Design and synthesis of 12-aza-epothilones (azathilones) - "Non-natural" natural products with potent anticancer activity

Article abstract of DOI:10.1002/anie.200601359

Improving on nature: Azathilones 1a and 1b are "non-natural" natural products derived from epothilones through C→N exchange at position 12. They are highly potent inducers of tubulin polymerization (part of a microtubule is shown in the picture) and inhib

Full text of DOI:10.1002/anie.200601359

Communications
Epothilones
DOI: 10.1002/anie.200601359
Design and Synthesis of 12-Aza-Epothilones
(Azathilones)—“Non-Natural” Natural Products
with Potent Anticancer Activity**
Fabian Feyen, Jürg Gertsch, Markus Wartmann, and
Karl-Heinz Altmann*
Natural products provide an immeasurable pool of lead
structures for drug discovery, and more than 50% of todayꢀs
[*] Dr. F. Feyen, Dr. J. Gertsch, Prof. Dr. K.-H. Altmann
ETH Zürich
Department of Chemistry and Applied Biosciences
Institute of Pharmaceutical Sciences
ETH Hönggerberg, HCI H 405
Wolfgang-Pauli-Strasse 10, 8093 Zürich (Switzerland)
Fax: (+41)44-633-1360
E-mail: karl-heinz.altmann@pharma.ethz.ch
Dr. M. Wartmann
Oncology DA
Novartis Institute for Biomedical Research
4002 Basel (Switzerland)
[**] We thank Bettina Sager, Jacqueline Loretan, and Robert Reuter for
their excellent technical assistance. The Electron Microscopy Centre
(EMEZ) of ETH Zürich is acknowledged for excellent technical
support.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
5880
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 5880 –5885

Angewandte
Chemie
prescription drugs are ultimately derived from compounds
first obtained from natural sources.[1] While the screening of
large collections of natural products will continue to be an
important strategy for the identification of newdrug leads or
compounds of biochemical and pharmacological interest,[2]
various approaches have been developed in the more recent
past that aim at expanding the structural scope of natural-
products-based drug discovery. In particular, this includes the
de novo construction of libraries of “natural-products-like”
compounds through diversity-oriented synthesis (DOS),[3] as
pioneered by Schreiber et al., and the design of natural-
products-based libraries, a concept that was introduced by
Waldmann et al. in 2002[4] and has been continuously refined
over the last fewyears. [5]
Our own research in the area of natural-products-based
drug discovery has been directed towards the development of
newbiologically active scaffolds (or chemotypes) through the
extensive structural modification (rather than simple periph-
eral derivatization) of existing natural product leads.[6] In this
context we have also investigated the biological activity of 12-
aza-epothilones, which are characterized by the replacement
of a backbone carbon atom by nitrogen in the epothilone
macrocycle.[6c,d,7] The resulting analogues may best be de-
scribed as “non-natural” natural products,[8] as they still retain
most of the (two-dimensional) structural features of the
natural product lead; at the same time they are structurally
unique, as they are outside of the general scope of natureꢀs
biosynthetic machinery for polyketide biosynthesis, which is
not programmed for the incorporation of single nitrogen
atoms in a regular polyketide backbone.[9] As a result of these
studies we discovered that 12-aza-epothilones 1 (termed
case, i.e. with R = tert-butyl (1a), was ca. 15–50 times lower
than that of Epo A. These shortcomings led us to investigate
alternative synthetic routes to this class of compounds, with
the concurrent incorporation of additional structural modifi-
cations designed to enhance biological activity. Apart from
available data on the structure–activity relationships (SARs)
for azathilones 1 with regard to the nature of the acyl
substituent attached to N12 (vide supra),[6d] the design of
these analogues was additionally guided by the results of
previous studies in our laboratory, which had revealed a
general activity-enhancing effect of a dimethylbenzimidazole
side chain in combination with the natural epothilone macro-
cycle.[6a,b,11] These considerations resulted in compounds 2a
and 2b as initial targets for total synthesis and biological
investigation, both of which eventually proved to be highly
potent inhibitors of human tumor growth in vitro. In this
communication we now wish to report on the total synthesis
of azamacrolides 2a and 2b,[10] either through ring-closing
olefin metathesis (RCM) or by
a macrolactonization
approach, and the preliminary biological characterization of
these compounds.
