вход по аккаунту


On the Remarkable Antitumor Properties of Fludelone How We Got There.

код для вставкиСкачать
S. J. Danishefsky et al.
Epothilone Derivatives
On the Remarkable Antitumor Properties of Fludelone:
How We Got There**
Alexey Rivkin, Ting-Chao Chou, and Samuel J. Danishefsky*
anticancer drugs · drug development · epothilones ·
natural products · total synthesis
Dedicated to Professor George Olah
mall-molecule natural products are presumably often biosynthesized with a view to optimizing their ability to bind to strategic proteins
or other biomolecular targets. Although the ultimate setting in which a
drug must function may be very different, the use of such natural
products as lead compounds can serve as a significant head start in the
hunt for new agents of clinical value. Herein we reveal the synergistic
relationship between chemical synthesis and drug optimization in the
context of our research program around the epothilones: how
synthesis led to the discovery of more-potent epothilone derivatives,
and discovery inspired the development of new synthetic routes, thus
demonstrating the value of target-directed total synthesis in the quest
for new substances of material clinical benefit.
1. Introduction
Nature, during the course of evolution, has provided
prospectors with a diverse and formidable collection of
structurally complex, small-molecule natural products. Indeed, it is likely that the many structures that have been
discovered to date represent only a miniscule fraction of what
nature, in principle, has to offer.[1]
Although the field of natural products is certainly
cluttered with uninspiring “me too” entries, from time to
time those busily engaged in the isolation, biological screening, purification, and structure determination of collections of
[*] Dr. A. Rivkin, Prof. S. J. Danishefsky
Laboratory for Bioorganic Chemistry
Sloan-Kettering Institute for Cancer Research
1275 York Avenue, New York, NY 10021 (USA)
Fax: (+ 1) 212-772-8691
Prof. S. J. Danishefsky
Department of Chemistry
Columbia University, Havemeyer Hall
3000 Broadway, New York, NY 10027 (USA)
Dr. T.-C. Chou
Preclinical Pharmacology Core Facility
Sloan-Kettering Institute for Cancer Research
1275 York Avenue, New York, NY 10021 (USA)
[**] This Minireview is dedicated to Professor George Olah for his
pioneering contributions to organofluorine chemistry.
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
natural products are rewarded. They
may well discover architecturally truly
novel compounds, whose properties
prompt the launching of new directions in biological and even clinical
To bring forth a new drug capable
of providing real patient benefit while meeting the appropriately discerning standards of regulatory agencies is a daunting
task characterized by high failure rates. Those committed to
drug discovery cant help but be sensitive to the staggering set
of risk factors that bestride the complex path from concept to
approved, valuable drug. The discovery and study of smallmolecule natural products has been a productive setting in the
discovery of drugs.[3] Clearly we are far from understanding
the evolutionary forces and developmental advantages in the
biosynthesis of small molecules in plants, corals, bacteria,
fungi, and higher organisms. At least for the present, it can be
said that leads from small molecule natural product structures
(for example, steroids, prostaglandins, b-lactams, polyketides,
aminoglycosides, and statins) allow one to enter the intimidating arena of drug discovery at a later stage in the
development progression than other modalities, including
the mass screening of pharmaceutical sample collections and
commercially available combinatorial libraries.
Why do such small-molecule natural product structures
present such an advantage? Presumably they are often
biosynthesized with a view to optimizing their ability to bind
to strategic proteins or, in some instances, other biomolecular
targets. Binding to strategic proteins is a key feature of
virtually all drugs. Aside from their often exquisitely novel
structural motifs, honed and fine-tuned by evolution for their
interactivity with pertinent biomolecules, small-molecule
natural products come with the built-in advantage that, by
definition, they have been maintained in some biologically
DOI: 10.1002/anie.200461751
Angew. Chem. Int. Ed. 2005, 44, 2838 –2850
Epothilone Derivatives
viable entity. This genealogy, however different in context
from the ultimate setting in which a drug must function, and
however imperfect a marker it is for predicting pharmacodynamics and safety profiles, can already be seen as a significant
head start in the hunt for new agents of clinical value.
2. Natural Product Anticancer Agents: From Taxol
to the Epothilones
2.1. Total Synthesis
Natural products have made a huge impact on drug
discovery, particularly in the quest for new antitumor agents.[4]
A well-known example of a clinically important agent,
initially discovered through the screening of plant natural
products, is taxol (5).[5] The value of taxol (subsequently
renamed paclitaxel) in the treatment of several types of earlystage cancers and even some metastatic tumors is well
established.[6] Also of established usefulness is a semisynthetic
variant of taxol termed taxotere.[7] Both compounds inhibit
microtubule depolymerization, which is an essential phase in
mitosis. Other semisynthetic taxoids are at various stages of
The involvement of our research group in the study of
taxol was initially inspired largely by chemical interests.[8] It
has been our custom to make use of the rich structural
diversity of small, biologically generated natural products to
organize and focus our thoughts about the strategy and
methodology of chemical synthesis. In the case of taxol, we
sought to take advantage of smooth access to either enantiomer of the Wieland–Miescher ketone (1; Scheme 1). Moreover, we hoped to integrate the various stereochemical biases
within the Wieland–Miescher ketone with a suitable degradative strategy to facilitate access to baccatin III (4), a
precursor to taxol. In the event, a key transformation (2!
3), masterminded by Dr. John Masters, did occur. However,
the confidence and optimism occasioned by the stunning
utilization of an intramolecular Heck reaction by Masters
soon gave way to near desperation, as the route from 3 to 4
proved exceptionally troublesome as a result of difficulties
associated with selectively excising the external methylene
group. It was only the imaginative persistence of Dr. Wendy
Young and Dr. J. T. Link, along with that of Masters, that
allowed us to chart and realize a path to reach the natural
Scheme 1. Synthesis of taxol. Bn = benzyl, Bz = benzoyl, Tf = trifluoromethanesulfonyl, TBS = tert-butyldimethylsilyl.
product. We note, parenthetically, that for all of the frustrations and disappointments of the total synthesis, we were able
to deliver the appropriate enantiomer of taxol (5) without the
need for resolution and without recourse to relay synthesis (in
which a fragment of the natural product is used as a starting
material). These particular conditions had not been met in the
two total syntheses of taxol completed previous to our
own.[9, 10]
Accomplishments on a chemistry level notwithstanding,
our taxol effort did not deliver on one of its central purposes.
Initially, we had hoped to synthesize late-stage intermediates
containing the requisite side chain and housing the pharmacophores critical for the bioactivity of taxol. The complications of the final steps of the total synthesis left us with
precious little material (and energy!) for structure–activity
This failing at the discovery level was particularly
unfortunate since taxol is far from an ideal drug.[11] Although
cytotoxic side effects are not unexpected in a tubulintargeting agent, taxol carries two additional liabilities. The
first arises from difficulties associated with its formulation.
These problems are overcome by recourse to various delivery
Alexey Rivkin received his BS in chemistry
and biochemistry from the University of California, San Diego in 1996. He completed
his PhD in 2001 at the University of Pittsburgh, where he focused on synthetic studies
of the structurally intriguing natural product
penitrem D under the guidance of Dennis P.
Curran. He then worked as an NIH postdoctoral fellow with Samuel Danishefsky at
the Memorial Sloan-Kettering Cancer Center, where his synthetic studies of epothilones led to the discovery and development
of the (E)-9,10-dehydroepothilones. He is
currently a medicinal chemist at Merck.
