C-3′-cyclopropanated taxol analogs: Synthesis, bioassay and biostructural analysis

Article abstract of DOI:10.1002/ejoc.200500243

Ten taxoids with a cyclopropanated side chain were synthesized by coupling a spirocyclopropanated oxazoline-5-carboxylic acid with 7-(triethylsilyl) baccatin III, followed by hydrolytic ring opening and benzoyl migration. The absolute configuration of the

Full text of DOI:10.1002/ejoc.200500243

FULL PAPER
C-3Ј-Cyclopropanated Taxol Analogs: Synthesis, Bioassay and Biostructural
Analysis[‡]
Changhui Liu,[a] Markus Tamm,[b] Marcus W. Nötzel,[b] Karsten Rauch,[b]
Armin de Meijere,[b] Jennifer K. Schilling,[a] Ami Lakdawala,[c] James P. Snyder,[c]
Susan L. Bane,[d] Natasha Shanker,[d] Rudravajhala Ravindra,[d] and David G. I. Kingston*[a]
Keywords: Medium-ring compounds / Natural products / Heterocycles / Amino acids / Drug design / Medicinal chemistry
Ten taxoids with a cyclopropanated side chain were synthe-
sized by coupling a spirocyclopropanated oxazoline-5-car-
boxylic acid with 7-(triethylsilyl)baccatin III, followed by hy-
drolytic ring opening and benzoyl migration. The absolute
configuration of the 2Ј-position was determined by NMR
analysis of the corresponding Mosher esters. These paclit-
axel analogs were active in A2780 mammalian and PC-3
prostate cancer cell lines, and also in a tubulin-assembly as-
say, but all the analogs were less active than paclitaxel itself.
To probe the basis for the uniform potency reduction shown
by the cyclopropanated taxoid series, we have examined the
conformational properties of compound 3b alone by molecu-
lar mechanics and in complex with tubulin by molecular dy-
namics. In addition, we have performed an NMR/NAMFIS
conformer deconvolution analysis for compound 3b. Both
modeling and NAMFIS approaches provide a satisfying un-
derstanding of the biological behavior of the series of cyclo-
propanated taxoids and provide further, though indirect, sup-
port for the T-form as characteristic of taxoids bound to β-
tubulin.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2005)
closed, and 4a was considered to be a potential precursor to
taxol analogs with a cyclopropane-containing side chain.[9]
The synthesis of a taxol analog with a cyclopropane-con-
taining side chain was of interest, since it was hoped that
the 1,1-disubstituted cyclopropyl group in the side chain
would restrain the conformational mobility of the side
chain and bring it closer to the biologically active binding
conformation of paclitaxel. The synthetic strategy, the prin-
ciple of which had been successfully executed before,[10] is
to couple an oxazolinecarboxylic acid of general type 4 to
Introduction
Taxol[1] (1), originally isolated from the Pacific yew
(Taxus brevifolia),[2] has become an important anticancer
drug, especially for the treatment of refractory ovarian can-
cer, small-cell lung cancer, and metastatic breast disease.[3]
Extensive structure-activity relationship (SAR) studies of
taxol have been carried out and several new analogs are in
clinical trials.[4] Cyclopropyl groups have proved to be
highly effective in improving the activity of many biolo-
gically active compounds,[5,6] and several cyclopropyl-bear-
ing analogs of taxol[7] and epothilone[8] have previously
been synthesized and shown to have improved or retained
anticancer activity. Recently, a new simple access to the spi-
rocyclopropanated oxazoline-5-carboxylic acid 4a was dis-
[‡] For one of us (A. d. M.) this is to be considered as Part 115 in
the series “Cyclopropyl Building Blocks for Organic Synthesis”.
Part 114: M. Kordes, M. Brands, M. Es-Sayed, A. de Meijere,
Eur. J. Org. Chem., submitted. Part 113: M. Schelper, O. Buis-
ine, S. I. Kozhushkov, C. Aubert, A. de Meijere, M. Malacria,
Eur. J. Org. Chem., submitted.
[a] Department of Chemistry, M/C 0212, Virginia Polytechnic In-
stitute & State University,
Blacksburg, VA 24061, USA
[b] Institut für Organische und Biomolekulare Chemie der Georg-
August-Universität Göttingen,
Tammannstrasse 2, 37077 Göttingen, Germany
Fax: +49-551-399475
E-mail: Armin.deMeijere@chemie.uni-goettingen.de
[c] Department of Chemistry, Emory University,
Atlanta, GA 30322, USA
[d] Department of Chemistry, State University of New York,
Binghamton, NY 13902, USA
3962
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/ejoc.200500243
Eur. J. Org. Chem. 2005, 3962–3972

C-3Ј-Cyclopropanated Taxol Analogs
FULL PAPER
7-(triethylsilyl)baccatin III (7-TES-baccatin III) (2), and
then hydrolytically open the oxazoline moiety to the desired
side chain.
Molecular modeling has proved to be a powerful tool to
elucidate the biologically active conformation of taxol, the
T-taxol conformation.[11] The initial computationally re-
fined electron crystallographic model[11] was employed in
the design, synthesis and subsequent bioanalysis of a novel
and highly active series of C-4 to C-3Ј bridged analogs pro-
viding clearcut experimental confirmation of the model.[12]
A synthetic approach employing less constrained C-2 to C-
3Ј linkers complemented by protein docking, conceived
around the same time by the French group, furnishes paral-
lel support for the T-concept,[13] while another recent ap-
proach based on a REDOR-taxol structure gives support
to aspects of the T-concept.[14] An additional successful ap-
plication of the model was inspired by the observation that
tubulin from yeast (S. cerevisiae) is not polymerized by
taxol.[15] Mutation of five key residues in contact with T-
taxol in the tubulin binding site as portrayed by the electron
crystallographic model restored the ability of the drug to
Scheme 1. Synthesis of cyclopropanes 4a–4c. (a) NaH, MeCN, 0 °C
to room temp., 20 h, 16–60%; (b) NaOH, THF, room temp., 20 h,
58–100%.
stabilize yeast microtubules.[16] It can be added that the rela- benzoyl migration from the 2Ј-hydroxy group to the 3Ј-
tive cytotoxicity of paclitaxel and docetaxel in a cell line amino group.[19] Thus, benzoyl group migration did not
made resistant by the Asp26Glu mutation can likewise be take place under neutral aqueous conditions and basic non-
explained by a model in which both compounds adopt the aqueous conditions, and the 2Ј-O-benzoyl derivative was
β-tubulin-bound T-conformation.[17] In this paper, the con- formed (structure not shown). Treatment of the 2Ј-O-ben-
formational properties of a cyclopropyl-bearing taxoid were zoyl derivative with 0.1  NaHCO3/dioxane (1:1) led to
examined in isolation by molecular mechanics and in com- clean rearrangement, however, and the four cyclopropane-
plex with tubulin by molecular dynamics. As a complement, containing paclitaxel analogs 11, 12, 3a, and 17 were ob-
an NMR/NAMFIS conformer deconvolution analysis was tained. High-resolution FAB mass spectra (HRFABMS)
performed. The combined computational approaches pro- showed that all four compounds had the same mass and
vide a better understanding of the biological behavior of elemental composition. The chemical shift differences be-
1
this series of cyclopropanated taxoids and offer further in- tween the 7-H and 10-H protons in the H NMR spectra
sight into how taxol binds to β-tubulin.
of compounds 11 and 12 (or 3a and 17) were quite large.
By comparison with literature values,[20] it could be con-
cluded that 12 and 17 sustain the (7R) (7-epi) configuration
rather than the normal (7S) configuration. The epimeri-
zation to form small amounts of the C-7 epimers is an ex-
Synthesis
The synthesis of ( )-2-phenylspiro[cyclopropane-1Ј,4-ox- pected occurrence under the slightly basic conditions of the
azoline]-5-carboxylic acids 4a–c was performed according reaction.[4a]
to the previously developed procedures (Scheme 1).[9,18]
Syntheses of the related taxol analogs 3b–c and 13–16
The coupling of 7-TES-baccatin III (2) with ( )-2-phen- were similar to those described above, except that the DCC/
ylspiro[cyclopropane-1Ј,4-oxazoline]-5-carboxylic acid (4a) 4-PP coupling conditions were replaced by EDC/DMAP
using DCC/4-PP occurs in good yield (85%). The two dia- conditions for easier workup. Small amounts of the C-7 epi-
stereomers 7a and 8a were obtained in a ratio of 2:3 and mers of 3b and 3c were also formed, but the quantities were
were easily separated by column chromatography too small for complete characterization. With a methyl or
(Scheme 2f).
isopropyl group on the cyclopropane ring, compounds 9b–
The products 7a and 8a proved to be resistant to hydroly- 9c and 10b–10c turned out to be slightly more resistant to
sis under the conditions previously used [0.1  HCl/dioxane acid hydrolysis, but reaction could be achieved by raising
(1:1)].[10] Only starting material and various decomposition the temperature and prolonging the reaction time. The ben-
products could be detected under the literature conditions, zoyl migration from the 2Ј-hydroxy group to the 3Ј-amino
while prolonged reaction times led to cleavage of the side group also occurred less readily, and a longer time was
chain. Surprisingly, however, when the triethylsilyl groups needed. In the case of compounds 9b and 9c, the final prod-
in 7a and 8a were first removed by treatment with HF/pyri- ucts 13–16 underwent epimerization at the taxol 7-OH
dine, and the product subsequently hydrolyzed with 0.1  group under slightly basic aqueous conditions, like com-
HCl/dioxane (1:1) at 50 °C, the 3Ј-amino-2Ј-O-benzoyl de- pounds 9a and 10a. Compounds 10b and 10c, on the other
rivatives were obtained. These compounds proved to be less hand, gave only 3b and 3c with the normal taxol C-7 config-
prone than other related taxol analogs to rearrangement by uration.
Eur. J. Org. Chem. 2005, 3962–3972
www.eurjoc.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3963

D. G. I. Kingston et al.
FULL PAPER
differences ∆δRS were significant (Scheme 3). From these
data it was concluded that the taxoid 12 must have the (S)
configuration at C-2Ј, which is the configuration of 2Ј-epi-
paclitaxel. Thus, the diastereomers 3a and 17 must be (2ЈR),
and 11 must be (2ЈS).
Scheme 3. Determination of the absolute configuration of 12. (a)
(R)-α-Methoxyphenylacetic acid or (S)-α-methoxyphenylacetic
acid, EDC, DMAP, CH2Cl2, room temp., 24 h, 85%. *∆δRS
δR(18) – δS(19).