Macrocycle formation through RCM featured as a
particularly attractive approach to the target azathilones, as
it offered simultaneous access to specific unsaturated ana-
logues (as the immediate cyclization products), which could
be interesting newantiproliferative agents in their own
right.[12] Our RCM-based synthesis of target structure 2a
and its 9,10-didehydro derivative 9 (Scheme 1) involved three
key strategic steps, namely, 1) the stereoselective aldol
reaction between aldehyde 3[13] and ketone 4[14] (d.r. 8:1),
2) esterification of carboxylic acid 7 with the unsaturated
alcohol 10,[15] and 3) RCM of bisolefin 8.
Initial attempts to cyclize 8 employing the first-generation
Grubbs catalyst[16] met with complete failure and no con-
version was observed. In contrast, the use of the dihydroimi-
dazol-2-ylidene-based second-generation catalyst[16] produced
the cyclic olefin in excellent yield (85%) and with exclusive E
selectivity. No trace of the corresponding Z product could be
isolated, and similar observations have been made for the
RCM-based cyclization of the analogue of 8 containing the
natural epothilone side chain.[24]
Unfortunately, the efficiency of the cyclization reaction
was thwarted by serious difficulties encountered in the
ꢀ
subsequent reduction of the C9 C10 double bond, which
proved to be extremely sluggish under all experimental
conditions investigated (thus leading to lowyields and also
side reactions such as reductive ester cleavage with H2/Pd-C
without reduction of the double bond). The only viable
approach for the transformation of 9 into 2a involved the use
of in situ generated diimide, which had been successfully
employed in the transformation of 9,10- and 10,11-didehy-
droEpo D, respectively, into Epo D (= 12,13-deoxyEpoB),[17]
and which produced 2a in 31% yield from 9 after purification
by preparative HPLC.
While this approach provided sufficient material for initial
biological testing, it was clear that an alternative strategy
would have to be developed for more extensive profiling and
eventual in vivo studies, should those appear to be warranted.
In light of the highly promising biological data obtained for
“azathilones”) can retain much of the antiproliferative
activity of natural epothilones, depending on the nature of
the acyl substituent attached to the backbone nitrogen
atom.[6d,10]
Unfortunately, our first-generation synthesis of these
compounds proved to be relatively inefficient and tedious.[6d]
At the same time, the activity of azathilones 1 even in the best
Angew. Chem. Int. Ed. 2006, 45, 5880 –5885
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5881

Communications
Scheme 2. Reaction conditions: a) H2/Pd-C, EtOAc, RT, 62 h, 86%;
b) HN3, DEAD, PPh3, THF, 08C, 25 min, RT, 30 min, 96%; c) H2/Pd-C,
MeOH, RT, 3 h, 92%; d) 1. 13 (1.1 equiv), NaBH(OAc)3 (1.6 equiv),
AcOH (2.0 equiv), 4-Š MS, RT, 2.5 h; 2. Boc2O, Et3N, THF, 08C,
45 min, 60% (twosteps); e) CSA (1.1 equiv), CH 2Cl2/MeOH 1:1, 08C,
3 h, 80%; f) PDC (15 equiv), DMF, RT, 24 h, 50%; g) TBAF (6 equiv),
THF, RT, 24 h; h) 2,4,6-Cl3C6H2C(O)Cl, Et3N, THF, 08C, 20 min, then
diluted with toluene and added to a solution of DMAP in toluene,
758C, 1 h, 44% (twosteps); i) HF·pyridine, pyridine, THF, RT, 2.5 h,
then preparative HPLC, 40%. j) ZnBr2 (4.0 equiv), CH2Cl2, RT, 2.5 h,
quant.; k) CH3CH2OC(O)Cl, Et3N, THF, 08C, 30 min; l) HF·pyridine,
pyridine, THF, RT, 3.5 h, then preparative HPLC, 32% (twosteps).
