (MT) binding.[14]
As an extension of this earlier work by Nicolaou et al., we
have recently communicated the synthesis and biological
evaluation of quinoline-based Epo B/D analogues 1 and 2
(Figure 1).[15] In accordance with Nicolaouꢅs earlier data, an-
alogue 1b was found to be a significantly more potent anti-
proliferative agent than its isomer 2b. At the same time, a
preliminary assessment of the effects of 1b and 2b on tubu-
lin assembly revealed similar EC50 values for the induction
of tubulin polymerization in vitro, which suggested that the
difference in cellular activity between 1b and 2b might not
arise from differences in their interactions with the tubulin/
MT system. No firm conclusions can be drawn from the
polymerization data, however, as compounds with signifi-
cantly different MT binding affinities may still exhibit simi-
lar EC50 values for the induction of tubulin polymerization,
if they both exceed a certain affinity threshold. In contrast
to 1b and 2b, the corresponding epoxides 1a and 2a
(Figure 1) were found to be equally potent inhibitors of
cancer cell growth in vitro and they showed similar EC50
values for tubulin polymerization induction.
Figure 1.
However, in spite of the vast body of SAR information
available for epothilones today, our understanding of their
detailed molecular interactions with their target protein b-
tubulin is still limited by the lack of high-resolution structur-
al data for complexes between tubulin/MTs and any epothi-
lone analogue (or any other type of MT stabilizer for that
matter). While the structure of a complex of Epo A with b-
tubulin in Zn2+-stabilized tubulin polymer sheets has been
determined by a combination of electron crystallography
(EC) at 2.89 ꢄ resolution and NMR-based conformational
analysis,[11] the tubulin-bound conformation of Epo A de-
duced from this analysis does not permit rationalization of a
number of important features of the epothilone SAR. A
more satisfactory explanation of the SAR data is provided
by the structural model proposed by Carlomagno and co-
workers on the basis of solution NMR studies on tubulin-
bound Epo A.[12,13] This NMR-based model deviates from
the EC-derived structure of the tubulin/Epo A complex in
several respects, including the type of interaction between
the side-chain thiazole moiety in epothilones and the side
chain of His-227 in b-tubulin. The latter has been concluded
to be hydrogen-bonded (in its protonated form) to the N-
atom of the thiazole ring based on the EC data;[11] in con-
trast, no such H-bond was found in the solution NMR stud-
ies, but it is suggested that the thiazole ring is involved in a
stacking interaction with the imidazole ring of the His-227
side chain.[13]
In this paper, we disclose full details of the synthesis of
Epo B analogues 1a/b and 2a/b. In addition, we report on
an extension of our previous studies on the interactions of
these analogues with the tubulin/MT system to the investi-
gation of their binding affinities to cross-linked MTs, in
order to provide a direct measure of the stability of the cor-
responding protein/ligand complexes. Finally, we also de-
scribe the synthesis and preliminary biological assessment of
the new epothilone analogues 3 and 4, which belong to the
same family of structures as 1a/b and 2a/b. Conformational-
ly constrained analogues of this type exhibit enhanced cellu-
lar potency, especially in the Epo D[16] series (e.g., 1b and
2b), and of the various side chains investigated so far the ac-
tivity-enhancing effect is clearly greatest for a dimethylben-
zimidazole moiety.[17–19] Based on these previous findings, we
have designed analogues 3 and 4, which incorporate addi-
tional functional groups on the benzimidazole 3-substituent
that are envisaged to be utilized for the reversible attach-
ment of different types of tumor-targeting moieties. In a
first step to determining the potential of these analogues to
serve as active drug moieties in tumor-targeted prodrugs, we
have assessed their intrinsic biological activities.
Results and Discussion
SAR studies on the importance of N-positioning in pyri-
dine-based analogues of Epo B (with different isomeric pyri-
dine moieties in place of the natural thiazole ring) have
shown that antiproliferative activity similar to that of Epo B
itself is only conserved if the pyridine nitrogen is located
ortho to the attachment point of the vinyl linker between
the heterocycle and the macrolactone ring.[14] Together with
data on the effect of these compounds on tubulin polymeri-
zation (which are, however, not fully quantitative in nature),
this led to the conclusion that N-positioning in the heterocy-
Chemistry: Our general retrosynthesis of analogues 1–4 is
depicted in Scheme 1. In all cases, ring-closure was envis-
aged to be achieved through Yamaguchi-type macrolactoni-
zation of a suitably protected seco acid,[20] which was to be
followed by removal of the protecting group and, in the
cases of 1a and 2a, by epoxidation of the C12–C13 double
bond. The requisite seco acid was to be obtained from the
C1–C11 fragment I-1 and vinyl iodides I-2 by Suzuki–
Miyaura coupling, a strategy that we had successfully uti-
lized in our previous syntheses of trans-Epo A[21] and related
Chem. Eur. J. 2009, 15, 10144 – 10157
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10145

K.-H. Altmann et al.
Scheme 2. a) SeO2 (0.66 equiv), 1558C, 18 h, 58%. b) i) Et3B (1.4 equiv),
CF3SO3H (1.3 equiv), CH2Cl2, RT, 25 min; ii) + Ac-Aux* (8) (1 equiv),
DIEA (1.3 equiv), CH2Cl2, 08C, 20 min; iii) +7 (1.4 equiv), CH2Cl2,
ꢀ788C, 3.5 h, 80% (mixture of diastereoisomers). c) TBS-Cl (1.5 equiv),
imidazole (3 equiv), DMF, 408C, 17 h, 63% (single isomer). d) DIBAL-H
(2.5 equiv), CH2Cl2, ꢀ788C, 3.5 h, 78%. e) i) [Ph3PCH
ACHTUNGTRENNUNG(CH3)I]I
(1.3 equiv), NaHMDS (1.2 equiv), THF, ꢀ788C, 45 min, ꢀ158C, 20 min;
ii) + 11, THF, ꢀ788C, 45 min, 41%.
Scheme 1. Retrosynthesis of target structures 1–4. X, Y, R=CH=CH, N,
(1); N, CH=CH, (2); N(CH2CH2NH2), N, CH3 (3);
H
H
N(CH2CH2NHC(O)CH2CH2COOH), N, CH3 (4). PG=protecting group
or H. Protecting groups could be varied independently.
methylquinoline by oxidation with SeO2 in 59% yield)
(Scheme 3). In keeping with our own experience and previ-
ous literature reports (see above), the lowest yielding step
of the sequence was once again the formation of the vinyl
iodide (17), which was obtained from aldehyde 16 in 30%
yield under virtually identical conditions to those employed
for the transformation of 11 to 12.
analogues[22,23] and that was first exploited for the synthesis
of epothilones by Danishefsky and co-workers.[24]
While we have previously described the synthesis of ester
5 (i.e., intermediate I-1 with PG=TBS),[21] vinyl iodides I-2
were envisaged to be prepared aldehydes I-3 through Zhao
iodo-olefination.[25] The latter would be obtained from alde-
hydes I-4 by stereoselective aldol or allylation reactions.
Scheme 2 summarizes the synthesis of vinyl iodide 12 as
the requisite C13–C15 building block for the elaboration of
analogues 1a and 1b. Oxidation of commercially available
7-methylquinoline (6) with SeO2 gave quinoline-7-carboxal-
dehyde (7),[26] which was then submitted to aldol reaction
with the boron-enolate of the acetylated Oppolzer sultam
8.[27,28] The aldol product 9 was obtained in 80% yield as a
5:1 mixture of isomers, which were not separated, but di-
rectly converted to the corresponding C15-O-TBS ethers
(epothilone numbering). At this stage, the isomers were
easily separable and the desired product 10 could be isolat-
ed in 50% overall yield for the two-step sequence from al-
dehyde 7.[29] While the conversion of 10 to aldehyde 11 with
DIBAL-H was straightforward, the subsequent Zhao olefi-
nation[25] provided the desired vinyl iodide 12 only in moder-
ate (but still acceptable) yield. Yields not exceeding 50%
have been consistently reported in the literature for this
type of iodo-olefination reaction for a variety of substrates
and appear to be an intrinsic feature of this transforma-
tion.[30]
Scheme 3. a) i) Et3B (1.4 equiv), CF3SO3H (1.3 equiv), CH2Cl2, RT,
25 min; ii) +Ac-Aux* (8) (1.3 equiv), DIEA (1.3 equiv), CH2Cl2, 08C,
20 min; iii) +13 (1 equiv), CH2Cl2, ꢀ788C, 3.5 h, 81% (mixture of diaste-
reoisomers). b) TES-Cl (1.5 equiv), imidazole (3 equiv), DMF, 458C,
18 h, 37% (2 steps, single isomer). c) DIBAL-H (4 equiv), CH2Cl2,
ꢀ788C, 5.5 h, 74%. d) i) [Ph3PCH
ACHTUNGRTEN(NUNG CH3)I]I (1.3 equiv), NaHMDS
(1.2 equiv), THF, ꢀ788C, 65 min, ꢀ158C, 20 min; ii) +16, THF, ꢀ788C,
45 min, 30%.
Compared to the synthesis of quinoline-derived building
blocks 12 and 17, the preparation of benzimidazole-contain-
ing vinyl iodide 28 was significantly more elaborate, as in
this case the required aldehyde I-4 (Scheme 1) could not be
simply obtained in one step from a commercially available
starting material. As illustrated in Scheme 4, this aldehyde
(i.e., 24 in Scheme 4) had to be prepared in a multistep se-
quence from 4-fluoro-3-nitrobenzoic acid (18), starting with
the conversion of the latter to its methyl ester 19
(Scheme 4). Nucleophilic displacement of the aromatic
Following an analogous sequence of reactions to that de-
scribed above, the isomeric vinyl iodide 17 was prepared
from quinoline-6-carboxaldehyde (13) (obtained from 6-
10146
www.chemeurj.org
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 10144 – 10157

Biological Evaluations of New Epothilone Analogues
FULL PAPER
As shown in Scheme 5, Suzuki–Miyaura coupling of the
borane derived in situ from alkene 5 and 9-BBN with vinyl
iodide 28 produced the fully protected seco acid 29 in excel-
Scheme 4. a) MeOH, H2SO4, reflux, 98%. b) BocNHCH2CH2NH2, Et3N,
CH2Cl2, RT, 96%. c) 1 atm H2, Pd/C, MeOH, RT, 99%. d) (EtO)3CCH3
(6 equiv), EtOH, reflux, 96%. e) DIBAL-H, CH2Cl2, ꢀ788C!RT, 78 %.
f) MnO2 (10 equiv), CHCl3, reflux, 1 h, 98%. g) i) C3H5MgBr (1.5 equiv),
(ꢀ)-Ipc2BCl (1.5 equiv), Et2O, 0 8C!RT, 1 h (solution A); ii. slow addi-
tion of solution A to a solution of 24 (1 equiv) in Et2O, ꢀ1008C, 3 h,
89%, 94% ee. h) TES-Cl, imidazole, DMAP, RT, 4 h, 98%. i) OsO4,
NaIO4, 2,6-lutidine, dioxane/water/tBuOH, RT, 23 h, 74%. j) i) [Ph3PCH-
Scheme 5. a) i) 5, 9-BBN, THF, RT; ii) Cs2CO3, [PdCl2ACHTNUTRGNE(NUG dppf)2], Ph3As, 28,
ꢀ58C!RT, 83%. b) LiOH, dioxane/water 4:1, 608C, 11.5 h, acidic work-
up, 85%. c) i) Et3N, 2,4,6-trichlorobenzoyl chloride, THF, 108C;
ii) DMAP, toluene, RT, 57%. d) CF3COOH, CH2Cl2, RT, 2 h, 44%.
e) ZnBr2, CH2Cl2, 08C, 72 h, 57%. f) Succinic anhydride, DIEA, DMF,
RT, 2 h, 92% (crude). g) CF3COOH, CH2Cl2, RT, 5 h, 23%.
(CH3)I]I (1.3 equiv), NaHMDS (1.2 equiv), THF, ꢀ788C, 45 min, ꢀ158C,
20 min; ii) +27, THF, ꢀ788C, 45 min, 42%.
fluoro substituent in 19 with mono-Boc-protected ethylene-
diamine gave nitroaniline 20, which was reduced with H2/
Pd-C to provide the phenylenediamine derivative 21 in es-
sentially quantitative yield. The latter was cyclized to benzi-
midazole 22, the ester group of which was reduced with
DIBAL-H followed by oxidation of the resulting alcohol 23
with MnO2. Aldehyde 24 was thus obtained from 18 in ex-
cellent overall yield (70% over eight steps). While 24 could
be successfully elaborated into the desired vinyl iodide 28
following the aldol-based approach that had been developed
for the synthesis of 12 and 17, the selectivity of the aldol re-
action in the case of aldehyde 24 proved to be very moder-
ate (2:1 vs. 5:1 and 3:1 for aldehydes 11 and 15, respective-
ly; see above), thus leading to unsatisfactory overall yields
for the elaboration of 24 into 28 (not shown). As a conse-
quence, an alternative approach for the stereoselective two-
carbon extension of 24 was developed, which was based on
lent yield (83%). The conditions for this coupling had been
optimized during our previous work on trans-Epo A and re-
lated analogues,[21–23] but it is worth emphasizing that the
success of the reaction critically depends on the exclusion of
moisture in the hydroboration step and thus requires careful
and extensive drying of ester 5. The coupling product 29 was
directly converted to the immediate cyclization precursor 30
through ester saponification with LiOH and exposure of the
resulting C15-O-TES derivative to aqueous acid during
work-up in 85% yield. Seco acid 30 was then cyclized at RT
according to the Yamaguchi protocol[20] to provide the fully
protected macrolactone 31 in 57% yield. Finally, treatment
of 31 with TFA gave target structure 3 in 44% yield after
HPLC purification. Alternatively, the Boc group in 31 could
be selectively removed by treatment with ZnBr2 in
[33]
CH2Cl2 and the resulting free amine was reacted with suc-
Brown allylation of 24 with (ꢀ)-Ipc2B
G
cinic anhydride to give 33, which was converted to target
structure 4 by deprotection with TFA. The latter structure 4
was obtained in 21% overall yield for the two-step sequence
from amine 32. The low yield in the transformation of 31 to
4 is largely a consequence of the difficulties associated with
the purification of the very polar products.
from 24.[31,32] Reaction of 27 with Ph3P=C
ACTHNUTRGEN(NUG CH3)I finally pro-
vided vinyl iodide 28; as for 12 and 17, the latter could be
obtained only in moderate yield (42%).
As for the reaction of 5 with vinyl iodide 28, the corre-
sponding Suzuki–Miyaura couplings with 12 and 17 proceed-
Chem. Eur. J. 2009, 15, 10144 – 10157
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10147

K.-H. Altmann et al.
ed with excellent efficiencies and provided the respective
coupling products 34 and 38 in yields of 91 and 94%, re-
spectively (Scheme 6).
products in the macrolactonization of 30 in this study, but it
should be noted that epothilone analogues incorporating
2,3-dimethylbenzimidazole-derived side chains have been
obtained in yields comparable to those for quinoline-bearing
macrocycles 37 and 40 (for the macrolactonization
step).[22,23]
The Epo D analogues 1b and 2b were converted to the
corresponding epoxide-based Epo B analogues 1a and 2a
using the ReO3/pyridine/H2O2 system developed by Sharp-
less[34] and Hermann.[35] In both cases, the epoxidation pro-
duced the desired epoxide isomer with about 6:1 selectivity,
but it was also accompanied by N-oxidation of the quinoline
side chain (which was, in fact, faster than the epoxidation of
the C12/C13 double bond). However, conditions could be
identified that allowed selective reduction of the N-oxide
without affecting the epoxide moiety (and, in particular,
without reductive cleavage of the benzylic ester moiety).