=
The absolute configurations at C-2Ј for 13–16 and 3b–c
were also checked using the Mosher ester analysis. From
these data it was concluded that taxoids 14 and 16 have the
(S) configuration at C-2Ј. The diastereomers 3b and 3c must
thus be (2ЈR), while 13 and 15 must be (2ЈS).
Cytotoxicity and Tubulin Assembly Assays
The bioactivities of taxoids 3a–c and 11–17 were com-
pared in the tubulin-assembly assay, the A2780 ovarian can-
cer cell line, and the PC-3 prostate cancer cell line (Table 1).
In general, taxoids with the unnatural configuration at the
2Ј position were much less active, while configuration at the
C-7 position plays a less important role for activity. The
most active compounds were the taxoids 3b and 3c with the
naturally occurring configuration at both C-2Ј and C-7. In
the tubulin-assembly assay, both compounds are less active
than taxol by a factor of 4–5. The cytotoxicities of com-
pounds 3b and 3c in the A2780 and PC-3 cell lines were
Scheme 2. Synthesis of C-3Ј-cyclopropanated taxol analogs. (a)
DCC or EDC, 4-PP, toluene, room temp., 24 h, 85%; (b) HF/pyri-
dine, THF, 0 °C to room temp., 24 h, 90%; (c) 0.1  HCl/1,4-diox-
ane (1:1), 50 °C, 1–3 h, 85%; (d) 0.1  NaHCO3/1,4-dioxane (1:1),
room temp., 6–24 h, 80%.
Results and Discussion
Determination of the Absolute Configuration at the
2Ј-Position
In order to determine the absolute configuration at the also less than that of taxol itself. In the A2780 cell line, all
2Ј-position, compound 12 was converted to two esters with of the cyclopropanated taxoids are much less active than
(R)- or (S)-α-methoxyphenylacetic acid (MPA), using the taxol (50–1000×), but 3b and 3c were only about an order
EDC/DMAP coupling conditions. Since the hydroxy of magnitude less active in the PC-3 cell line. The activities
groups at C-1 and C-7 in this compound are highly hin- appear to be related to the substituents on the cyclopropane
dered,[21] reaction occurred only at the 2Ј-OH position to ring. As the proton was changed to methyl, activity in-
yield the Mosher esters 18 and 19. This observation also creases by factors of 5 and 3 were observed in the tubulin-
supports the assigned configuration being (R) at C-7 for 12 assembly assay and the A2780 ovarian cancer cell line,
and 17. The resulting 2Ј-MPA esters 18 and 19 were sub- respectively. The same trend was observed for the change
jected to NMR analysis.[22] Fortunately, the chemical shift from methyl to isopropyl, although the activity does not
3964
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2005, 3962–3972

C-3Ј-Cyclopropanated Taxol Analogs
FULL PAPER
increase as dramatically in the same two assays. A plausible minal 2-phenyl group proved to be as active as PTX in the
explanation lies in the greater hydrophobicity of the ligands tubulin aggregation assay. A docking study concluded that
accompanied by an entropy gain in the binding free energy the T-conformer is the preferred form for optimal binding
as the ligand leaves the aqueous milieu and associates with to β-tubulin.
the hydrophobic binding pocket within the protein.
Table 1. Tubulin polymerization (ED50 values) and cytotoxicities
(IC50 values) of compounds 3a–c and 11–17.
Compound Tubulin assembly
A2780
IC50 [µ]
PC-3
IC50 [µ]
ED50 [µ]
Taxol (1)
0.25
6.7
1.3
0.92
Ͼ30
Ͼ30
9.5
8.4
1.8
1.7
12
0.024 0.012
8.2 0.6
2.9 0.8
1.2 0.1
Ͼ25
0.052
Ͼ6.6
0.69
0.69
Ͼ6.6
Ͼ6.6
6.45
10.0
0.88
0.95
Ͼ6.6
3a
3b
3c
11
12
13
14
15
16
17
Figure 1. Taxol analog conformations at the β-tubulin binding site.
a) T-Taxol conformer illustrating the close contact between the 4-
OAc methyl group and the ortho position of the phenyl group at
C-3Ј; b) A C-4–C3Ј-bridged taxane with activity equivalent to par-
ent taxol.
22
19
21
1
1
1
3.4 1.0
5.4 0.2
9.5 0.3
One of the consequences of this conformation as the bi-
oactive one is that the proton at C-3Ј is directed into the
concave cavity of paclitaxel when the molecule is bound to
the protein. As depicted by Figure 2a, 3Ј-H is well above
van der Waals contact with the 4-OAc methyl protons (i.e.
2.7 Å[27]). Were this proton to be replaced by an alkyl
group, it would have difficulty achieving the T-taxol local
minimum energy conformation as a result of steric repul-
sion. Indeed, 3Ј-methyltaxotere is devoid of activity in
microtubule-stabilization assays.[28]
The observation that compounds 11–17 are even less
active than diastereomeric 3b and 3c is consistent with the
above-mentioned configurational assignments, as taxoids
with a (2ЈS) configuration are usually less active than those
with a (2ЈR) configuration.[19,23]
The Conformation of Taxol on β-Tubulin: The C-3Ј
Problem
The electron crystallographic (EC) analysis of zinc-stabi-
lized microtubule sheets has provided a clear picture of the
secondary elements of the α,β-tubulin dimer[24] and con-
firmed the location of the taxoid binding site originally lo-
cated by photoaffinity labeling.[25] Subsequent analysis of
the EC structure and the corresponding density allowed
derivation of a model of the paclitaxel-binding conforma-
tion, namely T-taxol.[11] Several follow-up studies provided
support for this conformation, but the most definitive evi-
dence came from the design, synthesis and bioassay of a
series of paclitaxel analogs bridged between the 4-OAc
methyl group and the ortho position of the phenyl moiety
at C-3Ј.[26] Figure 1a illustrates the near-space alignment of
the C-4 and C-3Ј centers in the T-taxol form. Figure 1b
shows how a short bridge maintains the integrity of the
conformation while permitting the terminal phenyl rings
from the benzamido unit at C-3Ј and the benzoyl unit at C-
2 to surround His227 in β-tubulin.[11] Bridging in this man-
ner is highly effective at stabilizing the T-taxol shape. The
Figure 2. T-Taxol conformations illustrating the potential for steric
congestion between the 4-OAc methyl group and the C-3Ј center.
a) The H···H distances between 3Ј-H and the 4-OAc methyl group
in T-taxol are above the sum of the van der Waals radii (2.7 Å); b)
cyclopropyl H atoms as in 3b engage in severe van der Waals con-
tacts with the 4-OAc methyl group (2.1 Å).
Internal Strain in the C-3Ј-Cyclopropanated Taxoids
In the cyclopropyl series of Table 1, the 3Ј-phenyl group
compounds proved to match and surpass taxol in both their of paclitaxel is replaced by CHR1, namely the edge of the
microtubule-stabilizing capacity and cytotoxicity.[12,13] cyclopropane that would formally confer the (S) configura-
More recently, the French team reported an efficient route tion on C-3Ј. It is well known that alkyl and alkenyl substi-
to a highly active series of taxol analogs bridged between tutions at this center are well tolerated.[29–31] However, for
the C-2 benzoyl phenyl group and an alkyl replacement of 3a–c and 11–17 the 3Ј-H atom is also replaced with CH2
the 3Ј-benzamido functionality.[14] While all the compounds of the cyclopropane to generate a quaternary center. While
were significantly weaker cytotoxins than docetaxel, unsat- tetrasubstitution at C-2Ј has been shown to provide taxanes
urated (E) and (Z) analogs with a C7 spacer linking the 3Ј- with activity comparable to taxol,[32,33] to the best of our
NHC(=O) carbon atom and the meta position of the ter- knowledge the C-3Ј variant has not been previously re-
Eur. J. Org. Chem. 2005, 3962–3972
www.eurjoc.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3965

D. G. I. Kingston et al.
FULL PAPER
ported to deliver effective anti-tubulin agents. In fact, as several subsequent studies with taxane derivatives,[33,34,39,40]
speculated for the 3Ј-methyl group mentioned above, the conformational analyses were performed on the compounds
cyclopropane CH2 group most likely experiences a strongly in solution. The combination of a 2D NMR NOESY or
disfavored steric clash with the 4-OAc methyl group in the ROESY experiment, an exhaustive conformational analysis
T-taxol conformation (Figure 2b). While cyclopropanes 3a– and a NAMFIS (NMR analysis of molecular flexibility in
c and 11–17 cannot undergo full torsional relaxation within solution)[41,42] deconvolution of the averaged NMR spec-
the tubulin binding site, a molecular dynamics treatment trum provided minimum-energy conformers assigned spe-
of the 3b–tubulin complex demonstrates that, although the cific populations in the dynamic conformational equilib-
cyclopropyl methyl group overlaps nicely with the 3Ј-phenyl rium. In all these cases involving active taxanes, T-forms
group, the ligand’s reaction to internal steric compression were found among the set of solution conformations and
is to uncoil to a sufficient extent to reposition the 3Ј-benz- presumed to serve as the bioactive forms.
amidophenyl group 2.0–2.2 Å further away from the bacca-
Above, we have rationalized the lack of activity of the C-
tin core by comparison with taxol. As a result, the terminal 3Ј-cyclopropanated taxanes as due to an excessive energy
phenyl group of the taxoid suffers conflict with tubulin side penalty associated with achieving the T-taxol conformation.
chains in the hydrophobic binding pocket similar to what To substantiate this interpretation, we have performed a 2D
we reported for a pair of low-activity bridged taxoids,[34] ROESY analysis for 3b in CDCl3, the results of which are
although the conflict is with the side rather than the bottom given in Table 2. Sixteen well-resolved ROE cross-peak am-
of the taxoid pocket (Figure 3).
plitudes provided the same number of intramolecular sepa-
rations by comparison with the r[H(3)–H(10)] standard dis-
tance of 2.9 Å. Two additional weak cross peaks (see
Table 2) were arbitrarily assigned internal separations of
4.0 Å. Conformers of 3b were generated by an MMFF/
GBSA/H2O conformational search in MacroModel yield-
ing 139 fully optimized conformers, the global minimum
having been located 109 times during the 10000 step search.
NAMFIS intersection of the NMR spectroscopic data and
the conformational dataset identified five conformations
with populations of 31, 25, 24, 17 and 4% that match the
data quite well (SSD[41,42] = 29). None correspond to the
T-taxane conformation. The first, second and fifth forms
involve hydrophobic collapse of the terminal phenyl rings
of 3Ј-NHCOPh and 2-OCOPh as expected for the non-po-
lar taxol conformer.[43] The third and fourth forms locate
the side chain terminal phenyl and cyclopropyl groups in
very different regions of space by comparison with the cor-
responding side chain phenyl groups of taxol.