Bn=benzyl, Boc=tert-butyloxycarbonyl, DEAD=diethylazodicarboxy-
late, TBAF=tetrabutylammonium fluoride.
Scheme 1. Reaction conditions: a) 4, LDA, ꢀ788C, 5 h, then addition
of 3, ꢀ908C, 75 min, 76%, d.r. 8:1; b) PPTS, MeOH, RT, 20 h, 86%;
c) 1. TBSOTf, 2,6-lutidine, ꢀ788C ! RT, 1.5 h; 2. flash chromatogra-
phy; 76%; d) 1. H2/Pd-C, MeOH, RT, 20 h; 2. TPAP, NMO, 4-Š MS,
CH2Cl2, RT, 1 h; 3. MePPh3Br, LiHMDS, THF, 08C, 1.5 h, 79% (three
steps); e) CSA (1.0 equiv), CH2Cl2/MeOH 1:1, 08C, 1 h, 87%; f) PDC
(11 equiv), DMF, RT, 64 h, 85%; g) 10, DCC (1.2 equiv), DMAP
(0.3 equiv), CH2Cl2, 08C, 15 min, RT, 15 h, 60%; h) 2nd-generation
Grubbs catalyst (0.15 equiv, incremental addition), CH2Cl2, reflux, 8 h,
with
a slight excess of aldehyde 13 (1.1 equiv) and
NaBH(OAc)3 as the reducing agent in the presence of
AcOH (2 equiv) and 4-Š molecular sieves. Owing to its
pronounced polarity (arising from the presence of the
secondary amino group as well as the benzimidazole
moiety) the reductive-amination product was not purified
but directly converted into the corresponding N-tert-butyl-
oxycarbonyl derivative, which was obtained in 60% yield
(based on amine 12). Selective cleavage of the primary tert-
butyldimethylsilyl (TBS) ether with CSA followed by oxida-
tion of the resulting free alcohol with PDC and removal of the
=
85%; i) HF·pyridine, pyridine, THF, RT, 4 h, 70%; j) KO2C-N N-CO2K
(excess), AcOH, CH2Cl2, 31%, pure 1 obtained through purification by
preparative HPLC. CSA=(+)-camphorsulfonic acid, DCC=N,N’-dicy-
clohexylcarbodiimide, DMAP=4-dimethylaminopyridine, LDA=lithium
diisopropylamide, LiHMDS=lithium 1,1,1,3,3,3-hexamethyldisilazide,
NMO=4-methylmorpholine-N-oxide, PDC=pyridinium dichromate,
PMB=para-methoxybenzyl, PPTS=pyridinium p-toluenesulfonate,
TBS=tert-butyldimethylsilyl, TPAP=tetrapropylammonium perruthen-
ate.
ꢀ
TBS protecting group from C15 O with TBAF then led to
seco acid 14, which was cyclized under Yamaguchi condi-
tions[18] to produce fully protected 2a (44% based on C15-O-
TBS-protected 14). Subsequent selective removal of the TBS
protecting groups with HF·pyridine gave target structure 2a
in 40% yield (after HPLC purification).
2a (vide infra), we thus embarked on the elaboration of an
alternative route to 2a that would be based on macrolacto-
nization rather than RCM.
This approach employed the reductive amination of
aldehyde 13 with amine 12 (obtained in three steps from the
known protected tetrol 11[14a]) to assemble the heteroaliphatic
skeleton of 2a (Scheme 2). The reaction was best conducted
In a preliminary attempt to assess the importance of the
tert-butyl moiety of 2a for biological potency (vide infra), we
also prepared the closely related azathilone 2b, which
5882
www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 5880 –5885

Angewandte
Chemie
incorporates a N12-ethoxycarbonyl substituent in place of the
tert-butyloxycarbonyl moiety present in 2a.[19] Azathilone 2b
was obtained from bis-TBS-protected 2a through highly
selective cleavage of the tert-butyloxycarbonyl group from
N12 (ZnBr2, CH2Cl2, RT)[20] followed by acylation of the
resulting free amine 15 with ethyl chloroformate and sub-
sequent removal of the TBS protecting groups with HF·pyr-
idine (Scheme 2).