Thus, careful catalytic hydrogenation of the N-oxides of 1a
and 2a over Raney Ni gave the desired target compounds
1a and 2a, which were finally obtained in 39% yield in each
case for the two-step sequence from 37 and 40, respectively.
Antiproliferative activity: The effects of epothilone ana-
logues 1–4 on human cancer cell growth in vitro were inves-
tigated for three different cell lines and the corresponding
IC50 values are summarized in Table 1. Epoxide-containing
analogues 1a and 2a were both found to exhibit highly
potent antiproliferative activity, with IC50 values in the sub-
nm range against all three cell lines. Compared to Epo B,
the activities of these analogues are reduced only slightly
and they are more active than Epo A. Most significantly,
however, the growth inhibitory activities of 1a and 2a are
independent of the position of the N-atom in the quinoline
side chain. This finding clearly contrasts with the results of
previous studies on pyridine-based Epo B analogues, which
showed a 3-pyridyl derivative (corresponding to quinoline
derivative 2a) to be substantially less active than the corre-
sponding 2-isomer (corresponding to 1a).[14] In contrast to
epoxide-containing analogues 1a and 2a, a significant differ-
ence in potency was observed between the (side-chain) iso-
meric Epo D derivatives 1b and 2b. The antiproliferative
activity of 1b is, quite remarkably, almost comparable to
that of 1a, in spite of the lack of an epoxide moiety, thus
making this analogue significantly more potent than the cor-
responding parent compound Epo D (Table 1). Evidently,
the activity-enhancing effect of the 7-quinolyl moiety (rela-
tive to the natural side chain) is more pronounced in combi-
nation with the epoxide-free Epo D macrocycle than with
the epoxide-containing Epo B core structure; this is in line
with previous observations on the increase in antiprolifera-
tive activity associated with a 2,3-dimethylbenzimidazole
side chain, which was also found to be more pronounced in
the Epo D series than in the Epo B series.[22,23] Analogues
1a and 1b were also investigated in the highly P-gp-overex-
pressing, multidrug-resistant human cervix carcinoma cell
line KB-8511.[36] No loss in activity was observed against this
cell line compared to the drug-sensitive KB-31 parental line
Scheme 6. a) i) 5, 9-BBN, THF, RT; ii) Cs2CO3, [PdCl2ACHTNUTRGNE(UNG dppf)2], Ph3As, 12
or 17, ꢀ108C!RT, 91 % (34) and 94% (38). b) LiOH, iPrOH/water 4:1,
528C, 15 h, 85%. c) TBAF (3 equiv), THF, RT, 18 h, 82%. d) i) Et3N,
2,4,6-trichlorobenzoyl chloride, THF, 08C; ii) DMAP, toluene, 70–758C,
83% (37) and 75% (40). e) HF·pyridine, THF, RT, 22 h, 89% (1b) and
89% (2b). f) i) MeReO3, H2O2, pyridine, CH2Cl2, RT, 46 h; ii) Raney Ni,
MeOH, RT, 3 h, 39% (1a) and 39% (2a) (2 steps). g) LiOH, iPrOH/
H2O 4:1, 558C, 15 h, acidic work-up, 74%.
Fully protected seco acid 38, which bears a C15-O-TES
protecting group, was converted into Epo D analogue 2b
following the same sequence of transformations as described
above for the elaboration of 29 into 31. In particular, 38
could be converted into 40 in one step (74% yield) owing to
the selective cleavage of the C15-TES ether during work-up.
In contrast, the conversion of the coupling product 34 to the
macrolactonization substrate 36 involved the removal of the
C15-O-TBS group in a discrete step (after ester saponifica-
tion); 36 was obtained in 70% overall yield from 34. Com-
pared to the cyclization of benzimidazole-containing seco
acid 30, macrolactonization was more efficient for the quin-
oline-derived intermediates 36 and 39, which could be cy-
clized in yields of 83 and 75%, respectively (compared to
57% for 30). We have not investigated the formation of side
10148
www.chemeurj.org
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 10144 – 10157

Biological Evaluations of New Epothilone Analogues
FULL PAPER
Table 1. Cancer cell growth inhibition by epothilone analogues 1–4 (IC50
[nm])[a]
moiety, the pKa of which in 3 and 4 is unknown) could lead
to changes in cellular uptake and/or intracellular distribution
of the compounds that could adversely affect their cellular
activity. These issues have not been investigated, but it
should be noted that, based on preliminary data in one cell
line, the N-Boc derivative of 3 (which was obtained from
fully protected macrolactone 31 (Scheme 5) by partial de-
protection with HF·pyridine) appears to be equally as
potent as 3. This finding would indicate that the free pri-
mary amino group in 3 does not have a negative impact on
cellular potency; in addition, it also suggests that the reduc-
tion in activity associated with the succinylation of 3 (3 !
4) is not simply a consequence of the increased size of the
substituent on the benzimidazole moiety.
The design of analogues 3 and 4 was driven by the idea of
providing functionalized epothilone analogues that could be
readily converted to tumor-targeted prodrugs by the attach-
ment of appropriate targeting moieties to the amino or car-
boxyl groups, respectively. In principle, tumor-targeted con-
jugates could also be prepared from unmodified natural
epothilones, utilizing one of the two hydroxyl groups at C3
and C7 as an anchoring point; however, the hindered nature
of these nucleophiles may render such an approach less than
straightforward. While we had anticipated 3 and 4 to be
somewhat more potent inhibitors of cancer cell growth than
was eventually observed experimentally, it should be empha-
sized that both compounds are still potent antiproliferative
agents and thus represent interesting candidates for the de-
velopment of tumor-targeted prodrugs. The synthesis of
such analogues is currently in progress in our laboratory and
the results of these efforts will be reported in future publica-
tions.
HCT-116 (colon)
A549 (lung)
MCF-7 (breast)
1a
1b
2a
2b
3
4
Epo A
Epo B
Epo D[c]
0.22ꢁ0.04
0.82ꢁ0.07
0.57ꢁ0.07
112ꢁ8
0.46ꢁ0.08
0.91ꢁ0.09
0.49ꢁ0.01
107ꢁ5
0.59ꢁ0.08
1.21ꢁ0.16
0.74ꢁ0.14
134ꢁ11
[b]
–
13.0ꢁ4.8
108ꢁ14
10.5ꢁ3.0
65ꢁ12
[b]
–
[b]
2.8ꢁ0.4
–
2.9ꢁ0.3
0.16ꢁ0.01
4.48ꢁ0.47
0.34ꢁ0.03
0.33ꢁ0.01
2.31ꢁ0.55
4.62ꢁ2.19
[a] For compound structures, see Figure 1. Cells were exposed to the test
compounds for 72 h. Cell numbers were estimated by quantification of
the protein content of fixed cells by methylene blue staining (cf. Support-
ing Information). Values shown are the means of three independent ex-
periments (ꢁstandard deviation). For compounds 3 and 4, the bis- and
mono-TFA salts were used, respectively. Data for Epo D are from
ref. [15]. [b] Not determined. [c] Epo D=12,13-deoxyEpo B; see also
ref. [16].
(IC50 values of 1a and 1b against the KB-31/KB-8511 lines
were 0.1 nm/0.1 nm and 0.59 nm/0.38 nm, respectively), thus
indicating that neither compound is a substrate for the P-gp
efflux pump.
Compared to 1b, the antiproliferative activity of its side-
chain isomer 2b is >100-fold lower, which is in line with ex-
pectation based on the data previously reported for pyri-
dine-based Epo B analogues[14] (see above). As will be
shown below, the difference in cellular activity between 1b
and 2b is paralleled by a significant difference in MT-bind-
ing affinity. However, a similar difference in MT binding
also exists for epoxide-containing analogues 1a and 2a and
this does not translate into significantly different growth in-
hibitory activities for the two compounds. In addition, the
>100-fold activity increase observed upon incorporation of
a C12,C13-epoxide moiety into 2b (2b ! 2a) is substantial-
ly higher than what is usually observed for transitions from
the Epo D to the Epo B series (5–30-fold; cf., e.g., the activ-
ities of Epo B and Epo D in Table 1).[7] Thus, the very
potent antiproliferative activity of 2a is very surprising and
it cannot be readily accommodated within the general SAR
landscape that has been delineated for epothilones over the
last few years.[7]
Benzimidazole-based analogues 3 and 4 exhibit IC50
values for cancer cell growth inhibition in the range 10–
100 nm, which makes them significantly less potent than the
corresponding 2,3-dimethylbenzimidazole-based Epo B ana-
logue (IC50 values in the sub-nm range have been observed
for the latter against a variety of human cancer cell lines in
vitro[17,37]). While these findings might be simply attributed
to the increased size of the benzimidazole 3-substituent in 3
and 4 (compared to a simple methyl group), it should be
noted that Nicolaou and co-workers have recently demon-
strated a significant tolerance to steric bulk in the north-
eastern quadrant of the epothilone structure,[38] which
should also apply to the modified side chains present in ana-
logues 3 and 4. Alternatively, the presence of ionizable
groups in 3 and 4 (possibly in addition to the benzimidazole
Interactions with tubulin/microtubules: The interactions of
epothilone analogues 1–4 with the tubulin/MT system were
first assessed through a tubulin polymerization assay, which
provides a measure of the ability of a compound to induce
the assembly of soluble ab-tubulin heterodimers into MT-
like polymers.[39,40] The extent of tubulin polymerization is
determined for different ligand concentrations at a fixed
concentration of soluble tubulin (in our case 10 mm), with
the concentration leading to 50% of the maximum polymer-
ization observed (EC50) serving as a comparator for the tu-
bulin-polymerizing capacity of different ligands. While this
approach allows the rapid identification of compounds that
are devoid of any MT-stabilizing properties, due to the often
small differences in EC50 values the unambiguous ranking of
agents with measurable tubulin-polymerizing activity is
more difficult. In addition, it has been demonstrated by
Hamel and co-workers that the relative rank order of tubu-
lin-polymerizing activity within a series of compounds may
vary depending on the exact experimental conditions em-
ployed in the polymerization experiments.[41]
Within these limitations, the EC50 values summarized in
Table 2 for epothilone analogues 1–4 suggest the following
rank order of tubulin-polymerizing activities: 1a ~ 1b >
2a ~ 3 ~ 4 > 2b. For 1a/b and 2b/3/4, this rank order is
Chem. Eur. J. 2009, 15, 10144 – 10157
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10149

K.-H. Altmann et al.
(qualitatively) consistent with the pronounced difference in
antiproliferative activities between the two groups of com-
pounds, while compound 2a might have been expected to
be a less active inhibitor of cancer cell growth than was ac-
tually observed (Table 1). It should be emphasized, however,
that differences between the EC50 values shown in Table 2
are small (with overlapping standard deviations for several
compound pairs) and should not be overinterpreted.
pacity suggest that the assembly reaction is promoted most
efficiently by analogues 1a and 1b, while analogue 2b is
clearly least active and the activity of 2a is intermediate be-
tween those of 1a/b and 2b. As pointed out above, this rank
order does not reflect the rank order of the antiproliferative
activities of the compounds, thus indicating that the induc-
tion of tubulin polymerization in vitro may not be a very re-
liable predictor of growth inhibitory activity in cells. At the
same time, both EC50 values for
the induction of tubulin poly-
merization as well as apparent
Table 2. Interactions of epothilone analogues 1–4 with tubulin and stabilized MTs.
EC50 TubPol.[a] Cr[b] [mm] Kel TubPol.[c]
ꢀDGel MTs[d]
Kb MTs[e]
[107 mꢀ1
ꢀDG MTs[f]
Cr values for tubulin assembly
are spread only over a narrow
numerical range, which ham-
pers the detection of possible
correlations with IC50 values for
growth inhibition, at least for a
limited number of compounds.
In an attempt to achieve a
better resolution of the MT-in-
teraction potential of the quino-
line-based epothilone analogues
1 and 2, and thus to gain a
better understanding of the re-
lationship between N-position-
ing in the side chain and MT-
binding affinity, we also investi-
gated the binding of 1a/b and
2a/b to stabilized MTs.[45] These
binding studies were performed
[mm]
[105 mꢀ1
]
[kJmolꢀ1
]
]
[kJmolꢀ1
]
1a
1b
2a
2b
3
3.2ꢁ0.4
3.4ꢁ0.5
4.3ꢁ0.8
5.2ꢁ0.5
4.3ꢁ0.8
4.1ꢁ0.5
0.28ꢁ0.03 37.4ꢁ0.3
0.29ꢁ0.12 34.3ꢁ9.9
0.34ꢁ0.02 29.3ꢁ1.3
0.40ꢁ0.05 25.2ꢁ2.8
0.53ꢁ0.12 18.8ꢁ3.4
0.51ꢁ0.10 19.6ꢁ3.2
0.41ꢁ0.19 24.7ꢁ7.9
0.26ꢁ0.12 38.7ꢁ1.2
0.29ꢁ0.04 34.5ꢁ4.2
38.7ꢁ0.2
38.5ꢁ0.6
38.1ꢁ0.1
37.7ꢁ0.3
37.2ꢁ0.5
37.3ꢁ0.4
37.7ꢁ0.7
38.8ꢁ0.7
38.8ꢁ0.4
91.8ꢁ13.2
88.0ꢁ3.7
6.92ꢁ0.22
6.12ꢁ0.23
23.4ꢁ0.8
20.7ꢁ1.0
3.63ꢁ0.5[g]
75.0ꢁ7.4[g]
14.2ꢁ3.8
52.8ꢁ0.3
52.7ꢁ0.1
46.2ꢁ0.3
45.9ꢁ0.1
49.3ꢁ0.1
49.0ꢁ0.1
44.5ꢁ0.3[g]
52.6ꢁ0.5[g]
48.0ꢁ0.7
4
Epo A 3.9ꢁ0.6
Epo B 3.0ꢁ0.3
[h]
Epo D
–
[a] Concentration required to induce 50% of maximal tubulin polymerization at 258C (10 mm of porcine brain
a/b tubulin). For compounds 3 and 4, the bis- and mono-TFA salts were used, respectively. [b] Critical tubulin
concentration in the presence of the respective ligand at 378C. Cr in the absence of ligand was 3.30ꢁ1.16 mm,
corresponding to a Kel of 3.0ꢁ0.8ꢆ105 mꢀ1 and a DGel of ꢀ32.3ꢁ0.6 kJmolꢀ1. [c] Equilibrium constant for the
MT growth reaction, that is, for the addition of a new tubulin heterodimer to a pre-existing MT polymer at
378C. [d] Binding free energy for the MT growth reaction. [e] Association constant with glutaraldehyde-stabi-
lized MTs at 358C, as determined by the Epo B displacement method of Matesanz et al.[46] [f] Binding free
energy for the association with glutaraldehyde-stabilized MTs. Errors are standard deviations for EC50 values
and standard errors of the mean for all other parameters. [g] Data are from ref. [43]. [h] Not determined.