Angew. Chem. Int. Ed. 2005, 44, 2838 –2850
Ting-Chao Chou completed his PhD at Yale
University and is currently director of the
Preclinical Pharmacology Core Laboratory
at the Memorial Sloan-Kettering Cancer
Center in New York. He was Professor of
Pharmacology at Cornell University from
1988 to 2000, and has been honorary professor at the Chinese Academy of Medical
Sciences since 1993. The median-effect
equation he created and the combinationindex equation he created with P. Talalay,
along with the computer software Biosoft,
UK, have received over 2000 citations in
over 250 biomedical journals.
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
S. J. Danishefsky et al.
vehicles, which create their own safety and tolerance issues in
real treatment settings. Perhaps even more serious is the
vulnerability of paclitaxel to disablement at the clinical level
by innate or acquired multidrug resistance (MDR). All too
often it proves impossible to overcome MDR by increasing
the dosage of the toxic drug. These complications in the
aggregate often lead to “taxol failure” in the clinic with
reemergence of the disease in metastatic form.
It was in this setting of seeking “life after taxol” that we
heard a report about a new anticancer agent, which functioned by the taxol mechanism. The epothilone family of
natural products had been discovered at the Gesellschaft fr
Biotechnologische Forschung mbH (GBF) by Hfle and
associates, following an extremely well-executed pharmacognosy program. The Hfle group had postulated possible
applications of epothilones in various agricultural contexts.[12]
Broad multidisciplinary interest in the epothilones was
aroused following a milestone publication by a natural
products discovery group led by Bollag at Merck.[13] The
features of the Merck disclosure that were most provocative
were that the new family of natural products, known as
epothilones (Scheme 2), seemed to owe their cytotoxic
Scheme 2. Structures of epothilones.
properties against various cancer cell lines to the same
tubulin-interference mechanism exhibited by taxol. However,
importantly, unlike taxol, epothilones A (6) and B (7) seemed
to be remarkably effective against apparent MDR cell
lines.[14] As discussed above, taxol failure is often attributable
to the onset of MDR. Although the absolute configuration of
the epothilones was not known at that time, their overall
structures suggested that they would constitute a more
Samuel Danishefsky completed his BS at
Yeshiva University in 1956 and his PhD at
Harvard University with Peter Yates. After
postdoctoral studies at Columbia University
with Gilbert Stork he began his independent
academic career in 1963 at the University
of Pittsburgh, where he became professor in
1971. In 1980, he moved to Yale University,
but returned to New York in 1993 as Professor of Chemistry at Columbia University
and Kettering Professor at the Memorial
Sloan-Kettering Cancer Center. His research
interests include synthetic strategy, reagent
development, cytotoxic natural products, and fully synthetic carbohydratebased tumor antigens.
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
manageable challenge from the point of view of chemical
synthesis than taxol derivatives. Accordingly, we hoped that
total synthesis as a means of drug discovery would be more
productive with the epothilones than had been the case with
the taxol program.
A focused effort directed towards the synthesis of the
epothilones by chemical means could not be launched until
their configurations were defined. The Merck group was
apparently not pursuing the full definition of configuration.[13]
As noted above,[12] the actual discovery of the epothilones
went back to pioneering studies by Hfle et al. However, it
was not until a key publication, again from the research group
of Hfle, that the configurations of epothilones A and B were
revealed.[15] This information led to a disciplined effort
towards total synthesis with an ultimate view to biological
investigation. Indeed, since we had no access to epothilones in
any quantity for independent investigations, total synthesis
would be our only recourse. We identified epothilones A (6)
and B (7) as our targets for total synthesis. We started by
targeting epothilone A, which lacks the methyl group at C12,
in the hope that lessons learned in that foray would serve us
well in the synthesis of epothilone B, which was reported to be
the more potent compound.[16]
Herein we relate, in a retrospective way, the interactive
relationship between chemical synthesis and drug optimization in our research program around the epothilones. It was
this synergism that led to the discovery of the remarkable (E)9,10-dehydro-12,13-desoxyepothilones 8 and 9, particularly
the congener 9 bearing a trifluoromethyl group rather than a
methyl group at C12.[17] We describe how this discovery
flowed from a unique synergy between chemistry and
preclinical experimental pharmacology. We have reason to
believe that this type of interdisciplinary interaction offers
excellent prospects in a variety of projects.
It is important to emphasize a matter of policy that
influenced the design of our syntheses to a considerable
extent. From the outset, our orienting goal was to produce
adequate quantities of preclinical lead compounds to carry
out in vivo evaluations in xenograft mouse models. We
assigned to this goal a far higher degree of urgency than to the
mass screening of individual compounds in vitro. Since our
resources were clearly finite, the decision to place emphasis
on evaluations of in vivo efficacy implicitly meant that we
would make fewer compounds, albeit in larger quantities than
are generally prepared in total syntheses in an academic
environment. This decision in turn placed a very high
premium on the attainment of high levels of stereoselectivity
in the individual steps.
Accordingly, we could ill afford to deal with complex
mixtures of stereoisomers. It seemed probable to us from the
outset that ring-closing metathesis (RCM), by then a very
popular method for forming cyclic olefins,[18] would, in the
case of the epothilones, lead to an E/Z mixture of stereoisomers at the C12C13 double bond if a precursor such as 10
or 11 was used (Scheme 3).[19] This expectation proved
accurate. We thus came to favor cross-coupling strategies in
which the eventual C12C13 double-bond geometry was
already defined in the olefinic precursors. Incidentally, we
were aware that in placing emphasis on quality syntheses to
Angew. Chem. Int. Ed. 2005, 44, 2838 –2850
Epothilone Derivatives
12,13-alkene with dimethyldioxirane proceeded with very high regio- and stereoselectivity
(15–20:1 in favor of the b epoxide[23]) to
complete the first total synthesis of epothilone A (EpoA, 6).[24] Happily, the logic of the
synthetic route was readily adapted to the first
synthesis of EpoB (7).[16b] The key features of
the pioneering effort are adumbrated in
Scheme 4. Experimental protocols have been
described in detail in the literature.
Scheme 3. Synthesis of epothilones by ring-closing metathesis.
produce carefully thought-out structures, we were deviating
from the prevailing trends—based on stochastic approaches—
of those times. We return to this issue toward the end of this
We were attracted to the palladium-mediated Balkyl
Suzuki cross-coupling, a reaction that was underappreciated
at the time, to establish the C12(sp2)C11(sp3) bond.[20] The
synthesis of EpoA commenced with compound 16, prepared
by taking advantage of chemistry developed to a large extent
by our research group.[20] Thus, the Lewis acid catalyzed
diene–aldehyde cyclocondensation reaction[21] of 14 and 15
afforded dihydropyrone 16 in high yield and with high relative
and absolute stereoselectivity (Scheme 4). This compound
was converted into intermediate 18, which underwent the key
Balkyl Suzuki reaction with 19 to afford 21. A virtually
unprecedented ester enolate–aldehyde macrocyclization[22]
followed by appropriate functional-group manipulations led
us to desoxyepothilone A (dEpoA, 12). Epoxidation of the
2.2. Biological Activity
With synthetically derived material in hand, we soon
confirmed the various claims in the literature concerning
inhibition of microtubule disassociation and the cytotoxicity
of epothilones A and B.[25] Remarkably, in vitro experiments
revealed that the 12,13-desoxy precursor of EpoB (i.e.
dEpoB) also exhibited tubulin-stabilization properties within
the range of those of EpoB, though the cytotoxicity was
diminished.[12] By permuting the total synthesis described
above in various ways, we built a family of epothilone
congeners, which served to define the first in vitro SAR
(structure–activity relationship) map of the epothilones.[20]
We were able to determine which structural elements of the
epothilones could be modified without loss of cytotoxicity or
tubulin-binding characteristics.