Figure 4 illustrates the relationship of the five NAMFIS-
derived conformers with T-taxol. Four atoms of the rigid
baccatin rings of all six structures have been superimposed.
At the right of the graphical composite, the baccatin cores,
the 2-benzoyl groups and the 10-acetyl groups are nicely
aligned. By contrast, at the terminus of the C-13 side chain
in the foreground of the graphic, serious misalignment with
the phenyl rings of T-taxol rendered in blue is observed. In
two cases, the methyl groups on the cyclopropanes are situ-
ated near the center of the 3Ј-phenyl ring of taxol (Figure 4,
yellow and red). In all of the conformations, the benzamido
side chains of the same taxoids occupy entirely different
regions of space in comparison with the parent ligand. If
the T-taxol conformation of 3b is a torsional partner in the
conformer equilibrium mixture in solution, it would appear
to fall in the NMR noise; i.e. at or below a population of
3%.
Figure 3. T-Taxol (blue) and cyclopropyl analog 3b (red) docked
in the β-tubulin binding pocket and relaxed by low-temperature
molecular dynamics. In response to internal steric strain in the con-
cave region of the molecule, the molecular volume of the 3b ex-
pands by 2.0–2.2 Å from C-4 to C-3Ј and places the 3Ј-benzamido
phenyl group (Bz-Ph) in unfavorable contact with tubulin’s Glu22
and Val23 residues.
One possible resolution of the problem is to retain the
substituted cyclopropyl group, but eliminate the acetate
substituent at C-4. Both 4-deacetyl-[35,36] and 4-deacetoxy-
taxol[37] are known, but the compounds are ineffective in
tubulin assembly and cytotoxicity bioassays. By contrast,
the compound incorporating the simple and less congested
4-methoxy compound is nearly as active as parents taxol
and docetaxel.[38] Conceivably, the combination of 4-OR (R
= H, Me) and C-3Ј-cyclopropanated taxol modifications
might induce activity in the analog series of Table 1.
To ascertain whether T-forms exist at all within the 139
conformer dataset, we searched the latter against T-taxol
by requiring a 3-D match of five baccatin core atoms, the
Solution Conformations of Cyclopropane 3b: NMR/
NAMFIS Analysis
In the original proposal for a T-shaped taxol conforma- centroids of the 3Ј-NHCOPh and 2-OCOPh phenyl rings
tion as the bioactive binding form on β-tubulin[11] and in and the 3-Ph vs. the cyclopropyl–Me group, respectively.
3966
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2005, 3962–3972

C-3Ј-Cyclopropanated Taxol Analogs
FULL PAPER
Table 2. Interproton distances [Å] for 3b calculated from 2D NMR ROESY measurements in CDCl3.
Proton interactions
ROESY distances
NAMFIS average distances
Difference
H(5Јα or β)–H(2Ј)
H(6Ј)–H(2Ј)
H(4-OAc)–H(2Ј)
H(10)–H(3)
H(CONH)–H(5Ј)
H(6α)–H(6β)
H(3)–H(7)
H(2)–H(16)
H(10)–H(7)
H(10)–H(18)
H(13)–H(17)
H(2)–H(19)
H(6α)–H(7)
H(6α)–H(5)
H(CONH)–H(3Ј-Ph-o-H)
H(20β)–H(19)
H(20α)–H(19)
H(20α,β)–H(2-Ph-o-H)
H(20β)–H(2)
2.70
2.50
2.52
2.76
3.26
–
0.18
0.26
0.16
–
3.10
(2.90)[a]
3.00
3.77
1.73
2.35
2.59
2.36
2.57
2.58
2.80
2.40
2.24
2.79
4.23
2.55
4.04
3.04
0.77
0.27
0.15
0.09
0.04
0.27
0.18
0.20
0.60
0.16
0.01
1.13
0.15
0.04
0.96
2.00
2.50
2.50
2.40
2.30
2.40
2.60
3.00
2.40
2.80
3.10
2.40
weak[b] (4.00)
weak[b] (4.00)
[a] Standard cross peak/distance from which all other distances were obtained. [b] These weak peaks could not be quantified accurately.
Thus, the distances were assumed to be 4.0 Å.
Ͼ120-fold poorer than taxol, the cell-based assessments
show IC50(cyclopropane)/IC50(taxol) ratios from 13–1000
(Table 1). Although a number of interplaying factors can
account for both the differences between the isolated tubu-
lin and cell-based assays and the range of responses as a
function of cyclopropane substitution, the common factor
for all compounds is the incorporation of a quaternary cen-
ter at C-3Ј. In the C-13 side chain rotamer believed to repre-
sent the bioactive T-taxoid conformation, the methylene
group (CH2) of a cyclopropanated taxane would encounter
the methyl group of the 4-acetyl group to produce struc-
tures incompatible with productive binding to the protein.
This scenario explains the uniformly lowered activity for
the C-3Ј-cyclopropanated taxanes cited in Table 1. Relative
to taxol, the best compounds, 3b and 3c, are 4–6-fold and
Figure 4. Five NAMFIS conformations of cyclopropane 3b in
CDCl3 (green, magenta, yellow, red, cyan) superimposed on the T-
10–120-fold less potent in the tubulin polymerization and
cytotoxicity assays, respectively. The activity reduction is
taxol conformer (blue) within the baccatin core. At the C-13 ter-
very similar to that observed for bridged taxols incorporat-
ing less than optimal linkers between C-4 and C-3Ј (10-
fold).[34] In that case, the loss of activity was traceable to an
unfavorable interaction between ligand and protein analo-
gous to that depicted in Figure 3. An added perturbation
for the C-3Ј-cyclopropanated derivatives is the internal
strain caused by the short interatomic contacts shown in
the constrained T-form of Figure 2b. According to MD re-
laxation within the tubulin binding pocket, the structure of
3b does not remain in this rotameric state. The cyclopropyl
group twists away from the 4-acetyl group, displaces the 3Ј-
benzamido phenyl group and causes the unfavorable inter-
actions with the protein depicted in Figure 3. Thus, while
the cyclopropanated taxane ligand can still fit within the β-
tubulin binding pocket, there is an energy penalty contrib-
uting to the reduced activity. These considerations are also
consistent with the failure of the NAMFIS treatment to
detect T-taxoid forms for 3b in solution. The latter observa-
tion contrasts with previous NMR/NAMFIS solution
analyses of taxanes with moderate to high activity, all of
mini, neither the phenyl rings nor the cyclopropane rings of the
synthetic taxoids match the phenyl rings of taxol. Two methyl
groups (yellow and red) fall near the middle of taxol’s 3Ј-phenyl
group.
No structures fall within an overall root mean square devia-
tion of 1.0 Å. Graphical inspection of those exhibiting a 1.0
to 1.5 Å rms deviation from the T-taxol target illustrates
sufficient structural deviation to eliminate these conformers
from consideration as the bioactive form within the β-tubu-
lin binding site. In sum, neither the NAMFIS-determined
conformations in CDCl3 nor the MMFF database of mini-
mized torsional rotamers contains the T-taxoid form.
Conclusions
A series of C-3Ј-cyclopropanated taxols was synthesized
and assayed for microtubule stabilization and cytotoxicity
against two cell lines. While the tubulin assembly measure-
ments demonstrate compound activities ranging from 4 to
Eur. J. Org. Chem. 2005, 3962–3972
www.eurjoc.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3967

D. G. I. Kingston et al.
FULL PAPER
was stirred at ambient temperature for 17 h and then heated under
reflux for an additional 2 d. After cooling to ambient temperature,
cold water (900 mL) was added. The brown solution was extracted
with dichloromethane (6×150 mL), and the combined organic lay-
ers, containing trimethyl (E/Z)-2-chloro-2-(2Ј-isopropylcyclopro-
pylidene)orthoacetate, were concentrated to a volume of ca.
600 mL. Strongly acidic ion exchange resin (46 g) (Bayer) catalyst
K 2621, (styrene-divinylbenzene copolymer with sulfonic acid
groups) was added, and the mixture stirred vigorously for 21 h.
The catalyst was filtered off and washed with dichloromethane
which evidence at least a low population contribution of
the T-conformer.[33,34,39,40] The analysis for 3b implies that
a distorted T-form (Figures 3 and 4) is most likely the best
binding compromise for C-3Ј-cyclopropanated analogs, but
that it leads to weak activity in the most favorable cases.
The result provides indirect confirmation that the T-geome-
try is the bound form for unstrained taxanes on α,β-tubulin.
Within the context of the classic baccatin core, the loss of
activity by C-3Ј substitution would seem to be general. It
suggests, however, that taxol-like activity for the cyclopro- (100 mL). After removal of the solvent under reduced pressure, the
residue was subjected to column chromatography on silica gel
(200 g, 3.8×35 cm) eluting with pentane/ether (20:1), Rf = 0.32, to
yield 23.1 g (57%) of (E/Z)-2-chloro-2-(2Ј-isopropylcyclopropyli-
dene)acetate as a pale yellow liquid, (E)/(Z) ratio 2.64:1 according
panoids might be recoverable in 4-deacetylated analogs or
in the corresponding ethers.
to GC (column: CP-SIL 5CB, 25 m). IR (KBr, film): ν = 2959 cm–1,
˜
Experimental Section
2872, 1732, 1436, 1264, 1065, 908, 760. 1H NMR (600 MHz,
3
C6D6): Major isomer (E): δ = 0.76 (d, J = 6.6 Hz, 3 H), 0.93 (d,
General: 1H and 13C NMR: Spectra were recorded at 600 (1H), and
62.9, 75.5 (13C, additional DEPT or APT) MHz with Varian IN-
OVA 600, Bruker AM 250 and Varian Mercury 300 instruments in
CDCl3, C6D6 or [D6]acetone solution and, if not otherwise speci-
fied, solvent residual signals used as internal reference; δ in ppm,
J in Hz. 1H-NOE experiments were recorded at 600 MHz with a
Varian INOVA 600 instrument in C6D6 solution. IR: Bruker IFS
66 (FT-IR) instrument, measured as KBr pellets, oils as films be-
tween KBr plates. MS (EI) and HRMS (EI): Finnigan MAT 95
spectrometer. M.p.: Büchi 510 capillary melting point apparatus,
values are uncorrected. TLC: Macherey & Nagel precoated sheets,
0.20 mm Sil G/UV254. Column chromatography: Merck silica gel,
grade 60, 70–230 mesh ASTM. Starting materials: Anhydrous ace-
tonitrile was distilled from CaH2. Tetrachlorocyclopropene[44] and
J = 6.4 Hz, 3 H), 0.91–0.96 (m, 1 H), 1.07–1.12 (m, 1 H), 1.17–
1.25 (m, 2 H), 3.37 (s, 3 H); minor isomer (Z): δ = 0.65 (d, J =
6.7 Hz, 3 H), 0.88 (d, J = 6.7 Hz, 3 H), 1.08–1.11 (m, 1 H), 1.28–
1.34 (m, 1 H), 1.45–1.53 (m, 1 H), 1.53–1.58 (m, 1 H), 3.33 (s, 3
H) ppm. 13C NMR (75.5 MHz, CDCl3, add. APT): Major isomer
(E): δ = 13.7 (–), 21.5 (+), 21.7 (+), 26.0 (+), 30.7 (+), 52.9 (+),
114.1 (–), 143.5 (–), 162.9 (–); minor isomer (Z): δ = 8.9 (–), 19.9
(+), 21.4 (+), 29.7 (+), 29.9 (+), 52.7 (+), 112.9 (–), 143.9 (–), 162.6
(–) ppm. MS (EI, 70 eV): m/z (%) = 189/187 (11/26) [M]+, 175 (40),
173 (100), 157 (12), 121 (22), 105 (28), 93 (81), 91 (60), 77 (49), 69
(21), 59 (13).