An alternative measure of the ability of tubulin-interact-
ing agents to promote the tubulin assembly reaction is the
apparent critical tubulin concentration, Cr, in the presence
of an MT-stabilizing agent (Cr being the concentration
below which no significant formation of large polymers
occurs).[42–44] Assuming the assembly process to be ligand-
mediated (i.e., with ligand binding preceding MT assembly),
the inverse of the Cr represents a good approximation of
the equilibrium constant Kel for the growth reaction, that is,
for the addition of a new tubulin heterodimer to a pre-exist-
ing polymer [1/Cr=Kel; Eqs. (1) and (2)].[42,43] Two thermo-
dynamically equivalent pathways can be envisaged for the
ligand-mediated assembly reaction, with the ligand (“L”)
binding to either unassembled tubulin (“Tub”) [Eq. (1)] or
to the ends of the MT [Eq. (2)].[42,43]
according to a protocol that was recently developed by Ma-
tesanz et al.[46] and utilizes Epo B as a reference ligand, the
displacement of which from MTs by a given test compound
is quantified by HPLC. As illustrated by the data shown in
Table 2, this approach revealed a clear difference between
the MT-binding affinities of analogues 1a/1b and 2a/2b.
While 1a and 1b bind to MTs with similar affinities and
Kb values of the order of 109, the binding affinity of the re-
spective isomers 2a and 2b are 13- and 14-fold lower, re-
spectively. Interestingly, the difference in MT-binding affini-
ties between 1a/b and 2b mirrors the separation of these an-
alogues in the tubulin assembly experiments, although the
differences in Kb values are clearly more pronounced than
those between the EC50 values for polymerization induction
or the Cr values for tubulin assembly. For analogue 2a, the
correlation between MT-binding and the promotion of tubu-
lin assembly is less obvious; while 2a binds to MTs with sim-
ilar affinity as 2b, and thus one order of magnitude less
tightly than 1a/b, its ability to promote tubulin assembly ap-
peared to be intermediate between those of 1a/b and 2b
(Table 2). Based on the MT-binding data, the position of the
N-atom in the quinoline side chain seemingly has a signifi-
cant impact on the interactions of analogues 1 and 2 with
MTs, with the location of the N-atom in the “natural” posi-
tion leading to higher-affinity binding. While this may be
taken to indicate that the side chain N-atom in 1a/b (and,
by inference, the thiazole nitrogen in the natural side chain)
Kel
ƒ!
MT þ L þ Tub Ð MT þ L ꢂ Tub
MT þ L þ Tub Ð MT ꢂ L þ Tub
Tub-MT ꢂ L
Tub-MT ꢂ L
ð1Þ
ð2Þ
ƒ
Kel
ƒ!
ƒ
The results of Cr measurements in the presence of epothi-
lone analogues 1 and 2 are in excellent agreement with the
data obtained from the tubulin polymerization assay, with
the rank order of apparent Cr values in glycerol assembly
buffer (GAB) at 378C (Table 2) being identical to the rank
order of EC50 values from the polymerization assay. Thus,
both methods for the assessment of tubulin-polymerizing ca-
10150
www.chemeurj.org
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 10144 – 10157

Biological Evaluations of New Epothilone Analogues
FULL PAPER
is involved in hydrogen bonding to tubulin,[11] this conclu-
sion is not inevitable. Recent NMR-based structural studies
on the tubulin-bound conformations of 1a and 2a indicate
that the quinoline side chains in these ligands adopt distinct-
ly different conformations, thus affecting their stacking in-
teractions with His-227[47] (which have been suggested by
Carlomagno and co-workers to be more relevant for the tu-
bulin binding of epothilones than hydrogen bonding[13] (see
above)).
Experimental Section
General: All solvents used for reactions were purchased as anhydrous
grade from Fluka and were used without further processing. Solvents for
extractions, column chromatography, and TLC were commercial grade
and were distilled before use. TLC was performed on Merck TLC alumi-
num sheets (silica gel 60 F254). Spots were visualized with UV light (l=
254 nm) or through staining with phosphomolybdic acid or KMnO4.
Flash column chromatography (FC) was performed using Fluka silica gel
60 for preparative column chromatography (40–63 mm), unless specifical-
ly noted otherwise. NMR spectra were recorded on a Bruker AMX-300
Based on the critical tubulin concentrations for benzimi-
dazole derivatives 3 and 4, these compounds are less potent
promoters of tubulin assembly than quinoline-based ana-
logues 1a/b and 2a/b. With regard to 1a and 1b, this conclu-
sion is also supported by the EC50 values obtained in the tu-
bulin polymerization assay (which are lower for 1a and 1b
than for 3 or 4), although the latter would also suggest com-
pounds 3 and 4 to possess similar tubulin-polymerizing activ-
ities as analogue 2a and to be more potent inducers of tubu-
lin polymerization than 2b. The MT-binding affinities of 3
and 4 are higher than those of 2a and 2b, but still substan-
tially lower than those of 1a and 1b; in addition, and some-
what surprisingly, 3 and 4 also bind to MTs with comparable
affinity to Epo D, which contrasts with the substantially dif-
ferent activities of 4 and Epo D at the cellular level
(Table 1). It may be speculated that the reduced cellular po-
tency of 4, relative to Epo D, is related to reduced cellular
uptake due to the presence of the negatively charged N3-
substituent on the benzimidazole moiety, although we do
not have any experimental data that would directly support
this hypothesis.
(300 MHz),
a Bruker AV-400 (400 MHz), or a Bruker DRX-500
(500 MHz) spectrometer at room temperature (298 K). Infrared spectra
(IR) were recorded on a Jasco FT/IR-6200 instrument. Optical rotations
were measured on a Jasco P-1020 polarimeter. Melting points were mea-
sured on a Bꢀchi B-540 apparatus and are uncorrected. RP-HPLC analy-
ses were carried out on a Waters Symmetry column (C18, 3.5 mm, 4.6ꢆ
100 mm) at a flow rate of 1 mLminꢀ1 and with a detection wavelength of
254 nm. For the purification of analogues 1a/b and 2a/b, elution was per-
formed with water/CH3CN gradients without addition of TFA to the
mobile phase. For 3 and 4, the TFA-free system produced only broad
peaks and therefore compounds were eluted with 0.1% aqueous TFA/
0.1% TFA in CH3CN. Preparative RP-HPLC was carried out using a
Waters Symmetry column (C18, 5 mm, 19ꢆ100 mm) at a flow rate of
25 mLminꢀ1, using the same solvent systems as for the respective analyti-
cal separations.
Quinoline-7-carboxaldehyde
(7):
7-Methylquinoline
(6;
10.0 g,
69.84 mmol) was heated to 1608C and SeO2 (5.1 g, 45.96 mmol) was
added portionwise at this temperature (evolution of gas). The mixture
was then maintained at 150–1618C for 18 h. After cooling to RT, CH2Cl2
(60 mL) was added, leading to the formation of a dark precipitate, which
was filtered off. After concentration of the filtrate to a volume of about
25 mL, hexane (100 mL) was added, which produced a second viscous,
dark-red to brownish precipitate. Filtration of this mixture and concen-
tration of the filtrate gave the crude product, which was purified by FC
eluting with AcOEt/hexane 1:1 to give the target compound 7 as light-
1
yellow crystals (4.22 g, 58%). M.p. 81–838C; H NMR (400 MHz, CDCl3,
TMS): d=10.26 (d, J=0.6 Hz, 1H), 9.05 (dd, J=4.3, 1.7 Hz, 1H), 8.62 (s,
1H), 8.27 (d, J=8.4 Hz, 1H), 8.07 (dd, J=8.5, 1.6 Hz, 1H), 7.96 (d, J=
8.5 Hz, 1H), 7.58 ppm (dd, J=8.3, 4.2 Hz, 1H); IR (film): n˜ =1695, 1116,
840, 800, 775, 758, 754 cmꢀ1; MS (ESI): m/z (%): 158 (100) [M+H+];
HRMS: m/z: calcd for C10H7NO: 157.0528 [M+]; found: 157.0522.
Conclusions
We have accomplished the stereoselective synthesis of a
series of side-chain-modified epothilone analogues display-
ing varying antiproliferative activity, tubulin-assembling po-
tential, and MT-binding affinity. Based on the results of the
MT-binding studies, the position of the side chain N-atom in
quinoline-based epothilone analogues 1 and 2 is an impor-
tant determinant of MT-binding affinity. For 1a/b and 2b,
the binding data are consistent with the relative effects of
these analogues on tubulin polymerization, while the corre-
lation is less clear for 2a. For 1b and 2b, the difference in
MT-binding affinity is clearly reflected in their cellular po-
tencies, whereas 1a and 2a show virtually identical antiproli-
ferative activity, in spite of a more than ten-fold difference
in MT-binding affinity. The reasons for this discrepancy
remain to be elucidated. Benzimidazole-based analogues 3
and 4, although less potent than quinoline derivatives 1a/b
and 2a, are still potent antiproliferative agents in vitro. Due
to the presence of a free amino or carboxyl group, these an-
alogues may be readily conjugated with appropriately func-
tionalized tumor-targeting moieties and are thus attractive
building blocks for the preparation of tumor-targeted pro-
drugs. Experiments along these lines are currently ongoing
in our laboratory.
Acetylsultam 8: AcCl (8.25 mL, 116.13 mmol) was added to a solution of
(ꢀ)-10,2-camphorsultam (10.0 g, 46.44 mmol) in dry CH3CN (200 mL)
and the solution was heated under reflux for 19 h. After cooling to RT,
K2CO3 (12.83 g, 92.84 mmol) was added and the mixture was stirred for
2 h at RT. The solvent was then evaporated and the residue was parti-
tioned between CH2Cl2 (50 mL) and water (150 mL). The organic layer
was separated and the aqueous solution was extracted with CH2Cl2 (3ꢆ
200 mL). The combined organic extracts were dried over MgSO4, the sol-
vent was evaporated, and the residue was recrystallized from EtOH to
provide 8 as white needles (10.98 g, 84%). M.p. 129–1318C; [a]DRT
=
ꢀ104.078 (c=2.32 in CHCl3); 1H NMR (400 MHz, CDCl3, TMS): d=
3.87–3.81 (m, 1H), 3.46 (dd, J=28.4 Hz, J=13.9 Hz, 2H), 2.40 (s, 3H),
2.19–2.02 (m, 2H), 1.97–1.83 (m, 3H), 1.46–1.30 (m, 2H), 1.15 (s, 3H),
0.97 ppm (s, 3H); IR (film): n˜ =1691, 1330, 1283, 1166, 1137, 983, 746,
668 cmꢀ1; MS (ESI): m/z (%): 258.64 (100) [M+H+]; HRMS: m/z: calcd
for C12H19NO3S: 257.1086 [M+]; found: 257.1082.
Aldol product 9: CF3SO3H (1.8 mL, 20.63 mmol) was added to a solution
of Et3B in hexane (15%, 20.8 mL, 21.55 mmol) and the mixture was
stirred under Ar for 10 min at 408C. After cooling to RT, CH2Cl2 (6 mL)
was added and stirring was continued for an additional 15 min. The mix-
ture was then cooled to 08C and a solution of 8 (3.95 g, 15.34 mmol) in
CH2Cl2 (8 mL) was added over 10 min. Thereafter, a solution of diisopro-
pylethylamine (3.4 mL, 19.85 mmol) in CH2Cl2 (4 mL) at ꢀ58C was
added dropwise over a period of 10 min. After cooling this mixture to
ꢀ788C, a solution of aldehyde 7 (3.37 g, 21.44 mmol) in CH2Cl2 (17 mL)
was finally added dropwise over 35 min and the reaction mixture was
Chem. Eur. J. 2009, 15, 10144 – 10157
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10151

K.-H. Altmann et al.
stirred at ꢀ788C for 3 h. The mixture was then allowed to warm to
ꢀ108C, saturated aqueous NH4Cl solution (50 mL) was added, and the
solution was extracted with CH2Cl2 (3ꢆ140 mL). The combined organic
extracts were washed with water (140 mL), dried over MgSO4, and con-
centrated. The residue was purified by FC eluting with AcOEt/hexane
3:2 to yield 9 (5.97 g, 80%) as a 5:1 mixture of diastereoisomers (white-
yellow foam). This mixture was separated after conversion to the TBS
ethers 10 (see below). 1H NMR (400 MHz, CDCl3, TMS): d=8.91 (dd,
J=4.3, 1.7 Hz, 1H), 8.18–8.12 (m, 2H), 7.85–7.80 (m, 1H), 7.68–7.60 (m,
1H), 7.42–7.38 (m, 1H), 5.43 (dd, J=8.3, 4.1 Hz, 1H), 3.90 (dd, J=7.6,
4.9 Hz, 1H), 3.52–3.41 (m, 2H), 3.32–3.16 (m, 2H), 2.22–2.02 (m, 2H),
1.97–1.80 (m, 3H), 1.45–1.30 (m, 2H), 1.13–1.02 (m, 3H), 0.98–0.91 ppm
(400 MHz, CDCl3, TMS): d=8.92 (dd, J=4.3, 1.8 Hz, 1H), 8.16 (d, J=
8.2 Hz, 1H), 8.02 (s, 1H), 7.81 (d, J=8.4 Hz, 1H), 7.61 (dd, J=8.4,
1.7 Hz, 1H), 7.39 (dd, J=8.3, 4.3 Hz, 1H), 5.52–5.47 (m, 1H), 4.89 (dd,
J=7.2, 5.3 Hz, 1H), 2.65–2.49 (m, 2H), 2.49–2.46 (m, 3H), 0.91 (s, 9H),
0.07 (s, 3H), ꢀ0.09 ppm (s, 3H); IR (film): n˜ =2955, 2930, 2855, 2360,
2332, 1473, 1455, 1252, 1093, 947, 940, 836, 668 cmꢀ1; MS (ESI): m/z (%):
454.72 (100) [M+H+]; HRMS: m/z: calcd for C20H28INOS+H: 454.1063
[M+H+]; found: 454.1051.
Ester 34: 9-BBN (90 mg, 0.74 mmol) was added to a solution of alkene 5
(156 mg, 0.29 mmol) in THF (1.8 mL) and the mixture was stirred under
Ar at RT for 2.5 h (solution A). In a separate flask, a solution of vinyl
iodide 12 (0.100 g, 0.22 mmol) in THF (1 mL, 0.684 g, 2 mmol) was added
(m, 3H); IR (film): n˜ =2958, 2880, 1691, 1330, 1137, 1116, 840, 771 cmꢀ1
;
to
(144 mg, 0.44 mmol), AsPh3 (18 mg, 0.06 mmol), and [Pd-
(dppf)2Cl2]·CH2Cl2 (22 mg, 0.03 mmol) (solution B). Solution was
a mixture of H2O (48 mL, 2.67 mmol), DMF (1.5 mL), Cs2CO3
MS (ESI): m/z (%): 414.52 (100) [M+H+]; HRMS: m/z: calcd for
C22H26N2O4S+H: 415.1692 [M+H+]; found: 415.1677.