The data summarized in Table 1 indicate that azathilone
2a is a highly potent antiproliferative agent, which inhibits the
growth of different types of drug-sensitive human cancer cell
lines (A549, HCT-116, PC-3M, KB-31) with IC50 values in the
lownanomolar range.
observation that compound 9, which incorporates a trans
double bond between C9 and C10, is significantly less potent
than the fully saturated azathilone 2a (both at the levels of
tubulin polymerization as well as cellular activity). A similar
potency difference also exists between 1a and its trans 9,10-
didehydro derivative.[25] These findings are in marked contrast
to the effects observed for Epo B and D, where the
introduction of a trans double bond between C9 and C10
results in enhanced cellular potency,[12b,17a] and they may be
indicative of differences in the bioactive conformation
between azathilone-type analogues and natural epothilones.
Compared to azathilone 2a, analogue 2b exhibits some-
what reduced cellular activity (Table 1), but the compound
still remains a very potent antipro-
liferative agent against all drug-
Table 1: ab-Tubulin-polymerizing and antiproliferative activity of azathilones 2a, 2b, and 9.
sensitive cancer cell lines investi-
gated. In addition, as 2b can safely
be assumed to be more acid-stable
than 2a, the former could in fact
offer some advantage over the
latter in vivo (for oral administra-
tion). Based on the tubulin-poly-
merization data shown in Table 1,
2a is a more potent inducer of
tubulin polymerization than 2b,
which indicates that hydrophobic
interactions between the protein
and the N12-carbamate substituent
play an inportant role in the binding
of these compounds to a/b-tubu-
lin.[19] Whether the difference in
Cmpd
EC50 (Tubulin polym.)
IC50 [nm][b]
[mm][a]
A549
HCT-116
PC-3M
KB-31
KB-8511
2a
2b
9
1a
EpoA
3.9ꢁ0.6
5.4ꢁ0.4
9.1ꢁ0.7
5.6ꢁ0.4
4.6 ꢁ0.5
1.9ꢁ0.4
18.0ꢁ3.1
920ꢁ85
130ꢁ24
3.2ꢁ0.5
1.6ꢁ0.5
23.9ꢁ2.9
1009ꢁ71
110ꢁ19
2.2ꢁ0.3
2.3ꢁ0.6
12.3ꢁ1.1
973ꢁ64
126ꢁ22
3.4ꢁ0.4
0.34ꢁ0.15
12.6ꢁ1.0
n.d.[e]
222ꢁ48
> 1000
n.d.[e]
31[c]
105[c]
2.15[d]
1.91[d]
[a] Concentration required to induce 50% of the maximum ab-tubulin polymerization achievable with
the respective compound (10 mm of porcine brain tubulin). Tubulin polymerization was determined by
turbidity measurements at 340 nm (A340).[22] For a given compound concentration, the achievement of an
equilibrium state between soluble and polymerized tubulin is indicated by a stable plateau in A340
.
Maximum tubulin polymerization is reached when increases in compound concentration no longer
result in an increase of the plateau value for A340. Similar maximum values for A340 were observed for all
compounds investigated in this study. [b] IC50 values for inhibition of human cancer cell growth. KB-31,
KB-8511: cervix; A549: lung; HCT-116: colon; PC-3M: prostate. KB-8511 is a P-glycoprotein 170 (P-
gp170)-overexpressing multidrug-resistant subline of the KB-31 parental line. Cells were exposed to
compounds for 72 h. Cell numbers were determined by quantification of protein content of fixed cells by
staining with methylene blue.[23a] For further experimental details see reference [23b]. Values represent
the means of at least three independent experiments (ꢁstandard deviation). [c] Data from
reference [6d]. [d] Data from reference [6a]. [e] Not determined.
tubulin-polymerizing
activity
between 2a and 2b can fully
account for the difference in cellu-
lar potency is unknown at this
The antiproliferative activity of 2a is thus comparable to
that of Epo A, with the exception of the multidrug-resistant
KB-8511 line, where 2a is significantly less potent (vide infra).