Methyl (5ЈS*)-2Ј-Isopropyl-2-phenylspiro[cyclopropane-1Ј,4-oxazo-
line]-5-carboxylate (6c): A solution of methyl (E/Z)-2-chloro-2-(2Ј-
isopropylcyclopropylidene)acetate [ratio (E)/(Z) 2.64:1] (3.773 g,
20.0 mmol) and benzamide (2.423 g, 20.0 mmol) in anhydrous ace-
tonitrile (90 mL) was treated with NaH (0.920 g, 23.0 mmol, 60%
dispersion in mineral oil) at 0 °C. The suspension was subsequently
stirred at ambient temperature for 20 h. The whole reaction mixture
was filtered through silica gel (10 g, 2.5×4 cm), the silica gel was
rinsed with diethyl ether (250 mL), and the solvents were removed.
The oily residue was subjected to column chromatography on silica
gel (170 g, 3.7×30 cm) eluting with pentane/ether (10:1), Rf = 0.23,
to yield 834 mg (16%) of 6c as a colorless oil. The stereochemical
methyl
(E/Z)-2-chloro-2-(2Ј-methylcyclopropylidene)acetate[45]
were prepared according to published procedures. All other chemi-
cals were used as commercially available from Merck/VWR, Ald-
rich, Acros and Bayer. For the synthesis of all other compounds,
chemicals were obtained from Aldrich Chemical Co. and were used
without further purification. All solvents were of reagent grade or
HPLC grade. THF was distilled from sodium/benzophenone, and
1
CH2Cl2 was distilled from calcium hydride. All H NMR spectro-
scopic data were obtained in CDCl3 or CD3OD with Varian Unity
400 spectrometer (operating at 399.951 MHz for 1H and
100.578 MHz for 13C). Mass spectra were obtained at the Analyti-
cal Service in the Department of Chemistry (HRFABMS) or De-
partment of Biochemistry (MALDI-TOF MS) at Virginia Tech.
assignments rest on NOE experiments. IR (KBr): ν = 2956 cm–1,
˜
1763, 1745, 1653, 1636, 1451, 1287, 1061, 695. 1H NMR (600 MHz,
3
C6D6): δ = 0.66 (dd, J = 5.6, 6.7 Hz, 1 H), 0.83 (d, J = 6.7 Hz, 3
H), 0.91 (d, J = 6.7 Hz, 3 H), 1.09–1.19 (m, 1 H), 1.31 (ddd, J =
6.7, 9.6, 10.7 Hz, 1 H), 1.44 (dd, J = 5.5, 9.4 Hz, 1 H), 3.23 (s, 3
H), 4.90 (s, 1 H), 7.02–7.06 (m, 3 H), 8.19–8.23 (m, 2 H) ppm. 13C
NMR (62.9 MHz, CDCl3, add. DEPT): δ = 15.3 (–), 21.9 (+), 28.0
(+), 32.0 (+), 52.1 (+), 57.9 (Cquat), 76.1 (+), 126.9 (Cquat), 127.8
(+), 128.2 (+), 131.2 (+), 162.1 (Cquat), 169.8 (Cquat) ppm. MS (EI,
70 eV): m/z (%) = 273 (17) [M+], 230 (100), 217 (17), 186 (26), 158
(21), 105 (27), 77 (10). C16H19NO3: calcd. C 70.31, H 7.01, N 5.13;
found C 70.17, H 6.84, N 5.05.
Methyl
(E/Z)-2-Chloro-2-(2Ј-isopropylcyclopropylidene)acetate
(5c):[44] A sampling cylinder (320 mL, monel alloy) was charged
with tetrachlorocyclopropene.[44] (80.0 g, 0.449 mol), 3-methyl-1-
butene (94.6 g, 1.35 mol) and anhydrous potassium carbonate
(8.9 g, 64 mmol). The carefully sealed cylinder was heated at 170 °C
for 20 h. After cooling to ambient temperature, the cylinder was
emptied and rinsed with dichloromethane (2×30 mL). The brown
reaction mixture was filtered to remove inorganic salts. The excess
of 3-methyl-1-butene and the dichloromethane were removed under
reduced pressure and the liquid residue was bulb-to-bulb-distilled
(5S*)-2Ј-Isopropyl-2-phenylspiro[cyclopropane-1Ј,4-oxazoline]-5-car-
(70 °C, 0.01 mbar) to yield 86.2 g (77%) of (E/Z)-1-chloro-2-iso- boxylic Acid (4c): A solution of 6c (600 mg, 2.2 mmol) in tetra-
propyl-1-(trichlorovinyl)cyclopropane as a pale yellow liquid, (E)/
(Z) ratio 1.37:1 according to GC (column: CP-SIL 5 CB, 25 m).
This product was used for the subsequent transformation without
further purification. Into a stirred solution of sodium methoxide
[freshly prepared by disolving sodium (39.4 g, 1.71 mol) in anhy-
drous methanol (400 mL)] was added dropwise a solution of (E/Z)-
hydrofuran (41 mL) was stirred vigorously with 5  aqueous so-
dium hydroxide (6.2 mL) at ambient temperature for 22 h. After
addition of acetic acid (10 mL) and stirring for an additional
10 min, all solvents were removed under reduced pressure. The resi-
due was taken up in water (25 mL) and the aqueous solution was
extracted with diethyl ether (2×100 mL). The combined organic
layers were washed with water (25 mL). After removal of the sol-
vent, the residual colorless solid was filtered through silica gel (11 g,
1-chloro-2-isopropyl-1-(trichlorovinyl)cyclopropane
(53.1 g,
0.214 mol) in anhydrous methanol (60 mL) at 5 °C. The mixture
3968
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2005, 3962–3972

C-3Ј-Cyclopropanated Taxol Analogs
FULL PAPER
1.5×6 cm) eluting with dichloromethane to remove residual acetic
acid, then eluting the product with ether/methanol (10:1). Yield,
after careful drying under reduced pressure, 388 mg (68%) of 4c as
0.1 mL of anhydrous pyridine, then the solution was cooled to 0 °C,
and 0.1 mL of HF/pyridine (70% HF, 30% pyridine) was added.
The reaction mixture was allowed to warm to room temperature,
and stirred overnight. The reaction mixture was then diluted with
EtOAc, the organic phase was washed with sodium hydrogencar-
a colorless solid, m.p. 70–72 °C. IR (KBr): ν = 2958 cm–1, 1734,
˜
1
1646, 1452, 1264, 1069, 694. H NMR (600 MHz, [D6]acetone): δ
= 0.73 (dd, J = 5.7, 5.9 Hz, 1 H), 1.04 (d, J = 6.6 Hz, 3 H), 1.07 bonate, water, and brine, dried with sodium sulfate, and concen-
3
3
3
(d, J = 6.6 Hz, 3 H), 1.10–1.30 (m, 3 H), 4.91 (s, 1 H), 7.44–7.49
(m, 2 H), 7.51–7.55 (m, 1 H), 7.91–7.95 (m, 2 H) ppm. 13C NMR
(62.9 MHz, CDCl3, add. DEPT): δ = 14.9 (–), 22.0 (+), 27.9 (+),
31.8 (+), 56.9 (Cquat), 76.3 (+), 126.0 (Cquat), 128.2 (+), 128.4 (+),
131.9 (+), 163.8 (Cquat), 172.0 (Cquat) ppm. MS (EI, 70 eV): m/z
(%) = 259 (18) [M+], 216 (100), 203 (20), 158 (17), 144 (13), 105
(54), 77 (17). HRMS (EI, 70 eV) calcd. for C15H17NO3 [M+]
259.1208, found 259.1208.
trated in vacuo. The residue was applied to preparative TLC (50%
EtOAc/hexane) to give 9a (12.3 mg, 97%). Compounds 9b (99%),
9c (94%), 10a (99%), 10b (71%), and 10c (98%) were prepared
similarly.
(Baccatin III)-13-yl (5S)-2-Phenylspiro[cyclopropane-1Ј,4-oxazo-
line]-5-carboxylate (9a): 1H NMR (400 MHz, CDCl3): δ = 1.03 (m,
2 H, overlapped), 1.17 (s, 3 H), 1.26–1.30 (m, 1 H), 1.30 (s, 3 H),
1.47–1.52 (m, 1 H), 1.69 (s, 3 H), 1.88 (m, 1 H), 2.01 (d, J = 1.3 Hz,
3 H), 2.12 (s, 3 H), 2.18–2.25 (m, 1 H), 2.27 (s, 3 H), 2.37–2.58 (m,
2 H, overlapped), 3.79 (d, J = 7.1 Hz, 1 H), 4.17 (d, J = 8.4 Hz, 1
General Procedure for Acylation at C-13 of 7-TES-Baccatin III: To
a solution of 4a (70 mg, 0.32 mmol) in toluene (4 mL) was added
DCC (68 mg, 0.32 mmol). After stirring at room temperature for H), 4.30 (d, J = 8.4 Hz, 1 H), 4.44 (m, 1 H), 4.92 (dd, J = 7.6,
15 min, 4-PP (3 mg, cat.) was added and stirring was continued for
5 min before 7-TES-baccatin III (2) (75 mg, 0.11 mmol) was added.