A
B
TBS-protected alcohol 10: Imidazole (2.51 g, 36.87 mmol) and TBSCl
(2.82 g, 18.71 mmol) were added to a solution of aldol product 9 (5.07 g,
12.23 mmol) in DMF (80 mL) and the reaction mixture was stirred at
408C for 17 h. The solution was then concentrated, CH2Cl2 (50 mL) and
water (300 mL) were added, and the organic layer was separated. The
aqueous solution was additionally extracted with CH2Cl2 (300 mL) and
the combined organic extracts were dried over MgSO4 and concentrated.
The residue was purified by FC (two columns; hexane/AcOEt 3:2) to
give the protected alcohol 10 (4.06 g, 63%) as a light-yellow oil (single
isomer). [a]RDT =ꢀ54.708 (c=3.34 in CHCl3); 1H NMR (400 MHz, CDCl3,
TMS): d=8.89 (dd, J=4.3, 1.8 Hz, 1H), 8.13 (d, J=8.3 Hz, 1H), 7.98 (s,
1H), 7.79 (d, J=8.4 Hz, 1H), 7.69 (dd, J=8.5, 1.7 Hz, 1H), 7.37 (dd, J=
8.2, 4.3 Hz, 1H), 5.46 (t, J=6.7 Hz, 1H), 3.78 (dd, J=7.8, 4.9 Hz, 1H),
3.35 (s, 2H), 3.19 (dd, J=6.7, 2.5 Hz, 2H), 2.06–1.95 (m, 1H), 1.88–1.74
(m, 3H), 1.68 (t, J=3.9 Hz, 1H), 1.38–1.23 (m, 2H), 0.86 (s, 9H), 0.82 (s,
3H), 0.54 (s, 3H), 0.07 (s, 3H), ꢀ0.14 ppm (s, 3H); IR (film): n˜ =2958,
2930, 2858, 1695, 1334, 1133, 1091, 1076, 840, 775 cmꢀ1; MS (ESI): m/z
(%): 529.84 (100) [M+H+]; HRMS: m/z: calcd for C28H40N2O4SSi+H:
529.2556 [M+H+]; found: 529.2542.
cooled to ꢀ108C and solution A was added dropwise over a period of
10 min. The mixture was allowed to warm to RT and then stirred for 2 h.
It was then diluted with AcOEt (17 mL) and water (9 mL), the layers
were separated, and the aqueous solution was further extracted with
AcOEt (2ꢆ9 mL). The combined organic extracts were washed with
water (3ꢆ5 mL), dried over MgSO4, and the solvent was evaporated. Pu-
rification of the residue by FC eluting with hexane/Et2O 3:2 gave 34
(170.5 mg, 91%) as a dark-yellow oil. [a]RDT =ꢀ19.758 (c=3.06 in CHCl3);
1H NMR (400 MHz, CDCl3): d=8.91 (dd, J=4.3, 1.7 Hz, 1H), 8.17 (d,
J=7.8 Hz, 1H), 8.00 (s, 1H), 7.79 (d, J=8.5 Hz, 1H), 7.60 (d, J=8.4 Hz,
1H), 7.40 (dd, J=8.1, 4.3 Hz, 1H), 5.17 (t, J=7.1 Hz, 1H), 4.83 (t, J=
6.4 Hz, 1H), 4.37 (dd, J=6.9, 3.2 Hz, 1H), 3.73 (dd, J=7.1, 1.9 Hz, 1H),
3.65 (s, 3H), 3.14–3.05 (m, 1H), 2.70–2.33 (m, 4H), 2.25 (dd, J=16.1,
6.9 Hz, 1H), 1.95–1.81 (m, 3H), 1.64 (s, 3H), 1.34–1.21 (m, 3H), 1.19 (s,
3H), 1.03 (s, 3H), 1.02 (d, J=7.7 Hz, 3H), 0.91–0.88 (m, 3H), 0.89 (s,
9H), 0.87 (s, 9H), 0.85 (s, 9H), 0.06 (s, 3H), 0.05 (s, 3H), 0.04 (s, 3H),
0.03 (s, 3H), 0.00 (s, 3H), ꢀ0.11 ppm (s, 3H); IR (film): n˜ =2952, 2930,
2858, 1741, 1695, 1469, 1255, 1083, 986, 940, 833, 775, 671 cmꢀ1; MS
(ESI): m/z (%): 857.26 (100) [M+H+]; HRMS: m/z: calcd for
C48H85NO6Si3+H: 856.5763 [M+H+]; found: 856.5742.
Aldehyde 11:
A 1m solution of DIBAL-H in CH2Cl2 (19.00 mL,
19.00 mmol) was added dropwise to a solution of 10 (3.96 g, 7.49 mmol)
in CH2Cl2 (30 mL) over a period of 20 min at ꢀ788C under Ar and the
mixture was stirred at this temperature for 3 h. Water (20 mL) was then
added to quench the reaction, the mixture was diluted with additional
water (400 mL) and CH2Cl2 (400 mL), and the pH was adjusted to basic
with 1 n NaOH (40 mL). The organic layer was then separated and the
aqueous solution was additionally extracted with CH2Cl2 (2ꢆ300 mL).
The combined organic extracts were washed with water (300 mL), dried
over MgSO4, and the solvent was evaporated. Purification of the residue
by FC eluting with hexane/AcOEt 1:1 (two columns) gave aldehyde 11
(1.84 g, 78%) as a light-yellow, transparent oil. [a]RDT =ꢀ58.228 (c=2.72
in CHCl3); 1H NMR (400 MHz, CDCl3, TMS): d=9.84 (dd, J=2.7,
1.9 Hz, 1H), 8.93 (dd, J=4.3, 1.8 Hz, 1H), 8.16 (d, J=8.4 Hz, 1H), 8.05
(s, 1H), 7.83 (d, J=8.5 Hz, 1H), 7.59 (dd, J=8.5, 1.7 Hz, 1H), 7.41 (dd,
J=8.2, 4.3 Hz, 1H), 5.45 (dd, J=8.1, 4.2 Hz, 1H), 2.99–2.91 (m, 1H),
2.76–2.68 (m, 1H), 0.89 (s, 9H), 0.09 (s, 3H), ꢀ0.12 ppm (s, 3H); IR
(film): n˜ =2955, 2930, 2858, 1724, 1255, 1219, 1091, 833, 775 cmꢀ1; MS
(ESI): m/z (%): 316.65 (100) [M+H+]; HRMS: m/z: calcd for
C18H25NO2Si+H: 316.1733 [M+H+]; found: 316.1720.
Carboxylic acid 35: LiOH (57 mg, 2.38 mmol) was added to a solution of
the coupling product 34 (333 mg, 0.39 mmol) in iPrOH/H2O 4:1 (10 mL)
and the mixture was maintained at 528C for 15 h. After cooling to RT,
water (3.3 mL) was added and most of the iPrOH was removed by evap-
oration. The remaining solution was treated with CH2Cl2 (41 mL), water
(33 mL), and 1 n HCl (2.5 mL). The layers were separated and the aque-
ous solution was extracted with CH2Cl2 (2ꢆ25 mL). The combined organ-
ic extracts were washed with water (4 mL), dried over MgSO4, and con-
centrated in vacuo. The crude product was purified by FC eluting with
CH2Cl2/MeOH 97:3 to yield the target compound 35 as a light-yellow oil
(287 mg, 85%). 1H NMR (400 MHz, CDCl3): d=8.86 (s, 1H), 8.49 (s,
1H), 8.27 (s, 1H), 7.85 (d, J=8.2 Hz, 1H), 7.78 (d, J=7.9 Hz, 1H), 7.47
(s, 1H), 5.36–5.27 (m, 1H), 4.92 (dd, J=9.5, 3.0 Hz, 1H), 4.55 (dd, J=
7.4, 2.6 Hz, 1H), 3.68 (t, J=3.3 Hz, 1H), 3.26–3.15 (m, 1H), 2.47–2.29
(m, 5H), 1.91–1.80 (m, 1H), 1.73 (s, 3H), 1.65–1.50 (m, 4H), 1.25 (s, 3H),
1.24–1.18 (m, 1H), 1.10 (s, 3H), 1.09 (d, J=6.0 Hz, 3H), 1.07–1.04 (m,
1H), 0.91 (d, J=7.0 Hz, 3H), 0.89 (s, 9H), 0.88 (s, 9H), 0.87 (s, 9H), 0.17
(s, 3H), 0.13 (s, 3H), 0.09 (s, 3H), 0.02 (s, 3H), ꢀ0.01 (s, 3H), ꢀ0.10 ppm
(s, 3H); IR (film): n˜ =2955, 2930, 2858, 1702, 1466, 1252, 1087, 986, 836,
775 cmꢀ1; MS (ESI): m/z (%): 842.43 (100) [M+H+]; HRMS: m/z: calcd
for C47H83NO6Si3+H: 842.5606 [M+H+]; found: 842.5585.
Vinyl iodide 12:
(7.00 mmol) in THF (7 mL) was added dropwise over a period of 15 min
to a stirred mixture of [Ph3PCH(CH3)I]I (3.97 g, 7.30 mmol) and THF
A
1m solution of Na-bis(trimethylsilyl)amide
G
Seco acid 36: A 1m solution of TBAF in THF (1.4 mL, 1.4 mmol) was
added to a solution of 35 (375 mg, 0.45 mmol) in THF (7 mL) and the
mixture was stirred at RT for 18 h. Water and AcOEt were then added
and the layers were separated. The pH of the aqueous solution was ad-
justed to 4.5 and it was re-extracted with the AcOEt phase. The layers
were separated once more and the above process was repeated. The
aqueous solution was then additionally extracted with AcOEt (2ꢆ
42 mL), and the combined organic extracts were washed with water
(28 mL), dried over MgSO4, and concentrated. The residue was purified
by FC eluting with CH2Cl2/MeOH 94:6 to provide the target compound
36 as a transparent, slightly yellow oil (264 mg, 82%) . [a]RDT =ꢀ5.408
(c=0.5 in CHCl3); 1H NMR (400 MHz, CDCl3): d=8.86 (dd, J=4.4,
(180 mL) at ꢀ788C under Ar. The mixture was then stirred for an addi-
tional 35 min at ꢀ788C and thereafter for 20 min at ꢀ158C. After cooling
to ꢀ788C once more, a solution of 11 (1.84 g, 5.83 mmol) in THF
(10 mL) was added dropwise over 5 min and the mixture was stirred at
ꢀ788C for an additional 40 min. The reaction was then quenched by the
addition of saturated aqueous NH4Cl solution (40 mL) and the resulting
mixture was poured into
a mixture of Et2O (400 mL) and water
(200 mL). The organic layer was separated, washed with water (2ꢆ
100 mL), dried over MgSO4, and concentrated. Purification of the residue
by FC, eluting with hexane/Et2O 1:1, gave the target compound 12 as a
yellow oil (1.07 g, 41%). [a]RDT =ꢀ9.388 (c=0.96 in CHCl3); 1H NMR
10152
www.chemeurj.org
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 10144 – 10157

Biological Evaluations of New Epothilone Analogues
FULL PAPER
1.6 Hz, 1H), 8.52 (s, 1H), 8.25 (d, J=7.5 Hz, 1H), 7.85 (d, J=8.4 Hz,
1H), 7.69 (dd, J=8.4, 1.5 Hz, 1H), 7.45 (dd, J=8.2, 4.4 Hz, 1H), 5.33–
5.30 (m, 1H), 4.89 (dd, J=8.3, 4.8 Hz, 1H), 4.53 (t, J=5.6 Hz, 1H), 3.74
(dd, J=4.2, 2.8 Hz, 1H), 3.21–3.18 (m, 1H), 2.55–2.49 (m, 2H), 2.48–2.44
(m, 2H), 2.43–2.34 (m, 1H), 1.95–1.85 (m, 1H), 1.78 (s, 3H), 1.65–1.50
(m, 4H), 1.21 (s, 3H), 1.20–1.07 (m, 2H), 1.12 (s, 3H), 1.10 (d, J=6.2 Hz,
3H), 0.94 (d, J=6.5 Hz, 3H), 0.89 (s, 9H), 0.86 (s, 9H), 0.13 (s, 3H), 0.09
(s, 3H), 0.08 (s, 3H), 0.02 ppm (s, 3H); IR (film): n˜ =2955, 2926, 2858,
1713, 1695, 1465, 1255, 1219, 1080, 986, 836, 768 cmꢀ1; MS (ESI): m/z
(%): 729.18 (100) [M+H+]; HRMS: m/z: calcd for C41H70NO6Si2+H:
728.4742 [M+H+]; found: 728.4723.
46 h to a solution of 1b (51 mg, 0.11 mmol) in CH2Cl2 (0.500 mL). There-
after, excess MnO2 was added to the mixture and stirring was continued
for a further 1 h. The mixture was then diluted with CH2Cl2 (45 mL) and
water (45 mL), the aqueous layer was removed, and the organic solution
was dried over MgSO4. The solvent was evaporated and the residue was
purified by FC eluting with CH2Cl2/acetone 1:1 to provide the N-oxide of
1a (59.3 mg). A portion of this material (44.3 mg, 0.09 mmol) was hydro-
genated over Raney Ni in MeOH (8 mL) at atmospheric pressure for 3 h.