Although our initial syntheses of EpoA and EpoB were
quite long, our newly launched drug-discovery program
benefited greatly from their high stereoselectivity. With
considerable diligence, we were able to accumulate adequate
quantities of totally synthetic probe structures for meaningful
in vivo determinations. The experiments involved implanting
human tumor cells into immunodeficient mice and assessing
the effects of the drugs on the growth profiles of the tumors.
We also conducted parallel investigations on tumors that had
become insensitive to various known anticancer drugs as a
consequence of MDR. Hence, these fully synthetic drug
samples could be used to determine whether the claims of
Bollag and co-workers about MDR insensitivity (for example,
in the case of epothilone B)[12] were also true in an in vivo
setting. We conducted the first in vivo investigations with
epothilone B. Our findings were highly worrisome. Although
the potency of EpoB was certainly indicated, its toxicity
profile was quite serious: deaths and troubling weight losses
resulted even at doses as low as 0.6 mg kg1 (Table 1).
With appropriate changes in dosing, it might well have
been possible to learn to administer EpoB in such a way as to
achieve useful therapeutic indices that would justify its
Table 1: Toxicity of EpoB and dEpoB in normal nude mice.
Scheme 4. First reported total synthesis of EpoA and EpoB.
Angew. Chem. Int. Ed. 2005, 44, 2838 –2850
Dose [mg kg1][a]
EpoB (7)
dEpoB (13)
[a] QDx4 (administered every day, 4 doses in total), intraperitoneal (i.p.).
[b] Mice died of toxicity on day 5,6,6,7,7,7,7,7.
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
S. J. Danishefsky et al.
potential development as a drug. In fact, Novartis did enter
this drug into phase I clinical trials.[26] However, we were
concerned that toxicity issues would, in the end, prevent
EpoB from maturing into the sort of breakthrough drug we
sought to discover.
We hypothesized that perhaps the serious toxicity of
EpoB was partly the consequence of nonspecific toxicity
superimposed on tubulin-directed antimitotic properties. It
seemed that the 12,13-epoxide of EpoB could well be a source
of nonselective toxicity.[16] Thus, we wondered whether
“molecular editing” of the epoxide in favor of a cis-12,13alkene would give rise to agents with reduced toxicity and
therefore broader and more exploitable therapeutic indices.
3. Second-Generation Epothilones
3.1. Desoxyepothilone B: Synthesis and Activity
As dEpoB had demonstrated cytotoxicity and tubulinbinding characteristics comparable to those of taxol in vitro,
but with activity in MDR cell lines reminiscent of that of
EpoB, we decided to evaluate this compound in vivo.
Multigram quantities of synthetic dEpoB (13) were required
for this task and delivered by later-generation syntheses (see
Comprehensive in vivo studies showed dEpoB to be much
more promising than EpoB itself with respect to the usefulness of its therapeutic index (Table 1).[25e] Subsequently, we
expanded our efforts by examining the effects of various
formulations, routes, and schedules of intravenous (i.v.)
administration. We discovered that dEpoB performed similarly to paclitaxel in tumor xenografts such as MX-1, in which
each demonstrated a complete tumor remission. Similar
results were seen in the treatment of HT-29 colon tumor
and SK-OV-3 ovarian tumor with dEpoB and taxol. However,
strikingly superior effects of dEpoB relative to taxol were
observed against multiple-drug-resistant (MDR) tumors in
our own models. For example, dEpoB (30 mg kg1, 6-h i.v.
infusion, Q2Dx5 = administered every other day, 5 doses in
total) demonstrated a full curative effect when administered
to nude mice bearing the resistant human lymphoblastic Tcell leukemia CCRF-CEM/paclitaxel, which was 57-fold
resistant to paclitaxel (Figure 1 a).
Additionally, the superior effects of dEpoB relative to
other commonly used anticancer agents were clearly shown in
the treatment of adriamycin-resistant MCF-7/Adr tumor
xenografts with frontline chemotherapeutic agents, such as
paclitaxel (24 mg kg1), adriamycin (3 mg kg1), vinblastine
(0.8 mg kg1), and etoposide (VP-16, 30 mg kg1), each at the
maximal tolerated doses. In these evaluations, adriamycin
demonstrated a lack of therapeutic effect even at the nearly
lethal dose, and vinblastine, etoposide, and paclitaxel showed
little therapeutic effect. Upon treatment with dEpoB, a
significant decrease in tumor size was observed (Figure 1 b).
As a consequence of these promising preclinical successes
with dEpoB, it became important for us to develop a more
concise route for its synthesis. Some of the drawbacks of the
first-generation synthesis[20] included the awkwardness of the
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Figure 1. Activity of dEpoB against a) paclitaxel-resistant tumors and
b) an adriamycin-resistant mammary adenocarcinoma: a) ^ control;
& paclitaxel, 20 mg kg1 ; dEpoB, 30 mg kg1; b) ^ control;
& vinblastine, 0.8 mg kg1 ; ~ paclitaxel, 24 mg kg1 ; dEpoB,
30 mg kg1; + adriamycin, 3 mg kg1; * etoposide (VP-16), 30 mg kg1.
introduction of the C4 gem-dimethyl group, the multistep
chain extension of the C9 benzyl ether, and the nonstraightforward opening of the pyran system to expose the C3
aldehyde. We sought to overcome these issues while still
making use of the stereoselective Balkyl Suzuki route.
We were able to achieve these objectives in our secondgeneration synthesis of dEpoB (Scheme 5).[27] This route
commenced with the readily available b-ketoester 23. Following the conversion of 23 into enol ether 24, an aldol reaction
with aldehyde 25 and subsequent manipulations eventually
gave b-ketoester 26. The opposite enantiomer of aldehyde 25
had already been prepared by Overman and co-workers[28]
through recourse to Evans oxazolidinone methodology,[29]
which makes use of a suitable auxiliary to direct the
diastereofacial sense of C2C3 bond formation. Suzuki
coupling of 26 and 27 afforded 28. The stereocenter at C3
was installed by a Ru–binap-catalyzed reduction developed
by Noyori et al.[30] Ultimately, macrolactonization of 29,
followed by deprotection, led to dEpoB (13). A total of
about 60 g of fully synthetic dEpoB was prepared by this
We were now able to evaluate dEpoB relative to taxol,
epothilone B, and a semisynthetic 15-desoxy-15-aza congener,
which had also been entered into clinical trials.[31] We also
Angew. Chem. Int. Ed. 2005, 44, 2838 –2850
Epothilone Derivatives
Kosan scientists detected and isolated, albeit in small
amounts, a desoxy and dehydro derivative of epothilone B,
which they named epothilone 490 (Epo 490).[34] The Kosan
investigators recognized Epo 490 as the derivative of EpoB
without the 12,13-epoxide and with an additional E double
bond at C10C11. We looked upon Epo 490 as a dehydro
version of dEpoB (13). However, the chemical transformation of dEpoB into this new Kosan metabolite was far from a
straightforward matter. We were anxious to take advantage of
our previous extensive efforts in total synthesis to gain access
to meaningful amounts of Epo 490. In other words, we would
try to adapt the total synthesis we had developed for dEpoB
to target Epo 490 directly.
To understand the steps taken in the development of the
new synthesis, it is important to be familiar with the then
state-of-the-art synthesis of the vinyl iodide component
(compound 35) for the Suzuki coupling process. Since we
have described previously the many methods that were
explored during the course of our investigations, we can
quickly turn to the optimal method for preparing this alcohol.