Likewise, 2a induces tubulin polymerization in vitro with
potency similar to that of Epo A (Table 1, Figure 1A), which
strongly suggests that inhibition of human cancer cell
proliferation by 2a, as for natural epothilones, is a conse-
quence of interference with microtubule functionality. This
viewis further corroborated by the fact that treatment of
cancer cells with 2a results in cell cycle arrest at G2/M, which
mirrors the effects on the cell cycle observed upon treatment
with Epo A or B[7] (Figure 1B).
Upon analysis of the data presented in Table 1 in more
detail, it also becomes apparent that 2a is > 60 times more
potent against drug-sensitive human cancer cells than the
corresponding parent (natural-side-chain-containing) azathi-
lone 1a. Although the molecular origin of this activity
difference is unclear, it should be noted that the potency
increase observed for 2a over 1a dramatically exceeds the
potency-enhancing effects previously observed for the dime-
thylbenzimidazole side chain in combination with polyketide-
based macrocycles (2–15-fold).[6a,b,11] Equally intriguing is the
Figure 1. A: Compound 2a induces the formation of microtubules in
vitro. [20 mm of purified porcine ab-tubulin was incubated with 20 mm
of 2a for 30 min in BRB80 buffer at RT]. Electron micrograph shows
part of a single long microtubule. Scale bar: 100 nm. B: Cell-cycle
effects of 2a in PC-3M cells (G1 vs. G2/M). 2a (250 nm) was incubated
for 1 h with 2ꢀ105 cells and then removed. Cells were subsequently
grown for 24 h prior to analysis of DNA content with propidium
iodide. Vehicle control (red) shows cells in the G1 phase of the cell
cycle. In the presence of compound 2a (blue line) or Epo A (250 nm;
dotted green line) cells accumulate in G2/M. PI=propidium iodide.
Angew. Chem. Int. Ed. 2006, 45, 5880 –5885
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5883

Communications
Vourloumis, Angew. Chem. 1998, 110, 2120 – 2153; Angew.
Chem. Int. Ed. 1998, 37, 2014 – 2045.
[8] For a previous use of the term “non-natural natural product”
see: L. F. Tietze, H. P. Bell, S. Chandrasekar, Angew. Chem.
2003, 115, 4128 – 4160; Angew. Chem. Int. Ed. 2003, 42, 3996 –
4028.
point; the latter may also be affected by differences in non-
target-related parameters such as cellular uptake or intra-
cellular distribution.
Compared to their effects on drug-sensitive cancer cell
lines, both 2a and 2b exhibit reduced antiproliferative activity
against the P-gp-overexpressing human cervix carcinoma line
KB-8511, which indicates that both compounds are substrates
for the P-gp efflux pump. However, we have recently shown
that the susceptibiliy of polyketide-based epothilone ana-
logues to P-gp-mediated drug efflux can be modulated
through adjustments in compound lipophilicity; this strategy
will also be explored for lead structures 2a and 2b.[21]
In summary, we have achieved the total synthesis of two
representative examples of a newclass of highly potent
microtubule-stabilizing agents, which are based on an aza-
macrolide backbone and which we have termed azathilones.
While the conception of these compounds is closely con-
nected to the structure of natural epothilones (hence the
name “azathilones”), given the degree of structural diver-
gence from the natural epothilone template, they may be
considered as members of a distinct group of “non-natural”
natural products with unique structural features and, as
indicated by some preliminary SAR data, a unique SAR
profile. Both compounds investigated in this study are potent
growth inhibitors of drug-sensitive human cancer cells in vitro
and, thus, should be attractive newlead structures for
anticancer-drug discovery.
[9] For a review, cf., e.g.: C. A. Walsh,Curr. Opin. Chem. Biol. 1999,
3, 598 – 606.