The catalyst was then filtered off, the filtrate was evaporated to dryness,
and the residue was purified by FC eluting with CH2Cl2/acetone 3:1, to
provide 1a (19.7 mg, 39%) as a colorless, glassy resin. For biological ex-
periments, this material was further purified by preparative HPLC (20%
CH3CN/water for 2 min; then 20% CH3CN/water ! 95% CH3CN/water
Protected macrolactone 37: Et3N (0.271 mL, 1.95 mmol) was added to a
solution of seco acid 36 (235 mg, 0.32 mmol) in THF (8 mL) at 08C
under Ar, and then 2,4,6-trichlorobenzoyl chloride (0.254 mL, 1.63 mmol)
was added. After stirring at 08C for 30 min, the solution was diluted with
THF (25 mL) and then added over a period of 2.5 h to a previously pre-
pared solution of 4-dimethylaminopyridine (0.401 g, 3.29 mmol) in tolu-
ene (325 mL) at 708C under vigorous stirring. The reaction mixture was
then evaporated to dryness and the residue was treated with Et2O
(60 mL). Insoluble material was removed by filtration and the filtrate
was concentrated. The crude product was purified by FC (hexane/AcOEt
4:1; two columns) to afford bis-TBS-protected macrolactone 37 as a
light-yellow oil (190 mg, 83%). [a]RDT =ꢀ26.548 (c=1.8 in CHCl3);
1H NMR (400 MHz, CDCl3): d=8.92 (dd, J=4.3, 1.7 Hz, 1H), 8.16 (d,
J=8.2 Hz, 1H), 8.11 (s, 1H), 7.83 (d, J=8.5 Hz, 1H), 7.56 (dd, J=8.4,
1.6 Hz, 1H), 7.40 (dd, J=8.2, 4.4 Hz, 1H), 5.66 (d, J=10.1 Hz, 1H), 5.26
(t, J=8.3 Hz, 1H), 3.97 (dd, J=9.3, 2.8 Hz, 1H), 3.91 (d, J=8.9 Hz, 1H),
3.10–3.00 (m, 1H), 3.00–2.88 (m, 1H), 2.86–2.72 (m, 2H), 2.67–2.54 (m,
1H), 2.49–2.12 (m, 1H), 1.85–1.75 (m, 1H), 1.72 (s, 3H), 1.68–1.53 (m,
3H), 1.25–1.07 (m, 8H), 1.15 (s, 3H), 1.01 (d, J=6.6 Hz, 3H), 0.97 (s,
9H), 0.85 (s, 9H), 0.12 (s, 3H), 0.10 (s, 3H), 0.08 (s, 3H), ꢀ0.09 ppm (s,
3H); IR (film): n˜ =2952, 2930, 2858, 1741, 1695, 1466, 1380, 1252, 1155,
1097, 1019, 983, 833, 775 cmꢀ1; MS (ESI): m/z (%): 711.20 (100) [M+H+
]; HRMS: m/z: calcd for C41H67NO5Si2+H: 710.4636 [M+H+]; found:
710.4618.
in 8 min) to provide 1a (7.2 mg) as a white lyophilized powder. [a]DRT
=
ꢀ57.408 (c=0.5 in CHCl3); 1H NMR (500 MHz, [D6]DMSO): d=8.94
(dd, J=4.2, 1.6 Hz, 1H), 8.39 (d, J=8.2 Hz, 1H), 8.11 (s, 1H), 8.01 (d,
J=8.5 Hz, 1H), 7.74 (dd, J=8.5, 1.3 Hz, 1H), 7.56 (dd, J=8.3, 4.2 Hz,
1H), 6.08 (d, J=8.9 Hz, 1H), 5.19 (d, J=6.9 Hz, 1H), 4.56 (d, J=6.3 Hz,
1H), 4.28–4.15 (m, 1H), 3.56 (t, J=7.0 Hz, 1H), 3.29–3.20 (m, 1H), 3.02
(dd, J=9.2, 3.6 Hz, 1H), 2.61–2.50 (m, 2H), 2.45 (dd, J=15.7, 10.5 Hz,
1H), 2.30–2.21 (m, 1H), 2.20–2.08 (m, 2H), 1.77–1.65 (m, 1H), 1.62–1.43
(m, 2H), 1.42–1.25 (m, 2H), 1.27 (s, 3H), 1.21 (s, 3H), 1.12 (d, J=6.7 Hz,
3H), 0.97 (d, J=6.7 Hz, 3H), 0.95 ppm (s, 3H); 13C NMR (125 MHz,
[D6]DMSO): d=217.7, 170.5, 150.8, 147.5, 142.9, 135.8, 128.3, 127.2,
125.5, 124.6, 121.4, 75.7, 73.4, 70.6, 61.5, 61.1, 53.0, 45.2, 38.8, 36.2, 35.8,
32.1, 29.7, 23.2, 22.2, 22.1, 20.5, 18.6, 16.5 ppm; IR (film): n˜ =2955, 2930,
1735, 1691, 1459, 1377, 1252, 836, 758 cmꢀ1; MS (ESI): m/z (%): 499.01
(100) [M+H+]; HRMS: m/z: calcd for C29H39NO6+H: 498.2856 [M+H+
]; found: 498.2840; RP-HPLC: tR =6.47 min (20% CH3CN/water for
1 min; then 20% CH3CN/water ! 95% CH3CN/water in 8 min).
Protected macrolactone 40: Et3N (0.825 mL, 5.93 mmol) was added to a
solution of seco acid 39 (760 mg, 1.04 mmol) in THF (2 mL) at 08C, and
then 2,4,6-trichlorobenzoyl chloride (0.813 mL, 5.20 mmol) was added.
After stirring at 08C for 15 min, the solution was diluted with THF
(100 mL) and then added over a period of 3 h to a previously prepared
solution of 4-dimethylaminopyridine (1.27 g, 10.4 mmol) in toluene
(700 mL) at 758C under vigorous stirring. The reaction mixture was then
evaporated to dryness and the residue was treated with Et2O (200 mL).
Insoluble material was filtered off and the filtrate was concentrated. The
crude product was purified by FC (hexane/AcOEt 4:1) to afford bis-
TBS-protected macrolactone 40 as white crystals (550 mg, 75%). M.p.
Macrolactone 1b: HF·pyridine (14 mL in total) was added in five por-
tions over a period of 21.5 h to a solution of bis-TBS-protected 37
(180 mg, 0.25 mmol) in THF (21 mL) in a Teflon tube at 08C. The mix-
ture was stirred at RT between additions. The reaction mixture was then
added dropwise at 08C to saturated aqueous NaHCO3 solution (400 mL).
Additional solid NaHCO3 (10 g) was then added (pH 8–9) and the aque-
ous solution was extracted with AcOEt (3ꢆ160 mL). The combined or-
ganic extracts were washed with water (200 mL), dried, and concentrated
in vacuo. The crude product was purified by FC eluting with CH2Cl2/
MeOH 97:3 to provide 1b (108 mg, 89%) as a colorless, viscous foam. A
portion of this material (56 mg) was additionally purified by RP-HPLC
1
80–828C; [a]RDT =ꢀ4.88 (c=0.01 in CH2Cl2); H NMR (300 MHz, CDCl3):
d=8.89 (d, J=4.1 Hz, 1H), 8.16 (d, J=8.2 Hz, 1H), 8.11 (d, J=8.8 Hz,
1H), 7.81–7.74 (m, 2H), 7.39 (dd, J=8.3, 4.2 Hz, 1H), 5.67 (d, J=
10.1 Hz, 1H), 5.24 (t, J=8.0 Hz, 1H), 3.98 (t, J=5.6 Hz, 1H), 3.92 (d, J=
8.9 Hz, 1H), 3.07–2.92 (m, 1H), 2.74–2.56 (m, 2H), 2.21–2.14 (m, 1H),
1.83–1.55 (m, 5H), 1.72 (s, 3H) 1.36–0.99 (m, 15H), 0.95 (s, 9H), 0.85 (s,
9H), 0.12–0.08 (m, 9H), ꢀ0.10 ppm (s, 3H); 13C NMR (75 MHz, CDCl3):
d=215.0, 171.4, 150.6, 147.9, 141.1, 139.6, 136.1, 130.1, 128.0, 127.4, 125.1,
121.4, 118.8, 79.5, 77.0, 76.2, 53.3, 48.1, 39.3, 37.7, 35.2, 32.1, 31.4, 27.5,
26.4, 26.1, 24.4, 24.3, 23.2, 19.3, 18.7, 18.6, 17.8, ꢀ3.4, ꢀ3.6, ꢀ3.7,
ꢀ5.6 ppm; MS (ESI): m/z: 711.15 [M+H+].
(20% CH3CN/water for 2 min; then 20% CH3CN/water
!
95%
CH3CN/water in 8 min) to give 1b as a white lyophilisate (33 mg).
[a]RDT =ꢀ44.008 (c=0.7 in CHCl3); 1H NMR (400 MHz, [D6]DMSO): d=
8.91 (dd, J=4.1, 1.7 Hz, 1H), 8.35 (dd, J=8.4, 1.0 Hz, 1H), 8.09 (s, 1H),
7.97 (d, J=8.4 Hz, 1H), 7.70 (dd, J=8.5, 1.6 Hz, 1H), 7.53 (dd, J=8.2,
4.3 Hz, 1H), 5.88 (d, J=9.4 Hz, 1H), 5.24 (t, J=7.6 Hz, 1H), 5.18 (d, J=
6.9 Hz, 1H), 4.49 (d, J=6.1 Hz, 1H), 4.29–4.18 (m, 1H), 3.55 (t, J=7 Hz,
1H), 3.29–3.16 (m, 1H), 2.93–2.78 (m, 1H), 2.58–2.47 (m, 1H), 2.47–2.29
(m, 3H), 1.92–1.77 (m, 1H), 1.76–1.62 (m, 1H), 1.67 (s, 3H), 1.56–1.44
(m, 1H), 1.43–1.33 (m, 1H), 1.30–1.06 (m, 2H), 1.19 (s, 3H), 1.11 (d, J=
6.7 Hz, 3H), 0.94 (d, J=6.8 Hz, 3H), 0.92 ppm (s, 3H); 13C NMR
(100 MHz, [D6]DMSO): d=217.6, 170.4, 150.7, 147.6, 143.0, 138.7, 135.7,
128.1, 127.1, 125.4, 124.8, 121.3, 120.2, 75.7, 75.5, 70.8, 53.1, 44.7, 38.9,
36.6, 35.3, 31.7, 30.1, 26.1, 23.1, 22.3, 20.6, 17.8, 16.3 ppm; IR (film): n˜ =
1735, 1688, 1463, 1377, 1302, 1252, 1040, 936, 836, 750, 668, 607 cmꢀ1; MS
(ESI): m/z (%): 482.95 (100) [M+H+]; HRMS: m/z: calcd for
C29H39NO5+H: 482.2906 [M+H+]; found: 482.2892; RP-HPLC: tR
=8.25 min (20% CH3CN/water for 1 min; then 20% CH3CN/water !
95% CH3CN/water in 8 min).
Macrolactone 2b: HF·pyridine (9 mL, 344 mmol in total) was added in
three portions over a period of 15 min to a solution of bis-TBS-protected
41 (150 mg, 0.21 mmol) in THF (20 mL) in a Teflon tube at 08C and the
reaction mixture was stirred at RT for 22 h. It was then added dropwise
at 08C to saturated aqueous NaHCO3 solution (200 mL). Additional
solid NaHCO3 (10 g) was then added to the mixture. After stirring for
1 h, the aqueous solution was extracted with AcOEt (3ꢆ80 mL). The
combined organic extracts were then washed with water (100 mL), dried,
and concentrated in vacuo. The crude product was purified by FC eluting
with CH2Cl2/MeOH 98:2 to provide 2b as white crystals (90 mg, 89%).
M.p. 188.5–189.58C; [a]DRT =ꢀ11.78 (c=0.90 in AcOEt); 1H NMR
(500 MHz, [D6]DMSO): d=8.89 (dd, J=4.1, 1.6 Hz, 1H), 8.35 (dd, J=
8.2, 1.6 Hz, 1H), 8.02 (d, J=1.9 Hz, 1H), 7.99 (d, J=8.8 Hz, 1H), 7.83
(dd, J=8.8, 1.9 Hz, 1H), 7.54 (dd, J=8.5, 4.4 Hz, 1H), 5.84 (dd, J=9.1,
1.3 Hz, 1H), 5.21 (t, J=7.6 Hz, 1H), 5.16 (d, J=6.6 Hz, 1 Hz, 1H), 4.46
(d, J=6.3 Hz, 1H), 4.26–4.19 (m, 1H), 3.56–3.50 (m, 1H), 3.21–3.13 (m
(quint), 1H), 2.87–2.76 (m, 1H), 2.46 (dd, J=15.5, 3.2 Hz, 1H), 2.41–2.30
Epoxylactone 1a: 1.05 mL of a solution of pyridine (48 mL) and 30%
H2O2 (760 mL) in water (6.84 mL) and MeReO3 (19 mg, 0.019 mmol)
were each simultaneously added in seven equal portions over a period of
Chem. Eur. J. 2009, 15, 10144 – 10157
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10153

K.-H. Altmann et al.
(m, 3H), 1.89–1.77 (m, 1H), 1.74–1.60 (m, 1H), 1.66 (s, 3H), 1.50–1.32
(m, 2H), 1.28–1.04 (m, 2H), 1.21 (s, 3H), 1.11 (d, J=6.6 Hz, 3H), 0.93
(d, J=6.6 Hz, 3H), 0.91 ppm (s, 3H); 13C NMR (125 MHz, [D6]DMSO):
d=217.51, 170.34, 150.42, 147.11, 139.67, 138.71, 136.04, 128.89, 127.74,
127.51, 124.44, 121.57, 120.18, 75.55, 75.47, 70.77, 53.21, 44.62, 38.85,
36.50, 35.21, 31.63, 30.01, 26.00, 23.07, 22.63, 20.03, 17.74, 16.28 ppm; MS
(ESI): m/z: 482.9 [M+H+]; RP-HPLC: tR = 7.49 min (20% CH3CN/
water for 1 min; then 20% CH3CN/water ! 95% CH3CN/water in
8 min).
oration of the solvent afforded 20 as a yellow solid (0.88 g, 96%). M.p.
116.5–117.58C; 1H NMR (400 MHz, CDCl3): d=8.89 (d, J=2.4 Hz, 1H),
8.50 (br, 1H), 8.07 (dd, J=9.2, 2.0 Hz, 1H), 6.96 (d, J=9.2 Hz, 1H), 4.85
(br, 1H), 3.91 (s, 3H), 3.56–3.44 (m, 4H), 1.46 ppm (s, 9H); 13C NMR
(100 MHz, CDCl3): d=165.57, 147.76, 136.44, 129.54, 117.55, 113.46,
80.12, 52.12, 43.23, 39.56, 28.34 ppm; IR (film): n˜ =3363, 2976, 1707, 1623,
1566, 1524, 1440, 1362, 1287, 1227, 1163 cmꢀ1; MS (ESI): m/z: 362.73
[M+Na+]; HRMS (ESI): m/z: calcd for C15H21N3O6+Na: 362.1323
[M+Na+]; found: 362.1327.
Anilino ester 21: Nitrobenzene 20 (16.7 g, 49 mmol) was hydrogenated
over Pd/C (0.8 g) at RT and atmospheric pressure in MeOH (300 mL) for
17 h. Filtration of the mixture through a pad of Celite, evaporation of the
filtrate, and purification of the residue by FC eluting with hexane/AcOEt
2:1 ! 1:2 gave 21 as a red oil (15.1 g, 99%). 1H NMR (400 MHz,
CDCl3): d=7.53 (dd, J=8.4 Hz, 2 Hz, 1H), 7.36 (d, J=1.6 Hz, 1H), 6.52
(d, J=8.4 Hz, 1H), 5.04 (br, 1H), 3.83 (s, 3H), 3.47–3.41 (m, 2H), 3.28–
3.25 (m, 2H), 1.46 ppm (s, 9H); 13C NMR (100 MHz, CDCl3): d=167.71,
157.09, 142.46, 132.43, 123.98, 118.8, 117.61, 108.94, 79.86, 51.67, 44.77,
40.03, 28.45 ppm; IR (film): n˜ =3371, 2976, 2948, 1686, 1600, 1524, 1442,
1393, 1365, 1296, 1253, 1221, 1158, 1113 cmꢀ1; MS (ESI): m/z: 310.32
[M+H+]; HRMS (ESI): m/z: calcd for C15H21N3O6+H: 310.1761 [M+H+
]; found: 310.1759.