The route is summarized in Scheme 6.[35]
Scheme 5. A more streamlined synthesis of dEpoB. Troc = trichloroethoxycarbonyl.
looked into questions such as the stability of dEpoB to
lactone-ring opening in various in vivo models and the
performance of dEpoB in vivo against sensitive or resistant
tumors. In our judgment, based on xenograft models, dEpoB
demonstrated major advantages over potential competitors in
the range of its likely therapeutic index and in its robustness
with respect to MDR disablement. These results have been
documented extensively.[25c–e]
Another important event in the progression of desoxyepothilone B to clinical evaluation was the entry of the Kosan
Biosciences Company as a licensee of Sloan-Kettering for
further development of the compound. Brilliantly, and with
great diligence, our commercial collaborators solved the
problem of obtaining desoxyepothilone B by fermentation
methods.[32] Their success in doing so was much facilitated by
their highly advanced technology, which promoted the
expression of 13 through the modification of particular
polyketide metabolites by controlling the biosynthesis of
such polyketides.[32] Our collaborators were able to prepare
the compound for phase I clinical trials. The human clinical
evaluations, which commenced at the end of 2001, have
advanced to a phase II stage in a much expanded effort, which
now includes the participation of the company Hoffmann-La
3.2. (E)-9,10-Dehydroepothilone Derivatives
The very effective and timely involvement of Kosan in the
project also provided another important step in our progression to the (E)-9,10-dehydro compounds, including fludelone. In the course of their pivotal fermentation work,
Angew. Chem. Int. Ed. 2005, 44, 2838 –2850
Scheme 6. Synthesis of the left-hand fragment 35. TES = triethylsilyl,
LHMDS = lithium hexamethyldisilazide.
A postdoctoral colleague, Dr. Jon Njardarson, offered an
interesting proposal for the synthesis of epothilone 490. His
idea brought us back to a more careful consideration of the
RCM method, which had been left on the sidelines for the
reasons discussed above. The coupling of the iodoalkene 35
with a suitable vinyl derivative should give substrate 36.
Following esterification with 37, ring-closing metathesis could
well lead to the desired (E)-10,11-(Z)-12,13-diene, which,
upon appropriate functional-group transformations, should
allow access to epothilone 490 (40). Indeed, this synthesis was
soon completed (Scheme 7).[36] The exact method for introducing the vinyl group involved a Stille cross-coupling
reaction of 35 with tri-n-butylvinylstannane. Since Epo 490
could be smoothly converted into 13 by a selective imide
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
S. J. Danishefsky et al.
Scheme 7. Total synthesis of epothilone 490 by ring-closing metathesis.
Mes = 2,4,6-trimethylphenyl.
reduction, the chemistry described in Scheme 7 constitutes an
independent total synthesis of dEpoB.
Unfortunately, the excellent in vitro results observed with
epothilone 490 did not translate to in vivo experiments
conducted with mouse xenografts on our fully synthetic
material. Although clear evidence for the suppression of
tumor growth could be found, tumor shrinkage, let alone
disappearance, was not observed.[21] On the basis of biostability studies, we believed that this lack of translation to an in
vivo setting might be specific to the mouse tumor host.
Thus, pharmacostability studies suggested that Epo 490 is
particularly unstable with respect to esterase action in mouse
sera. Remarkably, the compound was actually quite stable in
human sera of cell homogenates. However, in the real world
of advancing oncostatic agents for development, progress
depends rather critically on the promising performance of the
agent in the mouse models. Although the Epo 490 project
continues to provide a forum for testing interesting chemistry,
efforts to utilize Epo 490 as a drug have been discontinued in
our laboratory.
Concurrently with the Epo 490 project, we undertook the
synthesis and evaluation of a 26-trifluoro epothilone derivative. In this connection, we were not unmindful of many
instances in which the strategic incorporation of fluorine
atoms can lead to major effects on drug activity, including
altered lipophilicity, incremental stabilization to metabolism,
and modified binding affinities.[37] Even in the absence of a
clear-cut rationale for doing so, we set out to synthesize
dEpoB derivatives with a trifluoromethyl group at C12.
Naively, we assumed that the hard-won lessons from the
synthesis of dEpoB could be readily transferred to the
problem at hand. Thus, at the outset we set as our targets
26-trifluoro compounds in both the Epo 490 family (see
compound 42) and the dEpoB series (see 43).[38] Surprisingly,
the strategies that had worked well in the syntheses of the
nonfluorinated congeners failed in the presence of the
trifluoromethyl group. Thus, with the trifluoromethyl group
installed, ring-closing metathesis (41!42), nucleophilic methylation of the Weinreb amide (44!45), and Balkyl Suzuki
coupling (44+26!46) were all unsuccessful (Scheme 8).[39]
Clearly, we had significantly underestimated the consequences of the inclusion of the trifluoromethyl group for our
synthesis. Since a Stille cross-coupling reaction had allowed us
to construct the precursor 41, we investigated the possibility
of introducing an allyl group at the future C12 vinylic carbon
atom en route to a 26-trifluoro epothilone derivative. We
hoped that the inclusion of even a single methylene spacer
would attenuate the effects of the trifluoromethyl group, thus
enabling the realization of a RCM reaction. Happily, with the
Scheme 8. Failed strategies for the synthesis of 26-trifluoro-dEpoB.
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 2838 –2850
Epothilone Derivatives
one carbon spacer in place, RCM of 47 did produce
compound 48 in a respectable yield (Scheme 9). Now, of
course, the epothilone functions were displayed in the context
of a 17- rather than a 16-membered ring. Not surprisingly, the
Scheme 9. Successful ring-closing metathesis.
same sequence could be carried out to give the analogue 51
containing a dehydrohomodesoxyepothilone B scaffold and a
methyl group at C12.[40] To broaden our SAR base, we also
introduced a butenyl function on the C12 vinylic carbon atom.
RCM of this compound led to the epothilone macrolide 52.
Remarkably, in all cases the double bonds formed in the RCM
reactions were exclusively of the E configuration.
In a fateful set of experiments, we examined the in vitro
activity of these ring-expanded epothilones. As it turned out,
the 18-membered-ring congener 52 was totally inactive. By
contrast, the 17-membered-ring compound 51 exhibited
substantial in vitro activity. This result at first seemed
puzzling, as Nicolaou and co-workers had reported that the
monodihydro congener of 51, in which the C10C11 bond was
saturated, was devoid of useful epothilone activity.[41] This
observation led us to speculate that perhaps this second
double bond of our synthetic construct could also impart
exploitable incremental biological function to epothilones
with the usual 16-membered macrocycle.
The limitations imposed on our operative synthetic
schemes by the trifluoromethyl group really left us with only
one option if we were to stay within the broad parameters of
our route. We would take advantage of the fact that the
trifluoromethyl group does not interfere with Stille-like
allylation. Furthermore, with the single methylene spacer in
place, olefin metathesis in the ring-closing mode was indeed a
viable reaction. To get back to the 16-membered ring, we
would now have to delete one carbon atom from the
component containing the carboxylic acid functionality in
the esterification reaction. A key reaction in the assembly of
this component would now be a chelation-controlled aldol
reaction of 53 and 54 (Scheme 10). The aldehyde 54 was
obtained smoothly from readily available 2-hydroxybutyric
acid—a charter member of the chiral pool. Hence, a fringe
benefit of targeting the dehydro-dEpoB system was that we
could now dispense with the need for recourse to a chiral
Scheme 10. Synthesis of 9,10-dehydroepothilones through ring-closing metathesis.