The reaction mixture was stirred at room temperature for 24 h. It
was then diluted with ethyl acetate (50 mL). The organic phase was
washed with sodium hydrogencarbonate, water, brine, dried with
sodium sulfate, and concentrated in vacuo. The residue was puri-
fied by preparative thin layer chromatography (30% EtOAc/hex-
2.1 Hz, 1 H), 5.09 (s, 1 H), 5.68 (d, J = 7.1 Hz, 1 H), 6.26 (td, J =
8.9, 1.3 Hz, 1 H), 6.32 (s, 1 H), 7.46–7.59 (m, 5 H, Ar, overlapped),
7.64 (m, 1 H, Ar), 7.99 (m, 2 H, Ar), 8.08 (m, 2 H, Ar) ppm. 13C
NMR (100 MHz, CDCl3): δ = 9.6, 11.2, 14.0, 15.1, 20.9, 22.16,
22.20, 26.9, 35.5, 36.3, 43.2, 45.5, 53.2, 58.5, 71.8, 72.1, 75.1, 75.4,
76.3, 79.6, 80.1, 80.9, 84.4, 126.7, 128.0, 128.6, 128.7, 129.1, 130.0,
131.9, 133.2, 133.8, 141.9, 163.5, 166.9, 168.7, 169.8, 171.3, 203.6
ane) to give 7a (35 mg, 36%) and 8a (48 mg, 49%). Compounds 7b ppm. HRFABMS: calcd. for C43H48NO13 [M + H+] 786.3126,
(65%), 7c (50%), and 8b (35%) and 8c (34%) were prepared simi-
found 786.3108.
larly using EDC.
(Baccatin III)-13-yl (5R)-2-Phenylspiro[cyclopropane-1Ј,4-oxazo-
line]-5-carboxylate (10a): 1H NMR (400 MHz, CDCl3): δ = 0.96
(m, 1 H), 1.16 (s, 3 H), 1.17–1.24 (m, 1 H), 1.28 (s, 3 H), 1.25–1.34
(m, 1 H), 1.38–1.45 (m, 1 H), 1.69 (s, 3 H), 1.88 (d, J = 1.3 Hz, 3
(7-TES-baccatin III)-13-yl (5S)-2-Phenylspiro[cyclopropane-1Ј,4-ox-
azoline]-5-carboxylate (7a): H NMR (400 MHz, CDCl3): δ = 0.64
(dq, J = 7.9, 2.3 Hz, 6 H), 0.95 (t, J = 7.9 Hz, 9 H), 1.05 (m, 1 H),
1
1.18–1.36 (m, 2 H, overlapped), 1.23 (s, 3 H), 1.26 (s, 3 H), 1.71 (s, H), 1.85–1.93 (m, 1 H, overlapped), 2.20–2.30 (m, 1 H, overlapped),
3 H), 1.90 (m, 1 H), 2.13 (d, J = 1.1 Hz, 3 H), 2.20 (s, 3 H), 2.21 2.24 (s, 3 H), 2.31 (s, 3 H), 2.40–2.60 (m, 2 H, overlapped), 3.79
(s, 3 H), 2.17–2.25 (m, 1 H), 2.40–2.58 (m, 2 H, overlapped), 3.81 (d, J = 7.1 H, 1 Hz), 4.18 (d, J = 8.3 Hz, 1 H), 4.32 (d, J = 8.3 Hz,
(d, J = 7.1 Hz, 1 H), 4.17 (d, J = 8.3 Hz, 1 H), 4.31 (d, J = 8.3 Hz,
1 H), 4.48 (dd, J = 10.7, 6.6 Hz, 1 H), 4.91 (dd, J = 9.4, 1.6 Hz, 1
H), 5.07 (s, 1 H), 5.71 (d, J = 7.1 Hz, 1 H), 6.17 (dt, J = 8.9, 1.3 Hz,
1 H), 6.48 (s, 1 H), 7.44–7.58 (m, 5 H, Ar, overlapped), 7.64 (m, 1
H, Ar), 7.99 (m, 2 H, Ar), 8.09 (m, 2 H, Ar) ppm. 13C NMR
(100 MHz, CDCl3): δ = 5.5, 6.9, 10.2, 11.5, 14.3, 14.8, 21.0, 21.4,
22.5, 26.8, 36.3, 37.4, 43.5, 46.9, 53.5, 58.7, 72.1, 72.4, 75.0, 75.1,
76.6, 79.4, 80.2, 81.2, 84.3, 126.9, 128.2, 128.7, 128.8, 129.4, 130.2,
131.9, 133.9, 134.2, 139.6, 163.6, 167.2, 168.9, 169.5, 169.8, 201.8
ppm. HRFABMS: calcd. for C49H62NO13S [M + H+] 900.3990,
found: 900.3983.
1 H), 4.42 (m, 1 H), 4.99 (dd, J = 7.6, 2.0 Hz, 1 H), 5.02 (s, 1 H),
5.69 (d, J = 7.1 Hz, 1 H), 6.08 (dt, J = 8.9, 1.3 Hz, 1 H), 6.27 (s, 1
H), 7.46–7.60 (m, 5 H, Ar, overlapped), 7.64 (m, 1 H, Ar), 8.04 (m,
2 H, Ar), 8.08 (m, 2 H, Ar) ppm. 13C NMR (100 MHz, CDCl3): δ
= 9.6, 10.3, 15.00, 15.04, 20.8, 21.9, 22.0, 26.9, 35.5, 36.2, 43.2,
45.6, 53.7, 58.5, 72.1, 72.4, 75.0, 75.5, 76.4, 79.5, 80.9, 84.4, 126.7,
128.1, 128.6, 128.7, 129.1, 130.1, 131.8, 133.1, 133.8, 142.3, 163.5,
167.0, 169.0, 169.6, 171.2, 203.7 ppm. HRFABMS: calcd. for
C43H48NO13 [M + H+] 786.3126, found 786.3115.
General Procedure for Hydrolysis of the Oxazoline Ring: To a solu-
tion of compound 9a (25 mg, 0.032 mmol) in 1,4-dioxane (5 mL)
was added HCl (0.1 , 5 mL) and stirring was continued at 50 °C
for 1 h. TLC showed that all the starting material was consumed,
and a much more polar compound was formed. Then, 84 mg of
NaHCO3 powder was added at room temperature. The reaction
mixture was stirred overnight until the TLC showed the highly po-
(7-TES-baccatin III)-13-yl (5R)-2-Phenylspiro[cyclopropane-1Ј,4-
oxazoline]-5-carboxylate (8a): 1H NMR (400 MHz, CDCl3): δ =
0.57 (m, 6 H), 0.92 (t, J = 7.9 Hz, 9 H), 1.21 (s, 3 H), 1.25 (s, 3 H),
1.19–1.44 (m, 4 H, overlapped), 1.71 (s, 3 H), 1.91 (m, 1 H), 1.99
(d, J = 1.2 Hz, 3 H), 2.18 (s, 3 H), 2.19–2.29 (m, 1 H), 2.35 (s, 3
H), 2.43–2.59 (m, 2 H, overlapped), 3.80 (d, J = 7.0 Hz, 1 H), 4.17 lar intermediate was converted to two less polar compounds. The
(d, J = 8.3 Hz, 1 H), 4.34 (d, J = 8.3 Hz, 1 H), 4.47 (dd, J = 10.5, reaction mixture was diluted with EtOAc (20 mL), and then
6.6 Hz, 1 H), 4.98–5.00 (2 H, overlapped), 5.71 (d, J = 7.0 Hz, 1 washed with water, brine, and dried with sodium sulfate. The or-
H), 6.00 (dt, J = 8.9, 1.3 Hz, 1 H), 6.42 (s, 1 H), 7.45–7.57 (m, 5 ganic phase was concentrated under reduced pressure, and the resi-
H, Ar, overlapped), 7.64 (m, 1 H, Ar), 8.09 (m, 2 H, Ar), 8.11 (m, due was applied to PTLC (50% EtOAc/hexane) to give 11 (7.7 mg,
2 H, Ar) ppm. 13C NMR (100 MHz, CDCl3): δ = 5.4, 6.9, 10.2,
10.3, 14.7, 15.3, 21.0, 21.2, 22.2, 26.9, 36.2, 37.4, 43.4, 47.1, 53.9,
48%) and 12 (5.1 mg, 32%). Compounds 3a (50%), 3b (80%), 3c
(78%), 13 (34%), 14, (48%), 15 (29%), 16 (45%), and 17 (40%)
58.6, 72.5, 72.7, 75.0, 75.2, 76.6, 79.3, 79.5, 81.1, 84.3, 126.9, 128.4, were prepared by a similar procedure.
128.6, 128.8, 129.4, 130.2, 131.8, 133.9, 134.0, 140.2, 163.7, 167.2,
3Ј-(N-Debenzoyl)-3Ј-dephenyl-3Ј,3Ј-ethylene-2Ј-O-benzoyl-2Ј-epi-taxol:
169.32, 169.33, 169.7, 201.9 ppm. HRFABMS: calcd. for
1H NMR (400 MHz, CDCl3): δ = 0.82 (d, J = 10 Hz), 1.00 (s, 2
H), 1.18 (s, 3 H), 1.22 (m, 2 H), 1.27 (s, 3 H), 1.72 (s, 3 H), 1.90
(m, 4 H), 2.06 (s, 3 H) 2.28 (s, 3 H), 2.32 (m, 1 H), 2.38 (s, 3 H),
2.44 (m, 1 H), 2.60 (m, 1 H), 3.88 (d, J = 6.8 Hz, 1 H), 4.21 (d, J
C49H62NO13Si [M + H+] 900.3999, found 900.3990.
General Procedure for Silyl Deprotection: To a solution of 7a
(15.1 mg, 0.0106 mmol), in 0.6 mL of dried THF, was added
Eur. J. Org. Chem. 2005, 3962–3972
www.eurjoc.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3969

D. G. I. Kingston et al.
FULL PAPER
= 8.4 Hz, 1 H), 4.33 (d, J = 8.4 Hz, 1 H), 4.48 (dd, J = 10, 6.8 Hz, 172.3, 172.6, 207.4 ppm. HRFABMS: calcd. for C43H50NO14 [M +
1 H), 4.55 (s, 1 H), 5.0 (d, J = 8 Hz, 1 H), 5.70 (d, J = 6.8 Hz, 1
H), 6.34 (t, J = 8.8 Hz, 1 H), 6.35 (s, 1 H), 7.52 (m, 4 H), 7.65 (t,
J = 7.6 Hz), 8.14 (m, 4 H, 2 H) ppm.
H+] 804.3231, found 804.3226.