Epoxylactone 2a:
A solution of pyridine (48 mL) and 30% H2O2
(760 mL) in water (6.84 mL) and MeReO3 (81 mg, 0.08 mmol) were each
simultaneously added in six equal portions over a period of 25 h to a so-
lution of 2b (62.9 mg, 0.13 mmol) in CH2Cl2 (0.82 mL). Thereafter,
excess MnO2 was added to the mixture and stirring was continued for a
further 1 h. The mixture was then diluted with CH2Cl2 (110 mL), the
aqueous layer was removed, and the organic solution was dried over
MgSO4. The solvent was evaporated and the residue was purified by FC
eluting with CH2Cl2/acetone 5:4 to provide the N-oxide of 2a as a glassy
resin (54.5 mg). A portion of this material (45.1 mg) was hydrogenated
over Raney Ni in MeOH (8 mL) at atmospheric pressure for 4.5 h. The
catalyst was then filtered off, the filtrate was evaporated to dryness, and
the residue was purified by FC eluting with CH2Cl2/acetone 3:1, which
gave 2a (26.4 mg, 39%) as a colorless, glassy resin. For biological experi-
ments, this material was further purified by preparative HPLC (20%
CH3CN/water for 2 min; then 20% CH3CN/water ! 95% CH3CN/water
in 8 min) to provide 2a (14.1 mg, 20%) as a white lyophilized powder.
Ester 22: A solution of 21 (100 mg, 0.33 mmol) and triethyl orthoacetate
(0.35 mL, 1.94 mmol) in EtOH (6 mL) was heated to reflux for 19.5 h.
Evaporation of the solvent gave a dark-brown solid, which was purified
by FC eluting with AcOEt/MeOH 95:5 to provide 22 as a white solid
(104 mg, 96%). M.p. 180.1–181.68C; 1H NMR (400 MHz, CDCl3): d=
8.32 (s, 1H), 7.98 (dd, J=8.4, 1.2 Hz, 1H), 7.35 (d, J=8.4 Hz, 1H), 4.96
(br, 1H), 4.33 (t, J=5.4 Hz, 2H), 3.94 (s, 3H), 3.51 (q, J=5.7 Hz, 2H),
2.67 (s, 3H), 1.43 ppm (s, 9H); 13C NMR (100 MHz, CDCl3): d=167.54,
156.32, 153.44, 141.82, 138.37, 124.15, 123.93, 120.99, 108.68, 80.08, 52.08,
43.56, 40.29, 28.51, 13.92 ppm; IR (film): n˜ =3223, 2977, 2952, 1705, 1619,
1522, 1435, 1396, 1366, 1336, 1285, 1253, 1209, 1165, 1082 cmꢀ1; MS
(ESI): m/z: 334.13 [M+H+]; HRMS (ESI): m/z: calcd for
C17H23N3O4+H: 334.1761 [M+H+]; found: 334.1759.
[a]2D5 =ꢀ45.798 (c=1.34 in CHCl3); H NMR (500 MHz, [D6]DMSO): d=
1
8.93 (dd, J=4.2, 1.7 Hz, 1H), 8.39 (dd, J=8.3, 1.9 Hz 1H), 8.08 (d, J=
1.4 Hz, 1H), 8.04 (d, J=8.8 Hz, 1H), 7.89 (dd, J=8.8, 1.8 Hz, 1H), 7.58
(dd, J=8.3, 4.3 Hz, 1H), 6.07 (d, J=9.1 Hz, 1H), 5.19 (d, J=5.5 Hz, 1H),
4.58 (d, J=6.1 Hz, 1H), 4.27–4.19 (m, 1H), 3.56 (t, J=6.8 Hz, 1H), 3.29–
3.20 (m, 1H), 3.01 (dd, J=9.2, 3.2 Hz, 1H), 2.59–2.49 (m, 1H), 2.44 (dd,
J=15.5, 10.7 Hz, 1H), 2.27 (d, J=14.9 Hz, 1H), 2.16–2.06 (m, 1H), 1.75–
1.65 (m, 1H), 1.60–1.17 (m, 6H), 1.27 (s, 3H), 1.23 (s, 3H), 1.13 (d, J=
6.6 Hz, 3H), 0.98 (d, J=6.6 Hz, 3H), 0.95 ppm (s, 3H); 13C NMR
(125 MHz, [D6]DMSO): d=218.1, 170.9, 151.0, 147.6, 140.1, 136.5, 129.5,
128.1, 127.9, 125.0, 122.1, 76.1, 73.7, 71.1, 61.9, 61.5, 53.5, 45.6, 39.3, 36.7,
36.1, 32.5, 30.1, 23.6, 22.7, 22.6, 20.7, 19.0, 17.0 ppm; IR (film): n˜ =2966,
2933, 1735, 1688, 1252, 1147, 1051, 1008, 980, 840, 750 cmꢀ1; MS (ESI):
m/z: 499.07 [M+H+]; HRMS: m/z: calcd for C29H39NO6+H: 498.2856
[M+H+]; found: 498.2841; RP-HPLC: tR =6.17 min (20% CH3CN/water
for 1 min; then 20% CH3CN/water ! 95% CH3CN/water in 8 min).
Alcohol 23: A 1m solution of DIBAL-H in CH2Cl2 (159 mL, 159 mmol)
was added dropwise to a solution of 22 (13.23 g, 39.7 mmol) in CH2Cl2
(170 mL) at ꢀ788C. The mixture was allowed to warm to RT and stirred
for 17 h. After re-cooling to ꢀ308C, MeOH (200 mL) was added and the
precipitate formed was removed by paper filtration. Concentration of the
filtrate and recrystallization of the residue from MeOH afforded 23 as a
beige-colored solid (10.03 g, 78%). M.p. 1308C (decomposition);
1H NMR (400 MHz, CDCl3): d=7.50 (m, 1H), 7.24 (m, 2H), 5.12 (br,
1H), 4.74 (s, 2H), 4.25 (t, J=5.4 Hz, 2H), 3.50–3.44 (m, 2H), 2.57 (s,
3H), 1.44 ppm (s, 9H); 13C NMR (100 MHz, CDCl3): d=156.16, 153.29,
142.63, 135.41, 134.57, 121.95, 117.80, 109.17, 80.25, 65.79, 43.62, 40.17,
28.50, 13.84 ppm; IR (film): n˜ =3310, 3218, 2976, 2932, 2869, 1689, 1623,
1519, 1436, 1404, 1356, 1332, 1277, 1251, 1164, 1084, 1035, 1014 cmꢀ1; MS
(ESI): m/z: 306.1 [M+H+]; HRMS (ESI): m/z: calcd for C16H23N3O3+H:
306.1812 [M+H+]; found: 306.1807.
Methyl 4-fluoro-3-nitrobenzoate 19: Concentrated H2SO4 (10 mL,
178 mmol) was added to a solution of 4-fluoro-3-nitrobenzoic acid (18)
(15 g, 108 mmol) in absolute MeOH (500 mL) and the mixture was
heated to reflux for 6 h. After cooling to RT, AcOEt (300 mL) was
added and MeOH was removed under reduced pressure. The mixture
was then diluted with water (300 mL) and the aqueous phase was neu-
tralized with 2 n aqueous sodium hydroxide. The organic layer was sepa-
rated and the aqueous solution was back-extracted with CH2Cl2 (2ꢆ
200 mL). The combined organic extracts were washed with brine and
dried over MgSO4. Evaporation of the solvent afforded 19 as a yellow
solid (15.7 g, 97%). M.p. 60.5–61.18C; 1H NMR (400 MHz, CDCl3): d=
8.74 (dd, J=7.4, 2.2 Hz, 1H), 8.32 (ddd, J=8.8, 4.2, 2.0 Hz, 1H), 7.39
(dd, J=10.2, 8.6 Hz, 1H), 3.98 ppm (s, 3H); 13C NMR (100 MHz,
CDCl3): d=164.24, 159.57, 156.86, 136.66 (d, J=10.1 Hz), 128.00 (d, J=
2.2 Hz), 127.38 (d, J=4.2 Hz), 118.95 (d, J=21.2 Hz), 53.06 ppm; IR
(film): n˜ =3076, 2956, 1729, 1615, 1540, 1439, 1351, 1276, 1233, 1197,
1116 cmꢀ1; MS (ESI): m/z: 199.75 [M+H+]; HRMS (EI): m/z: calcd for
C8H6FNO4+H: 199.0276 [M+H+]; found: 199.0274.
Aldehyde 24: MnO2 (704 mg, 7.8 mmol) was added to a solution of 23
(207 mg, 0.66 mmol) in CHCl3 (1.5 mL) and the mixture was refluxed for
1 h. After cooling to RT, the mixture was filtered through a pad of Celite
and the filtrate was concentrated to yield 24 as a white solid (193 mg,
96%). m.p. 1868C (decomposition); 1H NMR (400 MHz, CDCl3): d=
10.04 (s, 1H), 8.16 (d, J=1.2 Hz, 1H), 7.88 (dd, J=5.2, 1.2 Hz, 1H), 7.49
(d, J=5.0 Hz, 1H), 4.87 (br, 1H), 4.38 (m, 2H), 3.54 (m, 2H), 2.75 (s,
3H), 1.42 ppm (s, 9H); 13C NMR (100 MHz, CDCl3): d=192.03, 156.21,
154.31, 142.24, 139.55, 131.53, 123.47, 122.15, 109.63, 80.20, 43.66, 40.24,
28.49, 13.96 ppm; IR (film): n˜ =3339, 3226, 2974, 2929, 2854, 2736, 1687,
1613, 1586, 1522, 1422, 1395, 1366, 1335, 1285, 1253, 1165 cmꢀ1; MS
(ESI): m/z: 304.38 [M+H+]; HRMS (MALDI): m/z: calcd for
C16H21N3O3+H: 304.1656 [M+H+]; found: 304.1654.
Boc-protected amino ester 20: A solution of 19 (0.53 g, 2.7 mmol) in
CH2Cl2 (5 mL) was added to a stirred solution of BocNHCH2CH2NH2
(0.48 g, 3 mmol) in CH2Cl2 (5 mL) at RT; a yellow precipitate formed im-
mediately. Et3N was then added to the mixture in four portions over a
period of 25 h (0.8 mL, 5.7 mmol in total). CH2Cl2 (5 mL) was then
added and the solution was washed once with 2% aqueous KHSO4 solu-
tion (10 mL). The aqueous layer was back-extracted with CH2Cl2 (3ꢆ
10 mL) and the combined organic extracts were dried over MgSO4. Evap-
Homoallylic alcohol 25: A 1m solution of allylmagnesium bromide in
Et2O (60 mL, 60 mmol) was added to a solution of (ꢀ)-diisopinocamphe-
yl chloride (16.17 g, 50.3 mmol) in Et2O (100 mL) at 08C under Ar and
the mixture was stirred for 1 h. Dry pentane (80 mL) was then added, the
mixture was cooled to ꢀ308C and filtered under Ar through a sintered
10154
www.chemeurj.org
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 10144 – 10157

Biological Evaluations of New Epothilone Analogues
FULL PAPER
glass septum, and the residue was washed with dry pentane (20 mL). The
clear filtrate was cooled to ꢀ788C (solution A). In a separate flask, 24
(8.99 g, 29.6 mmol) was suspended in Et2O (200 mL) and the mixture was
cooled to ꢀ1008C (suspension B). Solution A was then added dropwise
to suspension B over a period of 1 h, while keeping the temperature at
ꢀ1008C, and then the mixture was stirred for a further 1 h at the same
temperature. Dry MeOH (15 mL) was then added and the temperature
was allowed to rise to ꢀ158C, whereupon ethanolamine (20 mL,
0.33 mol) was added and the mixture was stirred for 18 h at RT. Filtration
of the suspension and concentration of the filtrate afforded 41.87 g of a
yellow oil, which was purified by FC eluting with CH2Cl2/MeOH 20:1
(three runs) to afford 25 as a white solid (9.15 g, 89%). The absolute con-
figuration of 25 was confirmed by Mosher ester analysis; ee 94%, as esti-
mated from the 19F NMR spectrum of the Mosher ester. M.p. 132.8–
137.48C; [a]2D0 =ꢀ28.28 (c=1.23 in CHCl3); 1H NMR (400 MHz, CDCl3):
d=7.29 (m, 1H), 7.19 (m, 2H), 5.84–5.70 (m, 2H), 5.14–5.04 (m, 2H),
4.73 (t, J=6.4 Hz, 1H), 4.25–4.16 (m, 2H), 3.56–3.38 (m, 2H), 3.18–2.75
(br, 1H), 2.52 (s, 3H), 2.48 (t, J=7.2 Hz, 2H), 1.45 ppm (s, 9H);
13C NMR (100 MHz, CDCl3): d=156.25, 152.36, 142.31, 138.35, 134.94,
134.35, 120.07, 118.02, 116.36, 108.85, 79.97, 73.74, 44.40, 43.47, 40.16,
28.55, 13.76 ppm; IR (film): n˜ =3318, 3215, 2980, 2930, 1688, 1518, 1433,
1403, 1365, 1275, 1252, 1165, 1039 cmꢀ1; MS (ESI): m/z: 345.97 [M+H+],
367.92 [M+Na+]; HRMS (ESI): m/z: calcd for C19H27N3O3 345.2057 [M+];
found: 345.2050.
After stirring for a further 2 h at RT, the solvent was evaporated and the
residue was purified by FC eluting with hexane/AcOEt/MeOH 25:75:0
! 15:85:0 ! 0:1:0 ! 0:10:1 to give 31 as a white solid (11.4 mg, 57%).
M.p. 108.4–109.58C; [a]2D0 =ꢀ41.38 (c=1.00 in CHCl3); 1H NMR
(400 MHz, CDCl3): d=7.70 (s, 1H), 7.26 (m, 2H), 5.58 (d, J=9.2 Hz,
1H), 5.24 (t, J=8.0 Hz, 1H), 4.66 (br, 1H), 4.25 (br, 2H), 3.95 (t, J=
5.8 Hz, 1H), 3.90 (d, J=8.8 Hz, 1H), 3.46 (q, J=5.7 Hz, 2H), 3.08–2.98
(m, 1H), 2.98–2.88 (m, 1H), 2.71 (d, J=6 Hz, 2H), 2.67–2.55 (m, 1H),
2.58 (s, 3H), 2.18–2.08 (m, 1H), 1.84–1.72 (m, 2H), 1.70 (s, 3H), 1.68–
1.52 (m, 2H), 1.43 (s, 9H), 1.24–1.14 (m, 1H), 1.14–1.06 (m, 10H), 0.99
(d, J=6.8 Hz, 3H), 0.96 (s, 9H), 0.85 (s, 9H), 0.14–0.05 (m, 9H),
ꢀ0.09 ppm (s, 3H); 13C NMR (100 MHz, CDCl3): d=215.08, 171.40,
155.87, 152.63, 142.85, 140.82, 136.04, 134.63, 120.85, 119.24, 116.24,
108.91, 80.07, 79.57, 77.52, 76.35, 53.37, 48.15, 43.28, 39.96, 39.36, 37.73,
36.03, 32.10, 31.55, 28.34, 27.53, 26.41, 26.16, 24.61, 24.14, 23.09, 19.36,
18.71, 18.60, 17.76, 13.74, ꢀ3.27, ꢀ3.68, ꢀ3.70, ꢀ5.68 ppm; IR (film): n˜ =
2955, 2932, 2888, 2857, 1737, 1699, 1518, 1468, 1388, 1365, 1253, 1162,
1096, 1066, 1018, 985 cmꢀ1; MS (ESI): m/z: 856.62 [M+H+]; HRMS
(MALDI): m/z: calcd for C47H81N3O7Si2+H: 856.5686 [M+H+]; found:
856.5681.