Angew. Chem. Int. Ed. 2005, 44, 2838 –2850
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
S. J. Danishefsky et al.
auxiliary to reach the previously used aldehyde 25.[28] Ringclosing metathesis of 60 in the presence of the recently
developed catalyst 39 provided exclusively the E isomer 61
along with the corresponding 7-membered-ring side product
62 in a 1:3 ratio and 82 % combined yield.
At last, we had succeeded in preparing a 12,13-desoxy
compound with a 16-membered ring and the desired trifluoromethyl group. However, the major RCM product was the
unwanted cycloheptadienyloxy compound 62 arising from
extrusion of the thiazole side chain. Clearly, this result did not
meet our goal of a viable total synthesis for drug discovery
and, hopefully, development.
We postulated that this problem might be resolved by
resequencing the steps of the synthesis and conducting the
RCM with the ketone precursor to the thiazole side chain still
in place. We tested this proposal on the 12-methyl substrate 63
and were pleased to discover that the desired RCM adduct
was obtained in 78 % yield. Olefination of the ketone with
emplacement of the thiazole unit occurred in high yield to
give, after deprotection, the desired compound 8.[42]
When we carried out this synthesis, 8 was believed to have
been previously synthesized. Indeed, the compound claimed
to be 8 was reported to be only marginally cytotoxic.[43] We
were therefore surprised to find that the spectroscopic
properties of our synthesized 8 were not consistent with
those reported for the compound previously thought to be 8.
The actual structure of the compound originally assigned as 8
has been reevaluated and shown to be an isomer in which the
C12C13 double bond is in the E configuration.[44] The real
compound 8 had, in fact, never before been prepared and,
accordingly, never been evaluated as a potential antitumor
Having established the feasibility of the modified RCM–
olefination strategy with the methyl substrate 63, we refocused our attention on the ultimate goal of synthesizing 26trifluoro-9,10-dehydro-dEpoB (9). Happily, we were able to
utilize the same set of transformations to obtain the desired
compound 9, which we named fludelone, from ketone 64 in
high yield (Scheme 10).[17b, 42]
In the title of this retrospective, we posed the question as
to “How We Got There”. We had fashioned a special and
highly selective sequence of reactions for the total synthesis of
the epothilones. As matters transpired, the chemistry needed
for the incorporation of the 12-CF3 group required us to
access the 9,10-dehydro series, in which we found a new
family of epothilone drugs with extremely promising biological activity. We conclude with a summary of the rather
remarkable and promising findings from evaluations of these
9,10-dehydro compounds.
4. Biological Activity of Fludelone
Having synthesized the 12-proteo analogue 8 as a
chemical model, we set out to examine its biological activity,
in part to set a baseline for comparison with fludelone (9).
Indeed, compound 8 seems to be a promising secondgeneration drug candidate. As described elsewhere,[17] it is
significantly more potent than dEpoB (13) and is much more
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
stable in a variety of biological contexts than 13. The
treatment of various xenograft tumor models with compound
8 led to dramatic tumor shrinkage in these model in vivo
contexts (Figure 2). However, although the drug completely
suppressed tumor growth, there were obstacles to full tumor
eradication with 8. The problem was that the greater toxicity
of 8 (maximum tolerated dose in xenografts is about
4 mg kg1) made it difficult to dose with sufficient amounts
of the drug to remove the last traces of tumor. The exploitable
therapeutic index of 8, particularly in clinical settings, awaits
fuller exploration.
Figure 2. Therapeutic effect of 9,10-dehydro-dEpoB (8) in nude mice
bearing the HCT-116 xenograft (6-h i.v. infusion (except 5-mg kg1
dosage: i.v. injection), Q2Dx7 = administered every other day, 7 doses
in total, n = 3; arrows indicate drug administrations): * control;
~ 3 mg kg1 ; ^ 4 mg kg1, (n = 4); ~ 5 mg kg1.
By contrast, the 26-trifluoro derivative fludelone (9)
exhibits a remarkably broad therapeutic index in vivo in
xenografts. Dramatic results are observed for a range of
tumors. Indeed, fludelone has emerged as our star compound
in the epothilone series. The data given below form the basis
of our preclinical case.
All chemotherapeutic experiments in vivo were carried
out with human-tumor xenografts in immunodeficient nude
mice. For all its imperfections, this model is the one most
widely used in evaluating antitumor lead compounds prior to
clinical trials. Remarkably, the treatment of MX-1 xenografts
with 25-mg kg1 dosages of fludelone resulted in complete
tumor disappearance and the absence of any relapse for over
two months after the suspension of treatment (see Figure 3).
Most importantly, these therapeutic successes can be achieved
either by 6-h i.v. infusion (Figure 3) or by oral administration
(Figure 4). In contrast, the treatment of the MX-1 xenografts
by oral administration of taxol did not appreciably affect the
tumor, which highlights another significant advantage of
fludelone (see Figure 4). The treatment of taxol-resistant
tumor xenografts with fludelone administered by 6-h i.v.
infusion can lead to complete tumor remission (Figure 5),
Angew. Chem. Int. Ed. 2005, 44, 2838 –2850
Epothilone Derivatives
Figure 4. Therapeutic effects against the human mammary carcinoma
MX-1 xenograft by orally administered fludelone or paclitaxel (taxol).
Female nude mice were used. Fludelone (30 mg kg1, &, n = 3) was
given orally Q2Dx7 beginning D16 after tumor implantation and then
Q2Dx9 on D32 to D48, as indicated by arrows. The tumors of all three
mice disappeared (on D40, D45, and D48). As consolidation therapy, a
third cycle of treatment was given Q2Dx5 from D58 to D66 when all
mice were tumor free on D48. There was no relapse observed on D115
(49 days after stopping treatment). Control (*, n = 2) received the
vehicle only. A comparative experiment was carried out in parallel with
taxol (30 mg kg1, ~, n = 3) with oral dosing beginning D16, Q2Dx3;
the dose was then increased to 40 mg kg1, Q2Dx3 (D22–D26) and
then 60 mg kg1, Q2Dx3 (D28–D40).
Figure 3. Therapy of an extra-large MX-1 tumor xenograft: a) MX-1
tumor tissue (50 mg) was implanted subcutaneously (s.c.) on day 0.
Tumor-size changes in the vehicle-treated control (*) and fludelonetreated group (25 mg kg1, &; n = 5 in each group) were observed. On
day 22 (D22), when the size of the tumor had reached 960 132 mg
(about 3.4 % of body weight), fludelone treatment was commenced.
Fludelone (25 mg kg1, 6-h i.v. infusion, Q3Dx5) was given on D22,
D25, D28, D31, and D34, as indicated by the arrows. The second cycle
of treatment, following a 9-day rest, was given on D43, D46, D49, and
D52. For the fludelone-treated group, 5/5 tumors disappeared (on
D38, D40, D50, D50, D51). Observation was continued Q3D for up to
180 days, and no relapse was observed 128 days after cessation of
treatment on D52. b) Photographs of the nude mice (one mouse each
selected from the control group and the treated group) taken on D25,
D31, D37, D43, and D52.
whereas the treatment of human colon carcinoma (HCT-116,
Figure 6) with fludelone by 6-h i.v. infusion can lead to a
complete “cure”. The experiments with human-mammarycarcinoma (MX-1) and human-colon-carcinoma (HCT-116)
xenografts in nude mice lasted 6.0 and 6.6 months, respectively. There was no tumor relapse in either experiment
during 4.3 and 5.3 months, respectively, following the cessation of treatment. For the HCT-116 experiment, taxol and
fludelone were used at 20 mg kg1 and both caused tumor
disappearance. However, the taxol-treated group relapsed 1.1
months after treatment was discontinued, whereas the
animals treated with fludelone were tumor free for over
5.3 months.