(3ЈR)-3Ј-Dephenyl-3Ј,3Ј-[(1R)-methylethylene]-2Ј-epi-taxol
(13):
1
[α]2D0 = –76 (c = 0.61). H NMR (400 MHz, CDCl3): δ = 1.07 (m,
1 H), 1.16 (s, 3 H), 1.26 (s, 3 H), 1.34–1.47 (5 H, overlapped), 1.71
(s, 3 H), 1.78 (s, 1 H), 1.87–1.96 (m, 1 H), 2.14 (d, J = 1.2 Hz, 3
H), 2.19–2.26 (m, 1 H), 2.27 (s, 3 H), 2.29–2.36 (m, 1 H, over-
lapped), 2.35 (s, 3 H), 2.54–2.64 (2 H, overlapped), 3.87 (d, J =
7.1 Hz, 1 H), 4.00 (br. s, 1 H), 4.19 (d, J = 8.4 Hz, 1 H), 4.32 (d, J
= 8.4 Hz, 1 H), 4.48 (m, 1 H), 4.99 (dd, J = 9.4, 1.8 Hz, 1 H), 5.17
(br. s, 1 H), 5.68 (d, J = 7.1 Hz, 1 H), 6.18 (dt, J = 8.9, 1.2 Hz, 1
H), 6.37 (s, 1 H), 6.68 (s, 1 H), 7.39 (m, 2 H, Ar), 7.46–7.54 (m, 3
H, Ar), 7.64 (m, 1 H, Ar), 7.69 (m, 2 H, Ar), 8.07 (m, 2 H, Ar)
ppm. 13C NMR (100 MHz, CDCl3): δ = 9.5, 14.4, 15.7, 20.9, 21.0,
21.8, 22.6, 26.7, 35.6, 36.2, 39.9, 43.1, 45.8, 58.6, 71.6, 72.3, 74.2,
74.9, 75.7, 76.4, 79.4, 81.0, 84.4, 127.0, 128.71, 128.73, 129.1, 130.0,
132.3, 132.6, 133.2, 133.8, 142.8, 166.9, 170.0, 170.4, 171.3, 172.6,
3Ј-Dephenyl-3Ј,3Ј-ethylene-2Ј-epi-taxol (11): [α]2D0 = –60 (c = 0.23).
1H NMR (400 MHz, CDCl3): δ = 1.14 (s, 3 H), 1.24 (s, 3 H), 1.05–
1.40 (m, 4 H), 1.68 (s, 3 H), 1.88 (m, 1 H), 2.12 (d, J = 1.1 Hz, 3
H), 2.15–2.35 (m, 2 H), 2.24 (s, 3 H), 2.28 (s, 3 H), 2.51–2.61 (m,
1 H), 3.68 (s, 1 H), 3.84 (d, J = 7.1 Hz, 1 H), 4.16 (d, J = 8.3 Hz,
1 H), 4.29 (d, J = 8.3 Hz, 1 H), 4.45 (dd, J = 10.7, 6.6 Hz, 1 H),
4.96 (dd, J = 7.7, 1.8 Hz, 1 H), 5.66 (d, J = 7.1 Hz, 1 H), 6.15 (td,
J = 8.9, 1.3 Hz, 1 H), 6.34 (s, 1 H), 6.66 (s, 1 H), 7.33–7.52 (m, 5
H), 7.61 (m, 1 H), 7.67 (m, 2 H), 8.05 (m, 2 H) ppm. 13C NMR
(100 MHz, CDCl3): δ = 9.7, 13.7, 14.7, 15.9, 21.1, 21.9, 22.8, 26.9,
35.8, 36.3, 36.8, 43.3, 46.0, 58.8, 72.0, 72.5, 75.1, 75.9, 76.6, 77.3,
79.6, 81.2, 84.6, 127.2, 128.93, 128.94, 129.3, 130.3, 132.5, 132.8,
133.4, 134.0, 143.0, 167.2, 170.3, 170.4, 171.5, 172.3, 204.0 ppm.
HRFABMS: calcd. for C43H50NO14 [M + H+] 804.3231, found
804.3197.
203.8 ppm. HRFABMS: calcd. for C44H52NO14 [M
818.3388, found 818.3392.
+
H+]
(3ЈR)-3Ј-Dephenyl-3Ј,3Ј-[(1R)-methylethylene]-2Ј,7-epi-taxol (14):
3Ј-Dephenyl-3Ј,3Ј-ethylene-2Ј,7-epi-taxol (12): [α]2D0 = –56 (c = 0.41).
1H NMR (400 MHz, CDCl3): δ = 1.14 (s, 3 H), 1.17 (s, 3 H), 1.07–
1.31 (m, 4 H), 1.65 (s, 3 H), 1.65–1.75 (m, 1 H), 2.00 (d, J = 1.3 Hz,
3 H), 2.19 (s, 3 H), 2.20–2.39 (m, 3 H), 2.39 (s, 3 H), 3.66–3.73 (m,
1 H), 3.72 (s, 1 H), 3.94 (d, J = 7.4 Hz, 1 H), 4.33 (d, J = 8.6 Hz,
1 H), 4.38 (d, J = 8.6 Hz, 1 H), 4.70 (d, J = 10.3 Hz, 1 H), 4.92
(dd, J = 5.7, 3.3 Hz, 1 H), 5.73 (d, J = 7.4 Hz, 1 H), 6.11 (td, J =
8.9, 1.5 Hz, 1 H), 6.71 (s, 1 H), 6.83 (s, 1 H), 7.32–7.52 (m, 5 H),
7.62 (m, 1 H), 7.68 (m, 2 H), 8.06 (m, 2 H) ppm. 13C NMR
(100 MHz, CDCl3): δ = 14.0, 14.6, 15.9, 16.4, 21.1, 21.3, 22.7, 26.2,
35.6, 36.6, 37.0, 40.6, 42.8, 57.8, 71.5, 75.4, 75.9, 77.8, 78.4, 79.5,
82.2, 82.9, 127.3, 128.96, 129.00, 129.4, 130.2, 132.6, 133.1, 133.4,
134.0, 140.3, 167.2, 169.7, 170.9, 172.1, 172.3, 207.5 ppm.
HRFABMS: calcd. for C43H49NO14Na [M + Na+] 826.3050, found
826.3050;
[α]2D0 = –58 (c = 0.29). H NMR (400 MHz, CDCl3): δ = 1.02 (t, J
1
= 6.1 Hz, 1 H), 1.17 (s, 3 H), 1.20 (s, 3 H), 1.28–1.33 (m, 1 H),
1.39–1.47 (m, 1 H), 1.47 (d, J = 5.7 Hz, 3 H), 1.68 (s, 3 H), 1.74
(s, 1 H), 2.02 (d, J = 1.3 Hz, 3 H), 2.22 (s, 3 H), 2.24–2.42 (4 H,
overlapped), 2.45 (s, 3 H), 3.73 (m, 1 H), 3.98 (d, J = 7.4 Hz, 1 H),
4.03 (d, J = 6.7 Hz, 1 H), 4.34 (d, J = 8.7 Hz, 1 H), 4.41 (d, J =
8.7 Hz, 1 H), 4.74 (d, J = 11.6 Hz, 1 H), 4.96 (dd, J = 8.9, 3.3 Hz,
1 H), 5.68 (d, J = 6.7 Hz, 1 H), 5.76 (d, J = 7.4 Hz, 1 H), 6.16 (dt,
J = 8.7, 1.3 Hz, 1 H), 6.69 (s, 1 H), 6.87 (s, 1 H), 7.39 (m, 2 H,
Ar), 7.48–7.54 (m, 3 H, Ar, overlapped), 7.66 (m, 1 H, Ar), 7.71
(m, 2 H, Ar), 8.09 (m, 2 H, Ar) ppm. 13C NMR (100 MHz, CDCl3):
δ = 14.5, 15.7, 16.1, 20.9, 21.1, 21.3, 22.2, 22.6, 26.0, 35.4, 36.5,
40.1, 40.4, 42.6, 57.6, 71.0, 75.0, 75.2, 75.7, 77.6, 78.2, 79.3, 82.1,
82.7, 127.1, 128.76, 128.78, 129.2, 130.0, 132.4, 132.9, 133.1, 133.8,
140.2, 167.0, 169.4, 171.0, 172.1, 172.2, 207.3 ppm. HRFABMS:
calcd. for C44H52NO14 [M + H+] 818.3388, found 818.3405.
3Ј-Dephenyl-3Ј,3Ј-ethylenetaxol (3a): [α]2D0 = –68 (c = 0.30). 1H
NMR (400 MHz, CDCl3): δ = 1.15 (s, 3 H), 1.27 (s, 3 H), 1.07–
1.35 (m, 4 H), 1.69 (s, 3 H), 1.79 (d, J = 1.3 Hz, 3 H), 1.88 (m, 1
H), 2.23 (s, 3 H), 2.25–2.38 (m, 2 H), 2.46 (s, 3 H), 2.57 (m, 1 H),
3.82 (d, J = 7.1 Hz, 1 H), 3.91 (s, 1 H), 4.18 (d, J = 8.5 Hz, 1 H)
4.33 (d, J = 8.5 Hz, 1 H), 4.44 (dd, J = 10.8, 6.7 Hz, 1 H), 4.99
(dd, J = 7.5, 2.1 Hz, 1 H), 5.68 (d, J = 7.1 Hz, 1 H), 6.24 (td, J =
9.2, 1.5 Hz, 1 H), 6.25 (s, 1 H) 6.83 (s, 1 H), 7.44–7.60 (m, 5 H),
7.64 (m, 1 H), 7.74 (m, 2 H), 8.10 (m, 2 H) ppm. 13C NMR
(100 MHz, CDCl3): δ = 9.6, 13.6, 14.3, 14.8, 20.9, 22.01, 22.04,
26.7, 35.5, 35.9, 37.2, 43.2, 45.6, 58.5, 71.0, 72.1, 75.1, 75.6, 76.4,
77.9, 79.5, 80.7, 84.5, 127.2,128.7, 128.9, 129.2, 129.3, 130.1,
132.59, 132.60, 132.7, 132.8, 133.8, 142.8, 167.0, 170.6, 170.8,
171.2, 172.3, 203.8, 207.4 ppm. HRFABMS: calcd. for C43H50NO14
[M + H+] 804.3231, found 804.3239.