Macrolactone 3: Protected macrolactone 31 (20 mg, 0.023 mmol) was
added at 08C to a solution of CF3COOH (100 mL, 0.135 mmol) in CH2Cl2
(200 mL). The mixture was stirred at RT for 2 h and then the solvent was
evaporated to leave a yellow oil (29 mg). FC of this material eluting with
TES-protected alcohol 26: TES-Cl (3 mL, 17.8 mmol), imidazole (1.3 g,
19.4 mmol), and DMAP (0.2 g, 1.6 mmol) were added to a solution of al-
cohol 25 (5.6 g, 16.2 mmol) in DMF (40 mL) and the mixture was stirred
at RT for 4 h. It was then diluted with saturated aqueous NaHCO3 solu-
tion (200 mL) and the resulting mixture was extracted with AcOEt (3ꢆ
200 mL). The combined organic extracts were dried over MgSO4, the sol-
vent was evaporated, and the oily residue was purified by FC (AcOEt/
hexane 6:4 ! AcOEt ! AcOEt/MeOH 10:1) to afford 26 as a yellow
oil (7.3 g, 98%). M.p. 99.5–101.18C; [a]2D0 =ꢀ25.78 (c=1.06 in CHCl3);
1H NMR (400 MHz, CDCl3): d=7.58 (s, 1H), 7.23–7.20 (m, 2H), 5.81–
5.69 (m, 1H), 5.03–4.93 (m, 2H), 4.78 (m, 2H), 4.30–4.20 (m, 2H), 3.47
(q, J=6.0 Hz, 2H), 2.58–2.37 (m, 5H), 1.43 (s, 9H), 0.86 (t, J=7.8 Hz,
9H), 0.58–0.44 ppm (m, 6H); 13C NMR (100 MHz, CDCl3): d=156.05,
152.14, 142.59, 139.63, 135.53, 134.45, 120.63, 116.78, 116.59, 108.66, 80.04,
75.29, 46.07, 43.34, 40.13, 28.45, 13.87, 6.96, 4.98 ppm; IR (film): n˜ =3293,
2954, 2875, 1701, 1521, 1453, 1433, 1404, 1365, 1326, 1276, 1254, 1170,
1166, 1082, 1036, 1004 cmꢀ1; MS (ESI): m/z: 459.98 [M+H+]; HRMS
(ESI): m/z: calcd for C25H41N3O3Si+H: 460.2990 [M+H+]; found:
460.2990.
CHCl3/MeOH/water/AcOH 85:13:1.5:0.5 afforded
a pale-yellow oil
(16.8 mg, 64% HPLC purity), which was further purified by preparative
RP-HPLC (A/B 80:20 ! 50:50 over 6 min; A=0.1% aqueous TFA, B=
0.1% TFA in CH3CN) to afford pure 3 (bis-TFA salt) as a white solid
(7.6 mg, 44%; HPLC purity >99%). [a]2D0 =ꢀ36.98 (c=0.97 in CHCl3);
1H NMR (500 MHz, [D6]DMSO): d=8.26–8.09 (br, 3H), 7.89–7.82 (m,
2H), 7.61 (d, J=9.0 Hz, 1H), 5.78 (d, J=10.5 Hz, 1H), 5.20 (t, J=5.8 Hz,
1H), 5.14 (br, 1H), 4.59 (t, J=6.2 Hz, 2H), 4.19 (d, J=10.5 Hz, 1H), 3.52
(d, J=8.1 Hz, 1H), 3.38–3.29 (m, 2H), 3.17 (quint, J=7.2 Hz, 1H), 2.84–
2.72 (m, 4H), 2.48–2.42 (m, 1H), 2.41–2.30 (m, 2H), 2.27–2.20 (m, 1H),
1.85–1.76 (m, 1H), 1.73–1.62 (m, 4H), 1.49–1.40 (m, 1H), 1.39–1.31 (m,
1H), 1.27–1.14 (m, 4H), 1.13–1.03 (m, 1H), 1.09 (d, J=6.5 Hz, 3H), 0.93
(d, J=6.8 Hz, 3H), 0.90 ppm (s, 3H); 13C NMR (125 MHz, [D6]DMSO):
d=217.55, 170.36, 158.49 (q, J=32 Hz), 152.74, 139.36, 138.87, 131.64,
122.85, 120.15, 116.92 (q, J=297 Hz), 112.03, 111.76, 75.77, 75.55, 70.75,
53.14, 44.75, 41.99, 38.74, 37.53, 36.49, 35.80, 31.75, 30.12, 26.18, 23.11,
22.43, 20.38, 17.86, 16.35, 12.03 ppm; IR (film): n˜ =3417, 2928, 1733, 1659,
1502, 1460, 1414, 1388, 1255, 1198, 1177, 1132, 1096 cmꢀ1; MS (ESI): m/z:
528.26 [M+H+]; HRMS (ESI): m/z: calcd for C30H46N3O5+H: 528.3432
[M+H+]; found: 528.3429; RP-HPLC: tR =4.4 min (A/B 80:20 for 2 min;
then A/B 80:20 ! 50:50 over 6 min; A=0.1% aqueous TFA, B=0.1%
TFA in CH3CN).
Aldehyde 27: 2,6-Lutidine (510 mL, 4.4 mmol), a solution of OsO4 in
tBuOH (2.5% w/w, 0.6 mL, 0.1 mmol), and NaIO4 (1.9 g, 8.8 mmol) were
added to a solution of 26 (1.0 g, 2.2 mmol) in dioxane/water 3:1 (20 mL)
at RT. The mixture was stirred at RT for 23 h, and then brine (30 mL)
and Et2O (30 mL) were added, the layers were separated, and the aque-
ous layer was further extracted with Et2O (2ꢆ30 mL). The combined or-
ganic extracts were dried over MgSO4, the solvent was evaporated, and
the residue was purified by FC eluting with hexane/AcOEt/MeOH
6:4:0.1 to provide 27 as a brown oil (0.77 g, 74%). [a]2D0 =ꢀ44.58 (c=1.06
in CHCl3); 1H NMR (400 MHz, CDCl3): d=9.80 (t, J=2.2 Hz, 1H), 7.66
(s, 1H), 7.28–7.23 (m, 2H), 5.34 (dd, J=8.0, 4.4 Hz, 1H), 4.65 (br, 1H),
4.27 (t, J=5.6 Hz, 2H), 3.51–3.46 (m, 2H), 2.93–2.86 (m, 1H), 2.69–2.85
(m, 1H), 2.59 (s, 3H), 1.44 (s, 9H), 0.86 (t, J=8.0 Hz, 9H), 0.58–
0.48 ppm (m, 6H); 13C NMR (100 MHz, CDCl3): d=201.66, 152.5,
142.71, 138.20, 134.62, 120.09, 116.27, 108.99, 80.04, 70.86, 54.47, 43.25,
40.01, 28.31, 13.80, 6.71, 4.77 ppm; IR (film): n˜ =3731, 3624, 2957, 2882,
1705, 1515, 1455, 1399, 1363, 1247, 1165, 1088, 1005 cmꢀ1; MS (ESI): m/z:
Partially protected macrolactone 32: ZnBr2 (93 mg, 0.35 mmol) was
added to a solution of protected macrolactone 31 (101 mg, 0.12 mmol) in
CH2Cl2 (2.5 mL) at 0–48C under Ar and the mixture was stirred at 08C
for 72 h. It was then diluted with CH2Cl2 (5 mL) and washed with saturat-
ed aqueous NaHCO3 solution (5 mL). The aqueous layer was back-ex-
tracted with CH2Cl2 (2ꢆ6 mL). The combined organic extracts were
dried over MgSO4, the solvent was evaporated, and the residue was puri-
fied by FC eluting with CHCl3/MeOH/water/AcOH 85:13:1.5:0.5 to give
32 as a yellow oil (57 mg, 57%). In addition, 18 mg of 3 and 7 mg of the
starting material 31 were recovered. [a]2D0 =ꢀ41.68 (c=0.52 in CHCl3);
1H NMR (400 MHz, CDCl3): d=7.68 (m, 1H), 7.42 (d, J=8.1 Hz, 1H),
7.29 (d, J=8.1, 1H), 5.52 (d, J=9.9 Hz, 1H), 5.21 (t, J=7.6 Hz, 1H),
4.54–4.40 (m, 2H), 3.97–3.86 (m, 2H), 3.39–3.22 (m, 2H), 3.03 (t, J=
7.4 Hz, 1H), 2.95–2.82 (m, 1H), 2.80–2.50 (m, 6H), 2.15–2.02 (m, 1H),
1.83–1.70 (m, 2H), 1.70 (s, 3H), 1.69–1.50 (m, 2H), 1.20–1.00 (m, 10H),
0.99 (d, J=6.1 Hz, 3H), 0.94 (s, 9H), 0.84 (s, 9H), 0.14–0.01 (m, 9H),
ꢀ0.06–0.13 ppm (m, 3H); 13C NMR (100 MHz, CDCl3): d=215.16,
171.79, 152.70, 141.08, 140.97, 137.08, 134.03, 121.67, 119.13, 115.79,
109.66, 79.80, 77.37, 76.53, 53.47, 48.18, 42.51, 39.45, 39.20, 37.91, 36.01,
32.27, 31.71, 27.73, 26.53, 26.27, 24.71, 24.24, 23.23, 19.40, 18.81, 18.71,
7.92, 13.57, ꢀ3.21, ꢀ3.53, ꢀ5.51, ꢀ3.53 ppm; IR (film): n˜ =2953, 2928,
2856, 1737, 1697, 1578, 1545, 1522, 1467, 1442, 1403, 1386, 1364,
462.06
[M+H+];
HRMS
(ESI):
m/z:
calcd
for
C24H39N3O4Si+CH3OH+H: 494.3042 [M+H+MeOH+]; found: 494.3041.
Protected macrolactone 31: Et3N (16 mL, 0.137 mmol) and 2,4,6-trichloro-
benzoyl chloride (18 mL, 0.114 mmol) were added to a solution of dry 30
(20 mg, 0.023 mmol) in THF (250 mL) at 108C under Ar. The mixture
was stirred at 108C for 1 h and then diluted at 08C with toluene (2 mL).
This solution was added by means of a syringe pump over a period of 3 h
at RT to a solution of DMAP (29 mg, 0.23 mmol) in toluene (8 mL).
Chem. Eur. J. 2009, 15, 10144 – 10157
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10155

K.-H. Altmann et al.
1253 cmꢀ1; MS (ESI): m/z: 756.25 [M+]; HRMS (ESI): m/z: calcd for
C42H74N3O5Si2+H: 756.5162 [M+H+]; found: 756.5181.
[1] a) G. Hçfle, H. Reichenbach, in Anticancer Agents from Natural
Products (Eds.: G. M. Cragg, D. G. I. Kingston, D. J. Newman),
CRC, Boca Raton, 2005, pp. 413–450; b) K. Gerth, N. Bedorf, G.
Hçfle, H. Irschik, H. Reichenbach, J. Antibiot. 1996, 49, 560–564;
c) G. Hçfle, N. Bedorf, K. Gerth, H. Reichenbach, German patent
disclosure DE 4138042, 1993.
[2] For a recent book on epothilones, see: “The Epothilones: An Out-
standing Family of Anti-Tumor Agents”, in K.-H. Altmann, G.
Hçfle, R. Mꢀller, J. Mulzer, K. Prantz, Progress in the Chemistry of
Organic Natural Products, Vol. 90 (Eds.: A. D. Kinghorn, H. Falk, J.
Kobayashi), Springer, Heidelberg, 2009.
[3] D. M. Bollag, P. A. McQueney, J. Zhu, O. Hensens, L. Koupal, J.
Liesch, M. Goetz, E. Lazarides, C. M. Woods, Cancer Res. 1995, 55,
2325–2333.
[4] For recent reviews on microtubule-stabilizing agents, see: a) K.-H.
Altmann, J. Gertsch, Nat. Prod. Rep. 2007, 24, 327–357; b) D. C.
Myles, Annu. Rep. Med. Chem. 2002, 37, 125–132; c) K.-H. Alt-
mann, Curr. Opin. Chem. Biol. 2001, 5, 424–431.
Macrolactone 33: Diisopropylethylamine (8 mL, 0.05 mmol) and a solu-
tion of succinic anhydride (5 mg, 0.05 mmol) in DMF (0.2 mL) were
added to a solution of 32 (15 mg, 0.020 mmol) in DMF (0.2 mL) at RT.
After stirring at RT for 2 h, saturated aqueous NH4Cl solution (4 mL)
was added and the resulting mixture was extracted with AcOEt (3ꢆ
4 mL). The combined organic extracts were dried over MgSO4 and the
solvent was evaporated to provide 33 as a colorless oil (15.4 mg, 92%),
which was used in the next step without further purification. [a]D20
=
1
ꢀ31.28 (c=0.9 in CHCl3); H NMR (400 MHz, CDCl3): d=7.68 (m, 1H),
7.34 (m, 2H), 6.82 (br, 1H), 5.53 (d, J=10.7 Hz, 1H), 5.22 (m, 1H), 4.33
(t, J=5.3 Hz, 2H), 3.94–3.87 (m, 2H), 3.69–3.61 (m, 2H), 3.08–2.99 (m,
1H), 2.96–2.86 (m, 1H), 2.81–2.65 (m, 2H), 2.64–2.56 (m, 5H), 2.56–2.48
(m, 2H), 2.43–2.32 (m, 2H), 2.16–2.05 (m, 1H), 1.86–1.72 (m, 2H), 1.70
(s, 3H), 1.64–1.50 (m, 2H), 1.33–1.24 (m, 1H), 1.16–1.05 (m, 9H), 0.99
(d, J=6.9 Hz, 3H), 0.95 (s, 9H), 0.84 (s, 9H), 0.14–0.04 (m, 9H), ꢀ0.09–
0.12 ppm (m, 3H); 13C NMR (100 MHz, CDCl3): d=215.15, 173.62,
173.05, 171.80, 153.03, 141.15, 139.91, 137.54, 133.68, 122.12, 119.13,
115.28, 109.98, 79.72, 77.37, 76.99, 53.48, 48.31, 43.28, 39.37, 38.76, 37.82,
36.70, 35.88, 32.27, 31.27, 30.03, 29.05, 26.56, 26.29, 24.85, 24.25, 23.23,
19.43, 18.84, 18.75, 17.94, 13.01, ꢀ3.18, ꢀ3.49, ꢀ3.54, ꢀ5.54 ppm; IR
(film): n˜ =2932, 2857, 1735, 1697, 1668, 1545, 1468, 1436, 1408, 1384,
1364, 1253, 1201, 1161, 1096, 1066, 1020, 984 cmꢀ1; MS (ESI): m/z: 856.28
[M+]; HRMS (ESI): m/z: calcd for C46H77N3O8Si2+H: 856.5322 [M+H+];
found: 856.5317.
[5] a) K.-H. Altmann, M. Wartmann, T. OꢅReilly, Biochim. Biophys.
Acta Rev. Cancer 2000, 1470, M79–M91; b) R. J. Kowalski, P. Gian-
nakakou, E. Hamel, J. Biol. Chem. 1997, 272, 2534–2541.