[10] The term “azathilones” is meant to reflect removal of the
epoxide oxygen in natural epothilones with concomitant replace-
ment of C12 by nitrogen. However, the term is used herein
rather loosely, as the benzimidazole-containing analogues 2a
and 2b strictly speaking should not be named azathilones,
because of the absence of a thiazole moiety, from which the
“thil” in epothilone is derived (cf. K. Gerth, N. Bedorf, G. Höfle,
H. Irschik, H. Reichenbach, J. Antibiot. 1996, 49, 560 – 564).
[11] K.-H. Altmann, G. Bold, G. Caravatti, A. Flörsheimer, V.
Guagnano, M. Wartmann, Bioorg. Med. Chem. Lett. 2000, 10,
2765 – 2768: The magnitude of the activity-enhancing effect of
the dimethylbenzimidazole side chain strongly depends on the
exact nature of the polyketide macrocycle. It is significantly
more pronounced in the deoxyepothilone series (Epo C and D)
than for the corresponding analogues of Epo A and B. (Cf. also
references [6a,b]).
[12] a) A. Rivkin, T.-C. Chou, S. J. Danishefsky, Angew. Chem. 2005,
117, 2898 – 2910; Angew. Chem. Int. Ed. 2005, 44, 2838 – 2850;
b) A. Rivkin, F. Yoshimura, G. Fumihiko, A. E. Gabarda, Y. S.
Cho, T.-C. Chou, H. Dong, S. J. Danishefsky, J. Am. Chem. Soc.
2004, 126, 10913 – 10922; c) R. L. Arslanian, L. Tang, S. Blough,
W. Ma, R.-G. Qiu, L. Katz, J. R. Carney, J. Nat. Prod. 2002, 65,
1061 – 1064; d) I. H. Hardt, H. Steinmetz, K. Gerth, F. Sasse, H.
Reichenbach, G. Höfle, J. Nat. Prod. 2001, 64, 847 – 856.
[13] J. D. White, R. G. Carter, K. F. Sundermann, M. Wartmann, J.
Am. Chem. Soc. 2001, 123, 5407 – 5413. Aldehyde 3 was obtained
by Swern oxidation of the corresponding primary alcohol and
directly used in step (a) of Scheme 2 without purification. The
yield reported for this step is based on the assumption that the
oxidation was quantitative.
[14] a) K.-H. Altmann, G. Bold, G. Caravatti, D. Denni, A. Flör-
sheimer, A. Schmidt, G. Rihs, M. Wartmann, Helv. Chim. Acta
2002, 85, 4086 – 4110; b) D. Schinzer, A. Bauer, O. M. Böhm, A.
Limberg, M. Cordes, Chem. Eur. J. 1999, 5, 2483 – 2491.
[15] Olefin 10 was obtained in three steps from the acylsultam
precursor of aldehyde 13 (cf. Ref. [11]). Details of the synthesis
will be described elsewhere. For carboxylic acid 7 alternative
synthetic approaches have been described; cf., e.g., Ref. [12b].
[16] For recent reviews on olefin metathesis cf., e.g.: a) R. H. Grubbs,
Tetrahedron 2004, 60, 7117 – 7140; b) T. M. Trnka, R. H. Grubbs,
Acc. Chem. Res. 2001, 34, 18 – 29.
[17] a) F. Yoshimura, A. Rivkin, A. E. Gabarda, T.-C. Chou, H.
Dong, G. Sukenick, F. F. Morel, R. E. Taylor, S. J. Danishefsky,
Angew. Chem. 2003, 115, 2622 – 2625; Angew. Chem. Int. Ed.
2003, 42, 2518 – 2521; b) K. Biswas, H. Lin, J. T. Njardson, 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 – 9832.
[18] J. Inanaga, K. Hirata, H. Saeki, T. Katsuki, M. Yamaguchi, Bull.
Chem. Soc. Jpn. 1979, 52, 1989 – 1993.
Received: April 6, 2006
Published online: July 27, 2006
Keywords: antitumor agents · epothilones ·
.
microtubule stabilizers · natural products · total synthesis
[1] G. M. Cragg, D. J. Newman, Pure Appl. Chem. 2005, 77, 7 – 24.