(3ЈS)-3Ј-Dephenyl-3Ј,3Ј-[(1S)-methylethylene]taxol (3b): [α]2D0 = –57
1
(c = 0.43). H NMR (400 MHz, CDCl3): δ = 0.93 (t,, J = 6.1 Hz 1
H), 1.11 (s, 3 H), 1.20–1.26 (4 H, overlapped), 1.33–1.41 (m, 1 H),
1.43 (d, J = 6.1 Hz, 3 H), 1.65 (s, 3 H), 1.71 (d, J = 1.3 Hz, 3 H),
1.79 (s, 1 H), 1.80–1.89 (m, 1 H), 2.19 (s, 3 H), 2.21–2.36 (m, 2 H,
overlapped), 2.43 (d, J = 4.2 Hz, 1 H), 2.45 (s, 3 H), 2.48–2.58 (m,
1 H), 3.78 (d, J = 7.1 Hz, 1 H), 4.08 (d, J = 7.1 Hz, 1 H), 4.15 (d,
J = 8.5 Hz, 1 H), 4.30 (d, J = 8.5 Hz, 1 H), 4.40 (m, 1 H), 4.96
(dd, J = 9.6, 2.0 Hz, 1 H), 5.65 (d, J = 7.1 Hz, 1 H), 5.94 (d, J =
7.1 Hz, 1 H), 6.19 (s, 1 H), 6.22 (dt, J = 8.9, 1.3 Hz, 1 H), 6.75 (s,
1 H), 7.42–7.64 (m, 6 H, Ar, overlapped), 7.71 (m, 2 H, Ar), 8.08
(m, 2 H, Ar) ppm. 13C NMR (100 MHz, CDCl3): δ = 9.8, 14.9,
15.0, 21.1, 22.0, 22.2, 22.3, 22.4, 27.0, 35.7, 36.1, 40.9, 43.4, 45.8,
58.6, 71.2, 72.3, 75.4, 75.8, 76.4, 76.6, 79.8, 80.8, 84.7, 127.4, 128.9,
127.1, 129.4, 130.3, 132.7, 132.8, 132.9, 134.0, 143.1, 167.2, 170.8,
171.4, 171.5, 173.0, 204.1 ppm. HRFABMS: calcd. for C44H52NO14
[M + H+] 818.3388, found 818.3392.
3Ј-Dephenyl-3Ј,3Ј-ethylene-7-epi-taxol (17): [α]2D0 = –70 (c = 0.24).
1H NMR (400 MHz, CDCl3): δ = 1.16 (s, 3 H), 1.22 (s, 3 H), 1.08–
1.37 (m, 4 H) 1.66 (s, 3 H), 1.71 (d, J = 1.4 Hz, 3 H), 2.18 (s, 3 H),
2.23–2.41 (m, 3 H), 2.57 (s, 3 H), 3.69 (m, 1 H), 3.89 (s, 1 H), 3.92
(d, J = 7.5 Hz, 1 H), 4.35 (d, J = 8.5 Hz, 1 H), 4.41 (d, J = 8.5 Hz,
1 H), 4.78 (d, J = 11.7 Hz, 1 H), 4.96 (dd, J = 5.7, 3.4 Hz, 1 H),
5.76 (d, J = 7.5 Hz, 1 H), 6.21 (td, J = 9.0, 1.4 Hz, 1 H), 6.74 (s, 1
(3ЈR)-3Ј-Dephenyl-3Ј,3Ј-[(1R)-isopropylethylene]-2Ј-epi-taxol (15):
1
[α]2D0 = –74 (c = 0.54). H NMR (400 MHz, CDCl3): δ = 0.98–1.16
(8 H, overlapped), 1.23 (s, 3 H), 1.27–1.39 (5 H, overlapped), 1.88
(m, 1 H), 2.10 (d, J = 1.1 Hz, 3 H), 2.15–2.33 (m, 2 H, overlapped),
H) 6.80 (s, 1 H) 7.44–7.60 (m, 5 H) 7.65 (m, 1 H), 7.73 (m, 2 H), 2.23 (s, 3 H), 2.30 (s, 3 H), 2.56 (m, 1 H), 3.84 (d, J = 7.1 Hz, 1
8.12 (m, 2 H) ppm. 13C NMR (100 MHz, CDCl3): δ = 13.8, 14.6, H), 3.95 (d, J = 5.5 Hz, 1 H), 4.16 (d, J = 8.3 Hz, 1 H), 4.28 (d, J
14.8, 16.2, 20.9, 21.3, 22.1, 25.9, 35.3, 36.2, 37.5, 40.4, 42.7, 57.6,
70.9, 75.4, 75.7, 77.5, 78.2, 78.6, 79.5, 81.6, 82.8, 127.2, 128.7,
128.9, 129.3, 130.1, 132.5, 132.7, 133.8, 140.3, 167.1, 169.3, 171.2,
= 8.3 Hz, 1 H), 4.45 (dd, J = 10.8, 6.8 Hz, 1 H), 4.96 (dd, J = 9.5,
1.9 Hz, 1 H), 5.17 (d, J = 6.1 Hz, 1 H), 5.65 (d, J = 7.1 Hz, 1 H),
6.16 (dt, J = 8.9, 1.1 Hz, 1 H), 6.33 (s, 1 H), 6.60 (s, 1 H), 7.36 (m,
3970
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2005, 3962–3972

C-3Ј-Cyclopropanated Taxol Analogs
FULL PAPER
[5] Reviews see: a) J. Salaün, M. S. Baird, Curr. Med. Chem. 1995,
2, 511–542; b) J. Salaün, in: Topics in Current Chemistry, vol.
207 (Ed.: A. de Meijere), Springer, Heidelberg, 2000, p. 1–67.
[6] W. A. Donaldson, Tetrahedron 2001, 57, 8589–8627.
[7] I. Ojima, S. Lin, J. Org. Chem. 1998, 63, 224–225.
[8] J. Johnson, S.-H. Kim, M. Bifano, J. DiMarco, C. Fairchild, J.
Gougoutas, F. Lee, B. Long, J. Tokarski, G. Vite, Organic Lett.
2000, 2, 1537–1540.
[9] M. W. Nötzel, M. Tamm, T. Labahn, M. Noltemeyer, M. Es-
Sayed, A. de Meijere, J. Org. Chem. 2000, 65, 3850–3852.
[10] D. G. I. Kingston, A. G. Chaudhary, A. A. L. Gunatilaka,
M. L. Middleton, Tetrahedron Lett. 1994, 35, 4483–4484.
[11] J. P. Snyder, J. H. Nettles, B. Cornett, K. H. Downing, E. No-
gales, Proc. Natl. Acad. Sci. USA 2001, 98, 5312–5316.
[12] a) T. Ganesh, R. C. Guza, S. Bane, R. Ravindra, N. Shanker,
A. S. Lakdawala, J. P. Snyder, D. G. I. Kingston, Proc. Nat.
Acad. Sci. USA 2004, 101, 1006–10011; b) D. G. I. Kingston,
S. Bane, J. P. Snyder, Cell Cycle 2005, 4, 279–289.
[13] O. Querolle, J. Dubois, S. Thoret, F. Roussi, F. Guéritte, D. J.
Guénard, J. Med. Chem. 2004, 47, 5937–5944.
[14] R. Geney, L. Sun, P. Pera, R. J. Bernacki, S. Xia, S. B. Horwitz,
C. L. Simmerling, I. Ojima, Chem. Biol. 2005, 12, 339–348.
[15] G. Barnes, K. A. Louie, D. Botstein, Mol. Biol. Cell 1992, 3,
29–47.
[16] a) C. J. Bode, M. L. Gupta, Jr., E. A. Reiff, K. A. Suprenant,
G. I. Georg, R. H. Himes, Biochemistry 2002, 41, 3870–3874;
b) M. L. Gupta, C. J. Bode, G. I. Georg, R. H. Himes, Proc.
Natl. Acad. Sci. USA 2003, 100, 6394–6397.
[17] M. Hari, F. Loganzo, T. Annable, X. Tan, S. Musto, D. B. Mor-
illa, J. H. Nettles, J. P. Snyder, L. M. Greenberger, submitted.
2 H, Ar), 7.42–7.52 (m, 3 H, Ar, overlapped), 7.60 (m, 1 H, Ar),
7.67 (m, 2 H, Ar), 8.04 (m, 2 H, Ar) ppm. 13C NMR (100 MHz,
CDCl3): δ = 9.7, 15.9, 18.8, 21.1, 22.0, 22.8, 22.9, 23.0, 26.9, 29.8,
35.8, 36.3, 36.5, 40.8, 43.3, 46.0, 58.8, 71.8, 72.5, 74.3, 75.1, 75.9,
76.6, 79.6, 81.3, 84.6, 127.3, 128.91, 128.92, 129.3, 130.2, 132.5,
132.7, 133.5, 134.0, 143.0, 167.1, 170.3, 170.5, 171.5, 172.9, 204.0
ppm. HRFABMS: calcd. for C46H56NO14 [M + H+] 846.3701,
found 846.3725.
(3ЈR)-3Ј-Dephenyl-3Ј,3Ј-[(1R)-isopropylethylene]-2Ј,7-epi-taxol (16):
[α]2D0 = –58 (c = 0.17). H NMR (400 MHz, CDCl3): δ = 0.94 (m,
1
1 H), 1.11 (d, J = 6.1 Hz, 3 H), 1.13 (s, 3 H), 1.08–1.14 (m, 1
H, overlapped), 1.16 (s, 3 H), 1.20–1.26 (m, 1 H), 1.32–1.42 (4 H,
overlapped), 1.62 (s, 1 H), 1.65 (s, 3 H), 1.98 (d, J = 1.0 Hz, 3 H),
2.10–2.45 (4h, overlapped), 2.19 (s, 3 H), 2.40 (s, 3 H), 3.69 (m, 1
H), 3.94 (d, J = 7.6 Hz, 1 H), 3.98 (s, 1 H), 4.33 (d, J = 8.7 Hz, 1
H), 4.38 (d, J = 8.7 Hz, 1 H), 4.93 (dd, J = 9.1, 3.4 Hz, 1 H), 5.72
(d, J = 7.6 Hz, 1 H), 6.13 (dt, J = 8.8, 1.0 Hz, 1 H), 6.67 (s, 1 H),
6.82 (s, 1 H), 7.36 (m, 2 H, Ar), 7.44–7.51 (m, 3 H, Ar, overlapped),
7.62 (m, 1 H, Ar), 7.68 (m, 2 H, Ar), 8.05 (m, 2 H, Ar) ppm. 13C
NMR (100 MHz, CDCl3): δ = 15.9, 16.3, 19.1, 21.1, 21.3, 22.77,
22.82, 23.1, 26.1, 29.8, 35.6, 36.6, 36.8, 40.6, 41.0, 42.8, 57.8, 71.2,
75.1, 75.4, 75.9, 77.8, 78.4, 79.5, 82.3, 82.9, 127.3, 128.95, 128.98,
129.4, 130.2, 132.6, 133.2, 134.0, 140.4, 167.2, 169.6, 171.1, 172.3,
172.5, 207.5 ppm. HRFABMS: calcd. for C46H56NO14 [M + H+]
846.3701, found 846.3725.