Angew. Chem. Int. Ed. 2005, 44, 2838 –2850
Figure 5. Therapeutic effects against the taxol-resistant human T-cell
lymphoblastic leukemia CCRF-CEM/taxol xenograft by fludelone and
taxol. Tumor tissue of CCRF-CEM/taxol (44-fold resistant in vitro),
50 mg/mouse was implanted s.c. into nude mice on day 0. Treatment
(6-h i.v. infusion) started on D8 with fludelone (15 mg kg1, &, n = 3
and 30 mg kg1, ~, n = 4) or taxol (20 mg kg1, *, n = 4), Q2Dx7 (D8 to
D20), D22 dose was skipped, and then treatment was resumed Q2Dx5
on D24, D26, D28, D30, and D32, as indicated by arrows. The control
group (*, n = 4) received the vehicle only. With fludelone at a dose of
15 mg kg1, the tumor of 1/3 mice disappeared on D37, and at a dose
of 30 mg kg1 the tumors of 3/4 mice disappeared on D22, D22, and
These results are based on a particularly long and
thorough therapeutic study with xenografts and demonstrate
the longest periods of complete remission that have been
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
S. J. Danishefsky et al.
5. Concluding Remarks
Figure 6. Therapeutic effects against the human colon carcinoma
HCT-116 xenograft by fludelone and taxol. HCT-116 tumor tissue
(50 mg/mouse) was implanted s.c. into nude mice on day 0. Treatment
(Q2Dx4, 6-h i.v. infusion) was carried out for three cycles on D9, D11,
D13, D15; D19, D21, D23, D25; and D31, D34, D35, D37 with fludelone (20 mg kg1, & and 30 mg kg1, ~), and taxol (20 mg kg1, *);
control: *; n = 4 in each group. Complete tumor disappearance in all
mice occurred (on D33, D35, D41, D45 for fludelone (20 mg kg1), on
D21, D23, D33, D41 for fludelone (30 mg kg1), and on D33, D33,
D41, D45 for taxol (20 mg kg1)). There was no tumor relapse
observed in either fludelone-treated group on D200. However, the
group treated with taxol (*) suffered relapses on D71, D75, D81, and
D81, which correspond to the 34th, 38th, 41st, and 41st day after
stopping treatment.
reported either with parenteral or oral administration of a
single antitumor agent. It is relatively common to find
compounds that suppress tumor growth. It is rarer to find a
drug candidate that causes tumor shrinkage, and particularly
rare for a compound to shrink the tumor to the point of
nondetectability. The dramatic finding regarding fludelone is
that tumors did not relapse during observation for 4.3 months
or longer (i.e. > 20 % of a typical mouse lifespan). This type
of result has very few counterparts in the literature.[45] The
achievement of complete tumor disappearance and long-term
remission by oral treatment could well be of particular
significance, as it could lead to outpatient home usage.
Furthermore, the use for drug delivery of cremophor vehicles,
which can themselves lead to severe allergic reactions, could
be avoided.
It is important to keep in mind that the ultimate purpose
of chemotherapeutic research is to provide clinically valuable
treatment for cancer patients. Although the results of the
xenograft studies discussed herein are certainly very encouraging, it remains to be seen whether these dramatic findings
will translate to human patients. Thus, only progression to
clinical trials will fully establish the value of these novel, latergeneration epothilones as effective anticancer agents. However, we are prepared to predict that the dramatic preclinical
performance of the 9,10-dehydro-dEpoB epothilones will
spur much research in synthesis, compound optimization, and
biotarget identification.
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
We conclude our documentary of retrospection with some
thoughts of a more general nature. With continuing advances
in the technology of isolation, purification, and characterization, the number of natural products available in the future
for screening could increase dramatically. This expansion in
the reservoir could certainly lead to a substantial increase in
lead structures for drug discovery. The ongoing study of the
epothilones serves to underscore the value of target-directed
total synthesis in the quest for new substances of material
clinical benefit.
This point is well worth making in the current research
environment, which favors recourse to massive numbers of
compounds for screening in preference to smaller numbers of
hypothesis-driven candidate structures based on natural
product leads. Although we readily concur that it is not
inconceivable that purely chance-driven diversity collections
may give rise to successful drug candidates, we would assert
that the wisdom inherent in natural products, now augmented
by the growing field of chemical synthesis, represents a
valuable resource that has been little appreciated of late.[46] In
essence, we are suggesting that in the drug discovery process
there are two directions that can be taken. One possibility is
to hurl a mind-boggling number of compounds at a problem.
There is certainly merit in this approach. For reasons of
personal taste, we prefer to make fewer compounds by
tapping the generous hints provided by nature and exploiting
the resources of human creativity and improvisation. This
lesson may in itself prove to be a valuable fringe benefit of the
journey which led us to fludelone.
Received: August 20, 2004
[1] S. F. Brady, C. J. Chao, J. Clardy, J. Am. Chem. Soc. 2002, 124,
[2] a) V. Knight, J. J. Sanglier, D. DiTullio, S. Braccili, P. Bonner, J.
Waters, D. Hughes, L. Zhang, Appl. Microbiol. Biotechnol. 2003,
62, 446; b) D. J. Newman, G. M. Cragg, K. M. Snader, Nat. Prod.
Rep. 2000, 17, 215.
[3] a) D. C. Myles, Curr. Opin. Biotechnol. 2003, 14, 627; b) Y.-Z.
Shu, J. Nat. Prod. 1998, 61, 1053; c) L. A. Wessjohann, Curr.
Opin. Chem. Biol. 2000, 4, 303; d) D. J. Newman, G. M. Cragg,
K. M. Snader, J. Nat. Prod. 2003, 66, 1022.
[4] A. G. Ravelo, A. Estevez-Braun, H. Chavez-Orellana, E. PerezSacau, D. Mesa-Siverio, Curr. Top. Med. Chem. 2004, 2, 241, and
references therein.
[5] M. C. Wani, H. L. Taylor, M. E. Wall, P. Coggan, A. T. McPhail,
J. Am. Chem. Soc. 1971, 93, 2325.
[6] a) P. B. Schiff, S. B. Horwitz, Proc. Natl. Acad. Sci. USA 1980, 77,
1561; b) P. B. Schiff, S. B. Horwitz, Nature 1979, 277, 665; c) J.
Parness, S. B. Horwitz, J. Cell Biol. 1981, 91, 479; d) K. C.
Nicolaou, W.-M. Dai, R. K. Guy, Angew. Chem. 1994, 106, 38;
Angew. Chem. Int. Ed. Engl. 1994, 33, 15; e) E. K. Rowinsky,
Annu. Rev. Med. 1997, 48, 353.
[7] D. Guenard, F. Gueritte-Vogelein, P. Potier, Acc. Chem. Res.
1993, 26, 160.
[8] S. J. Danishefsky, J. J. Masters, W. B. Young, J. T. Link, L. B.
Snyder, T. V. Magee, D. K. Jung, R. C. A. Isaacs, W. G. Bornmann, C. A. Alaimo, C. A. Coburn, M. J. DiGrandi, J. Am.
Chem. Soc. 1996, 118, 2843.
Angew. Chem. Int. Ed. 2005, 44, 2838 –2850
Epothilone Derivatives
[9] a) K. C. Nicolaou, Z. Yang, J. J. Liu, H. Ueno, P. G. Nantermet,
R. K. Guy, C. F. Claiborne, J. Renaud, E. A. Couladouros, K.
Paulvannan, E. J. Sorenson, Nature 1994, 367, 630; b) K. C.