[6] P. Giannakakou, D. L. Sackett, Y. K. Kang, Z. R. Zhan, J. T. M.
Buters, T. Fojo, M. Poruchynsky, J. Biol. Chem. 1997, 272, 17118–
17125.
[7] For reviews on the epothilone SAR, see, for example: a) K.-H. Alt-
mann, B. Pfeiffer, S. Arseniyadis, B. A. Pratt, K. C. Nicolaou, Chem-
MedChem 2007, 2, 396–423; b) ref. [1a]; c) A. Rivkin, T. C. Chou,
S. J. Danishefsky, Angew. Chem. 2005, 117, 2898–2910; Angew.
Chem. Int. Ed. 2005, 44, 2838–2850; d) M. Wartmann, K.-H. Alt-
mann, Curr. Med. Chem.: Anti-Cancer Agents 2002, 2, 123–148;
e) K. C. Nicolaou, F. Roschangar, D. Vourloumis, Angew. Chem.
1998, 110, 2120–2153; Angew. Chem. Int. Ed. 1998, 37, 2014–2045.
[8] For reviews on the chemistry of epothilones, see, for example: a) J.
Mulzer, K.-H. Altmann, G. Hçfle, R. Mꢀller, K. Prantz, Compt.
Rend. Chim. 2008, 11, 1336–1368; b) E. B. Watkins, A. G. Chitti-
boyina, M. A. Avery, Eur. J. Org. Chem. 2006, 4071–4084; c) K.-H.
Altmann, Org. Biomol. Chem. 2004, 2, 2137–2152; d) K. C. Nico-
laou, A. Ritzen, K. Namoto, Chem. Commun. 2001, 1523–1535.
[9] For recent reviews on clinical trials with epothilones, see, for exam-
ple: a) M. Harrison, C. Swanton, Expert Opin. Invest. Drugs 2008,
17, 523–546; b) J. M. G. Larkin, S. B. Kaye, Expert Opin. Invest.
Drugs 2006, 15, 691–702.
[10] E. Kaminskas, X. Jiang, R. Aziz, J. Bullock, R. Kasliwal, R. Hara-
panhalli, S. Pope, R. Sridhara, J. Leighton, B. Booth, R. Dagher, R.
Justice, R. Pazdur, Clin. Cancer Res. 2008, 14, 4378–4384. For lead-
ing references on ixabepilone, see: a) E. S. Thomas, H. L. Gomez,
R. K. Li, H.-C. Chung, L. E. Fein, V. F. Chan, J. Jassem, X. B. Pivot,
J. Klimovsky, F. Hurtado de Mendoza, B. Xu, M. Campone, G. L.
Lerzo, R. A. Peck, P. Mukhopadhyay, L. T. Vahdat, H. H. Roche, J.
Clin. Oncol. 2007, 25, 5210–5217; b) F. Y. F. Lee, R. Borzilleri, C. R.
Fairchild, S. H. Kim, B. H. Long, C. Reventos-Suarez, G. D. Vite,
W. C. Rose, R. A. Kramer, Clin. Cancer Res. 2001, 7, 1429–1437;
c) R. M. Borzilleri, X. P. Zheng, R. J. Schmidt, J. A. Johnson, S. H.
Kim, J. D. DiMarco, C. R. Fairchild, J. Z. Gougoutas, F. Y. F. Lee,
B. H. Long, G. D. Vite, J. Am. Chem. Soc. 2000, 122, 8890–8897.
[11] J. H. Nettles, H. Li, J. M. Krahn, J. P. Snyder, K. H. Downing, Sci-
ence 2004, 305, 866–869.
Macrolactone 4: Protected macrolactone 33 (21 mg, 0.025 mmol) was
added at RT to a solution of CF3COOH (30 mL, 0.40 mmol) in CH2Cl2
(300 mL) and the mixture was stirred for 5 h. Evaporation of the solvent
and purification of the residue by FC eluting with CHCl3/MeOH/water/
AcOH 85:13:1.5:0.5 afforded two batches of material with 84% (9.1 mg)
and 74% (7.4 mg) HPLC purities. HPLC purification (A/B 75:25
!
55:45 over 6 min; A=0.1% aqueous TFA, B=0.1% TFA in CH3CN) of
the combined materials gave 3.4 mg of 4 with >99% HPLC purity and a
second fraction that was 88% pure (0.9 mg), corresponding to a total
yield of 23% (calculated for the mono-TFA salt of 4). All analytical data
are for the HPLC-purified material with >99% purity. This material was
also used in the tubulin polymerization, microtubule binding, and prolif-
eration experiments. [a]2D0 =ꢀ46.48 (c=1.0 in CHCl3); 1H NMR
(400 MHz, [D6]DMSO): d=12.05 (br, 1H), 8.09 (t, J=5.8 Hz, 1H), 7.86–
7.79 (m, 2H), 7.62 (d, J=8.9 Hz, 1H), 5.79 (d, J=9.4 Hz, 1H), 5.20 (t,
J=7.5 Hz, 1H), 5.13 (br, 1H), 4.56–4.39 (m, 3H), 4.20 (d, J=10.6 Hz,
1H), 3.60–3.30 (m, 3H), 3.17 (quint, J=7.2 Hz, 1H), 2.83–2.73 (m, 4H),
2.48–2.42 (m, 1H), 2.41–2.31 (m, 2H), 2.30–2.21 (m, 3H), 2.20–2.13 (m,
2H), 1.86–1.76 (m, 1H), 1.74–1.62 (m, 4H), 1.50–1.40 (m, 1H), 1.39–1.30
(m, 1H), 1.27–1.15 (m, 4H), 1.14–1.04 (m, 4H), 0.93 (d, J=6.6 Hz, 3H),
0.90 ppm (s, 3H); 13C NMR (125 MHz, [D6]DMSO): d=217.60, 173.72,
171.77, 170.35, 158.03 (q, J=31 Hz), 152.12, 139.54, 138.83, 131.58,
122.97, 120.19, 117.04 (q, J=298 Hz), 112.15, 111.53, 75.71, 75.54, 70.69,
53.14, 44.72, 44.26, 38.74, 37.42, 36.52, 35.76, 31.74, 30.11, 29.74, 28.68,
26.18, 23.12, 22.44, 20.35, 17.84, 16.35, 13.68 ppm; IR (film): n˜ =3338,
2927, 2858, 1729, 1671, 1555, 1539, 1526, 1461, 1424, 1376, 1337, 1299,
1256, 1198, 1180, 1141 cmꢀ1; MS (ESI): m/z: 628.14 [M+H+]; HRMS
(ESI): m/z: calcd for C34H49N3O8+H: 628.3592 [M+H+]; found:
628.3588; RP-HPLC: tR =5.2 min (A/B 75:25 for 2 min; then A/B 75:25
! 55:45 over 6 min; A=0.1% aqueous TFA, B=0.1% TFA in CH3CN).
[12] T. Carlomagno, M. J. J. Blommers, J. Meiler, W. Jahnke, T. Schupp,
F. Petersen, D. Schinzer, K.-H. Altmann, C. Griesinger, Angew.
Chem. 2003, 115, 2615–2619; Angew. Chem. Int. Ed. 2003, 42, 2511–
2515.
[13] M. Reese, V. M. Sanchez-Pedregal, K. Kubicek, J. Meiler, M. J. J.
Blommers, C. Griesinger, T. Carlomagno, Angew. Chem. 2007, 119,
1896–1900; Angew. Chem. Int. Ed. 2007, 46, 1864–1868.
[14] K. C. Nicolaou, R. Scarpelli, B. Bollbuck, B. Werschkun, M. M. Per-
eira, M. Wartmann, K.-H. Altmann, D. Zaharevitz, R. Gussio, P.
Giannakakou, Chem. Biol. 2000, 7, 593–599.
Acknowledgements
We are indebted to the Swiss National Science Foundation (S.A.D.; pro-
ject numbers 200021-107876 and 205320-117594) and the ETH Zꢀrich for
generous financial support. This work was supported in part by grant
BIO2007–61336 from MEC and BIPPED-CM from the Comunidad de
Madrid to J.F.D. The excellent technical assistance by Kurt Hauenstein is
gratefully acknowledged. Dr. Bernhard Pfeiffer is acknowledged for
expert NMR support.
10156
www.chemeurj.org
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 10144 – 10157

Biological Evaluations of New Epothilone Analogues
FULL PAPER
[15] G. Bold, S. Wojeik, G. Caravatti, R. Lindauer, C. Stierlin, J. Gertsch,
M. Wartmann, K.-H. Altmann, ChemMedChem 2006, 1, 37–40.
[16] Epo D is 12,13-deoxyEpo B, that is, instead of an epoxide moiety
Epo D contains a Z double bond between C12 and C13.
[17] K.-H. Altmann, G. Bold, G. Caravatti, A. Flçrsheimer, V. Guagna-
no, M. Wartmann, Bioorg. Med. Chem. Lett. 2000, 10, 2765–2768.
[18] F. Feyen, F. Cachoux, J. Gertsch, M. Wartmann, K.-H. Altmann,
Acc. Chem. Res. 2008, 41, 21–31.
[30] H. Arimoto, M. D. Kaufman, K. Kobayashi, Y. Qiu, A. B. Smith III,
Synlett 1998, 765–767.
[31] H. C. Brown, P. K. Jadhav, J. Am. Chem. Soc. 1983, 105, 2092–2093.
[32] The ee of the allylation product 25 was determined by Mosher ester
analysis to be 94%. The optical rotation of the TBS-analogue of al-
dehyde 27 (obtained from 25 through TBS-protection and ozonoly-
sis) matched that of material prepared by the Oppolzer sultam ap-
proach (unpublished data).
[19] Side-chain modifications of the general type present in compounds
1–4 have been independently investigated by the Bayer-Schering
group: U. Klar, B. Buchmann, W. Schwede, W. Skuballa, J. Hoff-
mann, R. B. Lichtner, Angew. Chem. 2006, 118, 8110–8116; Angew.
Chem. Int. Ed. 2006, 45, 7942–7948.
[20] J. Inanaga, K. Hirata, H. Saeki, T. Katsuki, M. Yamaguchi, Bull.
Chem. Soc. Jpn. 1979, 52, 1989–1993.
[21] K.-H. Altmann, G. Bold, G. Caravatti, D. Denni, A. Flçrsheimer, A.
Schmidt, G. Rihs, M. Wartmann, Helv. Chim. Acta 2002, 85, 4086–
4110.
[22] F. Cachoux, T. Isarno, M. Wartmann, K.-H. Altmann, Angew. Chem.
2005, 117, 7636–7640; Angew. Chem. Int. Ed. 2005, 44, 7469–7473.
[23] F. Cachoux, T. Isarno, M. Wartmann, K.-H. Altmann, ChemBio-
Chem 2006, 7, 54–57.
[24] a) D. F. Meng, D. S. Su, A. Balog, P. Bertinato, E. J. Sorensen, S. J.
Danishefsky, Y. H. Zheng, T. C. Chou, L. F. He, S. B. Horwitz, J.
Am. Chem. Soc. 1997, 119, 2733–2734; b) A. Balog, D. Meng, T. Ka-
menecka, P. Bertinato, D.-S. Su, E. J. Sorensen, S. J. Danishefsky,
Angew. Chem. 1996, 108, 2976–2978; Angew. Chem. Int. Ed. Engl.
1996, 35, 2801–2803.
[33] R. M. Williams, J. Cao, H. Tsujishima, Angew. Chem. 2000, 112,
2640–2644; Angew. Chem. Int. Ed. 2000, 39, 2540–2544.
[34] J. Rudolph, K. L. Reddy, J. P. Chang, K. B. Sharpless, J. Am. Chem.
Soc. 1997, 119, 6189–6190.
[35] F. E. Kꢀhn, A. Scherbaum, W. A. Herrmann, J. Organomet. Chem.
2004, 689, 4149–4164.
[36] T. Meyer, U. Regenass, D. Fabbro, E. Alteri, J. Rçsel, M. Mꢀller, G.
Caravatti, A. Matter, Int. J. Cancer 1989, 43, 851–856.
[37] M. Wartmann, K.-H. Altmann, unpublished data.
[38] K. C. Nicolaou, B. A. Pratt, S. Arseniyadis, M. Wartmann, A.
OꢅBrate, ChemMedChem 2006, 1, 41–44.
[39] J. Gertsch, S. Meier, M. Mꢀller, K.-H. Altmann, ChemBioChem
2009, 10, 166–175.
[40] E. Hamel, A. A. del Campo, J. Lustbader, C. M. Lin, Biochemistry
1983, 22, 1271–1279.
[41] E. Hamel, D. L. Sackett, D. Vourloumis, K. C. Nicolaou, Biochem-
istry 1999, 38, 5490–5498.
[42] J. F. Dꢁaz, M. Menendez, J. M. Andreu, Biochemistry 1993, 32,
10067–10077.
[43] R. M. Buey, J. F. Dꢁaz, J. M. Andreu, A. OꢅBrate, P. Giannakakou,
K. C. Nicolaou, P. K. Sasmal, A. Ritzꢇn, K. Namoto, Chem. Biol.
2004, 11, 225–236.
[44] R. M. Buey, I. Barasoain, E. Jackson, A. Meyer, P. Giannakakou, I.
Paterson, S. Mooberry, J. M. Andreu, J. F. Dꢁaz, Chem. Biol. 2005,
12, 1269–1279.
[45] J. F. Dꢁaz, R. Strobe, Y. Engelborghs, A. A. Souto, J. M. Andreu, J.
Biol. Chem. 2000, 275, 26265–26276.
[46] R. Matesanz, I. Barasoain, C.-G. Yang, L. Wang, X. Li, C. de Inꢇs,
C. Coderch, F. Gago, J. Jimꢇnez Barbero, J. M. Andreu, W. S. Fang,
J. F. Dꢁaz, Chem. Biol. 2008, 15, 573–585.
[25] J. Chen, T. Wang, K. Zhao, Tetrahedron Lett. 1994, 35, 2827–2828.
[26] W. D. Kingsbury, I. Pendrak, J. D. Leber, J. C. Boehm, B. Mallet,
H. M. Sarau, J. J. Foley, D. B. Schmidt, R. A. Daines, J. Med. Chem.
1993, 36, 3308–3320.
[27] S. Bond, P. Perlmutter, J. Org. Chem. 1997, 62, 6397–6400.
[28] W. Oppolzer, Pure Appl. Chem. 1990, 62, 1241–1250.
[29] It should be noted that we have not rigorously established the con-
figuration of 10, as we were unable to obtain crystals suitable for X-
ray crystallographic analysis. However, the preferred stereochemical
outcome of acetate aldol reactions with Oppolzer sultam 8 is well
established (see, for example, refs. [27,28]). In addition, we have
been able to unambiguously determine the configuration of the
major aldol product obtained from 8 and an aldehyde closely related
to 7 as S (unpublished results). In the light of these facts, it is safe to
assume that 10 has the desired S configuration.
[47] M. Erdꢇlyi, T. Carlomagno, C. Stierlin, K.-H. Altmann, unpublished
results.
Received: May 22, 2009
Published online: August 20, 2009
Chem. Eur. J. 2009, 15, 10144 – 10157
ꢃ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10157