[2] B. A. Posner, Curr. Opin. Drug Discovery Dev. 2005, 8, 487 – 494.
[3] M. D. Burke, S. L. Schreiber, Angew. Chem. 2004, 116, 48 – 60;
Angew. Chem. Int. Ed. 2004, 43, 46 – 58.
[4] R. Breinbauer, I. R. Vetter, H. Waldmann, Angew. Chem. 2002,
114, 3002 – 3015; Angew. Chem. Int. Ed. 2002, 41, 2878 – 2890.
[5] a) M. A. Koch, H. Waldmann, Drug Discovery Today 2005, 10,
471 – 483; b) M. A. Koch, A. Schuffenhauer, M. Scheck, S.
Wetzel, M. Casaulta, A. Odermatt, P. Ertl, H. Waldmann, Proc.
Natl. Acad. Sci. USA 2005, 102, 17272 – 17277; c) M. A. Koch,
L.-O. Wittenberg, S. Basu, D. A. Jeyaraj, E. Gourzoulidou, K.
Reinecke, A. Odermatt, H. Waldmann, Proc. Natl. Acad. Sci.
USA 2004, 101, 16721 – 16726.
[6] a) F. Cachoux, T. Isarno, M. Wartmann, K.-H. Altmann,
ChemBioChem 2006, 7, 54 – 57; b) F. Cachoux, T. Isarno, M.
Wartmann, K.-H. Altmann, Angew. Chem. 2005, 117, 7636 –
7640; Angew. Chem. Int. Ed. 2005, 44, 7469 – 7473; c) F.
Cachoux, F. Schaal, A. Teichert, T. Wagner, K.-H. Altmann,
Synlett 2004, 2709 – 2712; d) K.-H. Altmann, A. Flörsheimer, G.
Bold, G. Caravatti, M. Wartmann, Chimia 2004, 58, 686 – 690.
[7] For reviews on epothilones cf., e.g.: a) G. Höfle, H. Reichenbach
in Anticancer Agents from Natural Products (Eds.: G. M. Cragg,
D. G. I. Kingston, D. J. Newman), CRC, Boca Raton, 2005,
pp. 413 – 450; b) K.-H. Altmann, Curr. Pharm. Des. 2005, 11,
1595 – 1613; c) K.-H. Altmann, Org. Biomol. Chem. 2004, 2,
2137 – 2152; d) C. R. Harris, S. J. Danishefsky, J. Org. Chem.
1999, 64, 8434 – 8456; e) K. C. Nicolaou, F. Roschangar, D.
[19] SAR studies on azathilones
1 had also included the N-
unsubstituted version of these compounds, which proved to be
completely inactive. For this reason the free amine correspond-
ing to 2a and 2b (N12-unsubstituted) was not investigated in the
current study.
[20] R. M. Williams, J. Cao, H. Tsujishima, Angew. Chem. 2000, 112,
2640 – 2644; Angew. Chem. Int. Ed. 2000, 39, 2540 – 2544.
[21] F. Cachoux, T. Isarno, M. Wartmann, K.-H. Altmann, Synlett
2006, 1384 – 1388.
5884
www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 5880 –5885

Angewandte
Chemie
[22] E. Hamel, A. A. del Campo, J. Lustbader, C. M. Lin, Biochem-
istry 1983, 22, 1271 – 1279.
[23] a) T. Meyer, U. Regenass, D. Fabbro, E. Alteri, J. Rösel, M.
Mꢁller, G. Caravatti, A. Matter, Int. J. Cancer 1989, 43, 851 – 856;
b) K. C. Nicolaou, R. Scarpelli, B. Bollbuck, B. Werschkun,
M. M. Pereira, M. Wartmann, K.-H. Altmann, D. Zaharevitz, R.
Gussio, P. Giannakakou, Chem. Biol. 2000, 7, 593 – 599.
[24] F. Feyen, K.-H. Altmann, unpublished results.
[25] J. Gertsch, F. Feyen, K.-H. Altmann, unpublished results.
Angew. Chem. Int. Ed. 2006, 45, 5880 –5885
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5885