(3ЈS)-3Ј-Dephenyl-3Ј,3Ј-[(1S)-isopropylethylene]-taxol (3c): [α]2D0
=
–66 (c = 0.66). 1H NMR (400 MHz, CDCl3): δ = 0.89 (m, 1 H),
1.07 (d, J = 6.2 Hz, 3 H), 1.10 (s, 3 H), 1.04–1.12 (m, 1 H, over-
lapped), 1.15–1.20 (m, 1 H), 1.23 (s, 3 H), 1.32–1.42 (4 H, over-
lapped), 1.65 (s, 3 H), 1.70 (s, 1 H), 1.71 (d, J = 1.1 Hz, 3 H), 1.80–
1.88 (m, 1 H), 2.19 (s, 3 H), 2.20–2.38 (m, 2 H, overlapped), 2.43
(s, 3 H), 2.27–2.57 (m, 1 H), 3.77 (d, J = 7.1 Hz, 1 H), 4.06 (s, 1
H), 4.15 (d, J = 8.3 Hz, 1 H), 4.29 (d, J = 8.3 Hz, 1 H), 4.39 (dd,
J = 10.9, 6.6 Hz, 1 H), 4.95 (d, J = 9.6, 1.9 Hz, 1 H), 5.64 (d, J =
7.1 Hz, 1 H), 6.18 (s, 1 H), 6.20 (dt, J = 9.0, 1.1 Hz, 1 H), 6.79 (s,
1 H), 7.40–7.64 (m, 6 H, Ar, overlapped), 7.71 (m, 2 H, Ar), 8.07
(m, 2 H, Ar) ppm. 13C NMR (100 MHz, CDCl3): δ = 9.6, 14.7,
19.3, 20.9, 22.0, 22.2, 22.6, 23.1, 26.7, 29.6, 35.5, 35.9, 36.5, 41.5,
43.2, 45.6, 58.5, 71.0, 72.0, 75.2, 75.6, 76.1, 76.4, 79.6, 80.6, 84.5,
127.2, 128.7, 128.9, 129.2, 130.1, 132.5, 133.7, 142.9, 167.0, 170.5,
171.18, 171.24, 172.9, 203.9 ppm. HRFABMS: calcd. for
C46H56NO14 [M + H+] 846.3701, found 846.3708.
[18]
[19]
T. Liese, S. Teichmann, A. de Meijere, Synthesis 1988, 25–32.
W. Cabri, M. Curini, M. C. Marcotullio, O. Rosati, Tetrahedron
Lett. 1996, 37, 4785–4786.
[20]
[21]
[22]
[23]
[24]
[25]
G. N. Chmurny, B. D. Hilton, S. Brobst, S. A. Look, K. M.
Witherup, J. A. Beutler, J. Nat. Prod. 1992, 55, 414–423.
A. G. Chaudhary, J. M. Rimoldi, D. G. I. Kingston, J. Org.
Chem. 1993, 58, 3798–3799.
R. Riguera, E. Quiñoá, J. M. Seco, Tetrahedron: Asymmetry
2001, 12, 2915–2925.
F. Gueritte-Voegelein, D. Guenard, F. Lavelle, M.-T. Le Goff,
L. Mangatal, P. Potier, J. Med. Chem. 1991, 34, 992–998.
E. Nogales, S. G. Wolf, K. H. Downing, Nature 1998, 391, 199–
203.
a) S. Rao, N. E. Krauss, J. M. Heerding, C. S. Swindell, I. Rin-
gel, G. A. Orr, S. B. Horwitz, J. Biol. Chem. 1994, 269, 3132–
3134; b) S. Rao, G. A. Orr, A. G. Chaudhary, D. G. I. Kings-
ton, S. B. Horwitz, J. Biol. Chem. 1995, 270, 20235–20238; c)
C. Combeau, A. Commercon, C. Mioskowski, B. Rousseau, F.
Aubert, M. Goeldner, Biochemistry 1994, 33, 6676–6683.
T. Ganesh, R. C. Guza, S. Bane, R. Ravindra, N. Shanker, A. S.
Lakdawala, J. P. Snyder, D. G. I. Kingston, Proc. Natl. Acad.
Sci. USA 2004, 101, 10006–10011.
[1] The chemical compound represented by structure 1 was named
taxol in 1971 by its discoverers, but the name taxol was sub-
sequently registered as a trademark by Bristol-Myers Squibb
to refer to their formulation of taxol. The original chemical
name taxol is used in this paper to refer to compound 1; no
infringement of the Bristol–Myers Squibb trademark is im-
plied.
[26]
[27]
[28]
A. Bondi, J. Phys. Chem. 1964, 68, 441–451.
C. Lucatelli, F. Viton, Y. Gimbert, A. E. Greene, J. Org. Chem.
2002, 67, 9468–9470.
[29]
G. I. Georg, G. C. B. Harriman, D. G. Vander Velde, T. C.
Boge, Z. S. Cheruvallath, A. Datta, M. Hepperle, H. Park,
R. H. Himes, L. Jayasinghe, in: Taxane Anticancer Agents
(Eds.: G. I. Georg, T. T. Chen, I. Ojima, D. M. Vyas), Ameri-
can Chemical Society, Washington, 1995, p. 217–232.
I. Ojima, Y. H. Park, I. Fenoglio, O. Duclos, C.-M. Sun, S. D.
Kuduk, M. Zucco, G. Appendino, P. Pera, J. M. Veith, R. J.
Bernacki, M.-C. Bissery, C. Combeau, P. Vrignaud, J. G. Riou,
F. Lavelle, in: Taxane Anticancer Agents: Basic Science and
Current Status, ACS Symposium Series vol. 583 (Eds.: G. I.
Georg, T. T. Chen, I. Ojima, D. M. Vyas), American Chemical
Society, Washington, DC, 1995, p. 262–275.
[2] M. C. Wani, H. L. Taylor, M. E. Wall, P. Coggon, A. T.
McPhail, J. Am. Chem. Soc. 1971, 93, 2325–2327.
[3] a) F. A. Holmes, R. S. Walters, R. L. Theriault, A. D. Forman,
L. K. Newton, M. N. Raber, A. U. Buzdar, D. K. Frye, G. N.
Hortobagyi, J. Nat. Cancer Inst. 1991, 83, 1797–1805; b) M.
Markman, Yale J. Biol. Med. 1991, 64, 583–590; c) S. M.
Swain, S. F. Honig, M. C. Tefft, L. Walton, Invest. New Drugs
1995, 13, 217–222.
[4] a) D. G. I. Kingston, P. G. Jagtap, H. Yuan, L. Samala, in: Pro-
gress in the Chemistry of Organic Natural Products, vol. 84
(Eds.:W. Herz, H. Falk, G. W. Kirby), Springer, Wien, 2002, p.
53–225; b) D. G. I. Kingston, in: Antitumor Agents from Natu-
ral Sources (Eds.: G. M. Cragg, D. G. I. Kingston, D. J. New-
man), Taylor and Francis, London, 2005; c) R. Geney, J. Chen,
I. Ojima, Med. Chem. 2005, 1, 125–139.
[30]
[31]
L. Li, S. A. Thomas, L. L. Klein, C. M. Yeung, C. J. Maring,
D. J. Grampovnik, P. A. Lartey, J. J. Plattner, J. Med. Chem.
1994, 37, 2655–2663.
Eur. J. Org. Chem. 2005, 3962–3972
www.eurjoc.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3971

D. G. I. Kingston et al.
FULL PAPER
[32] a) J.-N. Denis, A. Fkyerat, Y. Gimbert, C. Coutterez, P. Man-
tellier, S. Jost, A. E. Greene, J. Chem. Soc. Perkin Trans. 1 1995,
1811–1815; b) J. Kant, W. S. Schwartz, C. Fairchild, Q. Gao,
S. Huang, B. H. Long, J. F. Kadow, D. R. Langley, V. Farina,
D. Vyas, Tetrahedron Lett. 1996, 37, 6495–6498.
[33] L. Barboni, C. Lambertucci, G. Appendino, D. G.
Vander Velde, R. H. Himes, E. Bombardelli, M. Wang, J. P.
Snyder, J. Med. Chem. 2001, 44, 1576–1587.
[34] B. B. Metaferia, J. Hoch, T. E. Glass, S. L. Bane, S. K. Chat-
terjee, J. P. Snyder, A. Lakdawala, B. Cornett, D. G. I. Kings-
ton, Org. Lett. 2001, 3, 2461–2464.
[35] K. A. Neidigh, M. M. Gharpure, J. M. Rimoldi, D. G. I.
Kingston, Y. Q. Jiang, E. Hamel, Tetrahedron Lett. 1994, 35,
6839–6842.
[36] S.-H. Chen, J.-M. Wei, B. H. Long, C. A. Fairchild, J. Carboni,
S. W. Mamber, W. C. Rose, K. Johnston, A. M. Casazza, J. F.
Kadow, V. Farina, D. Vyas, T. W. Doyle, Bioorg. Med. Chem.
Lett. 1995, 5, 2741–2746.
[37] M. D. Chordia, A. G. Chaudhary, D. G. I. Kingston, Y. Q. Ji-
ang, E. Hamel, Tetrahedron Lett. 1994, 35, 6843–6846.
[38] S.-H. Chen, Tetrahedron Lett. 1996, 37, 3935–3938.
[39] J. P. Snyder, N. Nevins, D. O. Cicero, J. Jansen, J. Am. Chem.
Soc. 2000, 122, 724–725.
[40] L. Barboni, A. Datta, D. Dutta, G. I. Georg, D. G.
Vander Velde, R. H. Himes, M. Wang, J. P. Snyder, J. Org.
Chem. 2001, 66, 3321–3329.
[41] D. O. Cicero, G. Barbato, R. Bazzo, J. Am. Chem. Soc. 1995,
117, 1027–1033.
[42] N. Nevins, D. Cicero, J. P. Snyder, J. Org. Chem. 1999, 64,
3979–3986.
[43] a) J. Dubois, D. Guenard, F. Gueritte-Voeglein, N. Guedira, P.
Potier, B. Gillet, J.-C. Betoeil, Tetrahedron 1993, 49, 6533–6544;
b) H. J. Williams, A. I. Scott, R. A. Dieden, C. S. Swindell,
L. E. Chirlian, M. M. Francl, J. M. Heerding, N. E. Krauss,
Tetrahedron 1993, 49, 6545–6560; c) R. E. Cachau, R. Gussio,
J. A. Beutler, G. N. Chmurny, B. D. Hilton, G. M. Muschik,
J. W. Erickson, Supercomputer Applications and High Perform-
ance Computing 1994, 8, 24–34.
[44] C. Glück, V. Poignee, H. Schwager, Synthesis 1987, 260–262.
[45] T. Liese, F. Jaekel, A. de Meijere, Org. Synth. 1990, 69, 144–
153.
Received: April 6, 2005
Published Online: August 4, 2005
3972
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2005, 3962–3972