Nicolaou, P. G. Nantermet, H. Ueno, R. K. Guy, E. A. Couladouros, E. J. Sorenson, J. Am. Chem. Soc. 1995, 117, 624; c) K. C.
Nicolaou, J. J. Liu, Z. Yang, H. Ueno, E. J. Sorenson, C. F.
Claiborne, R. K. Guy, C. K. Hwang, M. Nakada, P. G. Nantermet, J. Am. Chem. Soc. 1995, 117, 634; d) K. C. Nicolaou, Z.
Yang, J. J. Liu, P. G. Nantermet, C. F. Claiborne, J. Renaud, R. K.
Guy, K. Shibayama, J. Am. Chem. Soc. 1995, 117, 645; e) K. C.
Nicolaou, H. Ueno, J. J. Liu, P. G. Nantermet, Z. Yang, J.
Renaud, K. Paulvannan, R. Chadha, J. Am. Chem. Soc. 1995,
117, 653.
[10] a) R. A. Holton, C. Somoza, H. B. Kim, F. Liang, R. J. Biediger,
P. D. Boatman, M. Shindo, C. C. Smith, S. C. Kim, H. Nadizadeh,
Y. Suzuki, C. L. Tao, P. Vu, S. H. Tang, P. S. Zhang, K. K. Murthi,
L. N. Gentile, J. H. Liu, J. Am. Chem. Soc. 1994, 116, 1597;
b) R. A. Holton, H. B. Kim, C. Somoza, F. Liang, R. J. Biediger,
P. D. Boatman, M. Shindo, C. C. Smith, S. C. Kim, H. Nadizadeh,
Y. Suzuki, C. L. Tao, P. Vu, S. H. Tang, P. S. Zhang, K. K. Murthi,
L. N. Gentile, J. H. Liu, J. Am. Chem. Soc. 1994, 116, 1599.
[11] a) E. K. Rowinsky, E. A. Eisenhauer, V. Chaudhry, S. G. Arbuck, R. C. Donehawer, Semin. Oncol. 1993, 20, 1; b) B. S.
Fletcher, D. A. Kujubadu, D. M. Perrin, H. R. Herschman, J.
Biol. Chem. 1992, 267, 4338; c) M. Tsuji, R. N. Dubois, Cell,
1995, 3, 493; d) D. M. Essayan, A. Kagey-Sobotka, P. J. Colarusso, L. M. Lichtenstein, R. F. Ozols, E. D. King, J. Allergy Clin.
Immunol. 1996, 97, 42; e) P. Giannakakou, D. L. Sackett, Y.-K.
Kang, Z. Zhan, J. T. Buters, T. Fojo, M. S. Poruchynsky, J. Biol.
Chem. 1997, 272, 17 118, and references therein.
[12] G. Hfle, N. Bedorf, K. Gerth, H. Reichenbach (GBF), DE-B
4138042, 1993; Chem. Abstr. 1993, 120, 52841.
[13] 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. We thank Professor Gunda Georg of the University of
Kansas for calling this publication to our attention, and Dr.
D. M. Bollag of Merck Sharp and Dohme for providing us with
about 1 mg of epothilone A to commence studies (ultimately
unsuccessful) towards the determination of its configuration.
[14] a) See reference [12]; b) R. J. Kowalski, E. Terhaar, R. E. Longley, S. P. Gunasekera, C. M. Lin, B. V. Day, E. Hamel, Mol. Biol.
Cell 1995, 6, 2137; c) E. K. Rowinsky, E. A. Eisenhauer, V.
Chaudhry, S. G. Arbuck, R. C. Donehawer, Semin. Oncol. 1993,
20, 1; d) B. S. Fletcher, D. A. Kujubadu, D. M. Perrin, H. R.
Herschman, J. Biol. Chem. 1992, 267, 4338; e) M. Tsuji, R. N.
Dubois, Cell 1995, 3, 493; f) D. M. Essayan, A. Kagey-Sobotka,
P. J. Colarusso, L. M. Lichtenstein, R. F. Ozols, E. D. King, J.
Allergy Clin. Immunol. 1996, 97, 42; g) P. Giannakakou, D. L.
Sackett, Y.-K. Kang, Z. Zhan, J. T. Buters, T. Fojo, M. S.
Poruchynsky, J. Biol. Chem. 1997, 272, 17 118, and references
[15] G. H. Hfle, N. Bedorf, H. Steinmetz, D. Schomburg, K. Gerth,
H. Reichenbach, Angew. Chem. 1996, 108, 1671; Angew. Chem.
Int. Ed. Engl. 1996, 35, 1567.
[16] a) A. Balog, D. F. Meng, T. Kamenecka, P. Bertinato, D.-S. Su,
E. J. Sorensen, S. J. Danishefsky, Angew. Chem. 1996, 108, 2976;
Angew. Chem. Int. Ed. Engl. 1996, 35, 2801; b) D.-S. Su, D. F.
Meng, P. Bertinato, A. Balog, E. J. Sorensen, S. J. Danishefsky,
Y. H. Zheng, T.-C. Chou, L. F. He, S. B. Horwitz, Angew. Chem.
1997, 109, 775; Angew. Chem. Int. Ed. Engl. 1997, 36, 757;
c) D. F. Meng, P. Bertinato, A. Balog, D.-S. Su, T. Kamenecka,
E. J. Sorensen, S. J. Danishefsky, J. Am. Chem. Soc. 1997, 119,
10 073.
[17] a) A. Rivkin, F. Yoshimura, A. E. Gabarda, T.-C. Chou, H.
Dong, W. P. Tong, S. J. Danishefsky, J. Am. Chem. Soc. 2003, 125,
2899; b) T.-C. Chou, H. Dong, A. Rivkin, F. Yoshimura, A. E.
Gabarda, Y. S. Cho, W. P. Tong, S. J. Danishefsky, Angew. Chem.
Angew. Chem. Int. Ed. 2005, 44, 2838 –2850
2003, 115, 4910; Angew. Chem. Int. Ed. 2003, 42, 4761; c) F.
Yoshimura, A. Rivkin, A. E. Gabarda, T.-C. Chou, H. Dong, G.
Sukenick, S. J. Danishefsky, Angew. Chem. 2003, 115, 2622;
Angew. Chem. Int. Ed. 2003, 42, 2518.
In an early RCM-based approach studied in our laboratory, the
12,13-epoxide was already in place. We also investigated RCM to
reach the 12,13-alkene, but only in the context of surveying
various possibilities. For reviews of RCM, see: a) R. H. Grubbs,
S. J. Miller, G. C. Fu, Acc. Chem. Res. 1995, 28, 446; b) T. M.
Trinka, R. H. Grubbs, Acc. Chem. Res. 2001, 34, 18; c) A.
Frstner, Alkene Metathesis in Organic Chemistry, Springer,
Berlin, 1998; d) A. Frstner, Angew. Chem. 2000, 112, 3140;
Angew. Chem. Int. Ed. 2000, 39, 3012; e) R. R. Schrock, Top.
Organomet. Chem. 1998, 1, 1.
See reference [16c] and A. Rivkin, Y. S. Cho, A. E. Gabarda, Y.
Fumihiko, S. J. Danishefsky, J. Nat. Prod. 2004, 67, 139.
C. R. Harris, S. J. Danishefsky, J. Org. Chem. 1999, 64, 8434, and
references therein.
a) S. J. Danishefsky, Aldrichimica Acta 1986, 19, 59; b) S. J.
Danishefsky, Chemtracts 1989, 2, 273.
Naturally, when we conceived this remarkable aldolization we
were aware of the non-enolizability of the “C3” aldehyde in the
C2–C3 seco structure.
S. J. Stachel, S. J. Danishefsky, Tetrahedron Lett. 2001, 42, 6785.
See reference [16a]. Shortly after our synthesis, two other
efforts, each exploiting olefin metathesis, were described: a) Z.
Yang, Y. He, D. Vourloumis, H. Vallberg, K. C. Nicolaou, Angew.
Chem. 1997, 109, 170; Angew. Chem. Int. Ed. Engl. 1997, 36, 166;
b) D. Schinzer, A. Limberg, A. Bauer, O. Bhm, M. Cordes,
Angew. Chem. 1997, 109, 543; Angew. Chem. Int. Ed. Engl. 1997,
36, 523.
For the first SAR (structure–activity relationship) map, see:
a) D.-S. Su, A. Balog, D. Meng, P. Bertinato, S. J. Danishefsky,
Y.-H. Zheng, T.-C. Chou, L. He, S. B. Horwitz, Angew. Chem.
1997, 109, 2178; Angew. Chem. Int. Ed. Engl. 1997, 36, 2093; see
also: b) K. C. Nicolaou, D. Vourloumis, T. Li. , J. Pastor, N.
Winssinger, Y. He, S. Ninkovic, F. Sarabia, H. Vallberg, F.
Roschanger, N. P. King, M. Ray, V. Finlay, P. Giannakakou, P.
Verdier-Pinard, E. Hamel, Angew. Chem. 1997, 109, 2181;
Angew. Chem. Int. Ed. Engl. 1997, 36, 2097; for the earliest in
vivo assays of dEpoB, see: c) T.-C. Chou, X. G. Zhang, C. R.
Harris, S. D. Kuduk, A. Balog, K. A. Savin, J. R. Bertino, S. J.
Danishefsky, Proc. Natl. Acad. Sci. USA 1998, 95, 15 798; d) T.-C.
Chou, X. G. Zhang, A. Balog, D. Su, D. F. Meng, K. Savin, J. R.
Bertino, S. J. Danishefsky, Proc. Natl. Acad. Sci. USA 1998, 95,
9642; e) T.-C. Chou, O. A. OConnor, W. P. Tong, Y. Guan, Z.-G.
Zhang, S. J. Stachel, C. Lee, S. J. Danishefsky, Proc. Natl. Acad.
Sci. USA 2001, 98, 8113.
Novartis is the sponsor of the clinical trials with EpoB. For EpoB
preclinical data reported by Novartis, see: a) K.-H. Altmann, M.
Wartmann, T. OReilly, Biochim. Biophys. Acta 2000, 1470, M79;
b) J. Rothermel, M. Wartmann, T. Chen, J. Hohneker, Semin.
Oncol. 2003, 6(Suppl.30), 51.
C. R. Harris, S. D. Kuduk, A. Balog, K. Savin, P. W. Glunz, S. J.
Danishefsky, J. Am. Chem. Soc. 1999, 121, 7051.
N.-H. Lin, L. E. Overman, M. H. Rabinowitz, L. A. Robinson,
M. J. Sharp, J. Zablocki, J. Am. Chem. Soc. 1996, 118, 9062.
D. A. Evans, J. V. Nelson, T. R. Taber, Top. Stereochem. 1982, 13,
R. Noyori, T. Ohkuma, M. Kitamura, H. Takaya, N. Sayo, H.
Kumobayahshi, S. Akutagawa, J. Am. Chem. Soc. 1987, 109,
Bristol-Myers Squibb (BMS) is sponsoring clinical trials of the
15-desoxy-15-aza congener: a) R. M. Borzilleri, X. 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; b) F. Y. F. Lee, R. Borzilleri, C. R.
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
S. J. Danishefsky et al.
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; BMS
is also sponsoring clinical trials of another epothilone derivative
(BMS310705): c) R. M. Borzilleri, G. D. Vite, Drugs Future 2002,
27, 1149.
L. Tang, S. Shah, L. Chung, J. Carney, L. Katz, C. Khosla, B.
Julien, Science 2000, 287, 640.
For more information about the clinical trials of dEpoB, visit:
R. L. Arslanian, L. Tang, S. Blough, W. Ma, R. G. Qiu, L. Katz,
J. R. Carn, J. Nat. Prod. 2002, 65, 570.
M. D. Chappell, S. J. Stachel, C. B. Lee, S. J. Danishefsky, Org.
Lett. 2000, 2, 1633.
K. Biswas, H. Lin, J. T. Njardarson, M. D. Chappell, T.-C. Chou,
Y. Guan, W. P. Tong, L. He, S. B. Horwitz, S. J. Danishefsky, J.
Am. Chem. Soc. 2002, 124, 9825.
a) B. E. Smart, J. Fluorine Chem. 2001, 109, 3; b) I. Ojima, T.
Inoue, S. Chakravarty, J. Fluorine Chem. 1999, 97; c) R. A.
Newman, J. Yang, M. R. V. Finlay, F. Cabral, D. Vourloumis,
L. C. Stephens, P. Troncoso, X. Wu, C. J. Logothetis, K. C.
Nicolaou, N. M. Navone, Cancer Chemother. Pharmacol. 2001,
48, 319; d) Special issue on “Fluorine in the Life Sciences”,
ChemBioChem 2004, 5, 557 – 726.
A. Rivkin, K. Biswas, T.-C. Chou, S. J. Danishefsky, Org. Lett.
2002, 4, 4081.
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
[39] Unpublished results.
[40] A. Rivkin, J. T. Njardarson, K. Biswas, T.-C. Chou, S. J.
Danishefsky, J. Org. Chem. 2002, 67, 7737.
[41] K. C. Nicolaou, F. Sarabia, S. Ninkovic, M. Ray, V. Finlay,
C. N. C. Boddy, Angew. Chem. 1998, 110, 85; Angew. Chem. Int.
Ed. Engl. 1998, 37, 81.
[42] A. Rivkin, F. Yoshimura, A. E. Gabarda, Y. S. Cho, T.-C. Chou,
H. Dong, S. J. Danishefsky, J. Am. Chem. Soc. 2004, 126, 10 913.
[43] J. D. White, R. G. Carter, K. F. Sundermann, M. Wartmann, J.
Am. Chem. Soc. 2001, 123, 5407.
[44] J. D. White, R. G. Carter, K. F. Sundermann, M. Wartmann, J.
Am. Chem. Soc. 2003, 125, 3190.
[45] For a recent and impressive case of long-term cure, see: M. J.
Towle, K. A. Salvato, J. Budrow, B. F. Wels, G. Kuznetsov, K. K.
Aalfs, S. Welsh, W. Zheng, B. M. Seletsky, M. H. Palme, G. J.
Habgood, L. A. Singer, L. V. DiPietro, Y. Wang, J. J. Chen, D. A.
Quincy, A. Davis, K. Yoshimatsu, Y. Kishi, M. J. Yu, B. A.
Littlefield, Cancer Res. 2001, 61, 1013. Our case is particularly
noteworthy in that the cures were established consistently.
Furthermore, they were established on high tumor burdens, and
the drug shows oral-administration efficacy.
[46] R. Breinbauer, I. R. Vetter, H. Waldmann, Angew. Chem. 2002,
114, 3002; Angew. Chem. Int. Ed. 2002, 41, 2878.
Angew. Chem. Int. Ed. 2005, 44, 2838 –2850
Без категории
Размер файла
664 Кб
fludelone, properties, remarkable, got, antitumor
Пожаловаться на содержимое документа