Synthesis and biological activity of new functionalized epothilones for prodrug design and tumor targeting

Article abstract of DOI:10.2533/chimia.2010.136

Epothilones are potent antiproliferative agents, which have served as successful lead structures for anticancer drug discovery. However, their therapeutic efficacy would benefit greatly from an increase in their selectivity for tumor cells, which may be a

Full text of DOI:10.2533/chimia.2010.136

136ꢀ CHIMIAꢀ2010,ꢀ64,ꢀNo.ꢀ3ꢀ
doi:10.2533/chimia.2010.136ꢀꢀ
Laureates: awards and Honors, sCs FaLL Meeting 2009
Chimiaꢀ64ꢀ(2010)ꢀ136–139ꢀ ©ꢀSchweizerischeꢀChemischeꢀGesellschaft
Synthesis and Biological Activity of New
Functionalized Epothilones for Prodrug
Design and Tumor Targeting
SilviaꢀAnthoineꢀDietrich§,ꢀLindaꢀRiediker,ꢀJürgꢀGertscha,ꢀandꢀKarl-HeinzꢀAltmann*ꢀ
§SCSꢀMettler-ToledoꢀAwardꢀWinnerꢀ(OralꢀPresentation)
Abstract:ꢀEpothilonesꢀareꢀpotentꢀantiproliferativeꢀagents,ꢀwhichꢀhaveꢀservedꢀasꢀsuccessfulꢀleadꢀstructuresꢀforꢀ
anticancerꢀdrugꢀdiscovery.ꢀHowever,ꢀtheirꢀtherapeuticꢀefficacyꢀwouldꢀbenefitꢀgreatlyꢀfromꢀanꢀincreaseꢀinꢀtheirꢀse-
lectivityꢀforꢀtumorꢀcells,ꢀwhichꢀmayꢀbeꢀachievedꢀthroughꢀconjugationꢀwithꢀaꢀtumor-targetingꢀmoiety.ꢀThreeꢀnovelꢀ
epothiloneꢀanalogsꢀbearingꢀvariouslyꢀfunctionalizedꢀbenzimidazoleꢀsideꢀchainsꢀwereꢀsynthesizedꢀusingꢀaꢀstrategyꢀ
basedꢀonꢀpalladium-mediatedꢀcouplingꢀandꢀmacrolactonization.ꢀTheꢀsynthesisꢀofꢀtheseꢀcompoundsꢀisꢀdescribedꢀ
andꢀtheirꢀin vitroꢀbiologicalꢀactivityꢀisꢀdiscussedꢀwithꢀrespectꢀtoꢀtheirꢀinteractionsꢀwithꢀtheꢀtubulin/microtubuleꢀ
systemꢀandꢀtheꢀinhibitionꢀofꢀhumanꢀcancerꢀcellꢀproliferation.ꢀTheꢀadditionalꢀfunctionalꢀgroupsꢀmayꢀbeꢀusedꢀtoꢀ
synthesizeꢀconjugatesꢀofꢀepothiloneꢀderivativesꢀwithꢀaꢀvarietyꢀofꢀtumor-targetingꢀmoieties.
Keywords:ꢀAntiproliferativeꢀagentꢀ·ꢀDrugꢀconjugateꢀ·ꢀEpothiloneꢀ·ꢀTotalꢀsynthesisꢀ·ꢀTumorꢀtargeting
1. Introduction
Fig.ꢀ1.ꢀ
R
R
O
S
N
S
N
Epothilones (Fig. 1) are microtubule-stabi-
lizing agents[1] which display very potent
antiproliferative activity in vitro[2] and high
in vivo antitumor activity in a variety of
human tumor models.[3] They have been at
the focus of extensive research for the past
decade, which has led to the synthesis of
a large number of diverse epothilone ana-
logs;[4] several compounds of this family
have reached clinical development and one
has been approved recently for breast can-
cer therapy.[5,6]
HO
HO
O
O
O
O
OH
O
OH
O
R = H
Epothilone C
R = H
Epothilone A
R = CH3 Epothilone B
R = CH3 Epothilone D
parison to the majority of healthy tissues. gates of antiproliferative drugs for tumor
The latter are inevitably affected at some targeting.[7]
Notwithstanding their promising prop-
erties, epothilones lack any inherent se-
lectivity for malignant cells with respect
to normal ones. As for many other cancer
chemotheraputics, their clinical efficacy
relies heavily on the increased prolifera-
tion rate of many types of tumors in com-
level of cytotoxicity during therapy, which
The natural epothilones framework,
often leads to significant side effects; in however, offers little opportunity for easy
the case of epothilones, these include neu- conjugation chemistry. Two obvious func-
rological and gastrointestinal toxicities.[3] tionalities that may lend themselves to de-
The resulting need to balance efficacy rivatization are the two hydroxyl groups at
against toxicity in oncological treatments the C(3) and C(7) positions, which might
is a well-known problem, which often lim- be converted, for instance, to esters or car-
its the dosage that can be administered, and bonates. Functionalization of these groups,
may ultimately lead to failure of therapy.
however, appeared likely to be difficult,
To reduce their side effects, it would given that they are secondary alcohols in
be very desirable to confer to epothilones a relatively crowded steric environment.
an intrinsic ability to discriminate between Moreover, derivatization at these positions
healthy and malignant cells. This may be entails the risk of side reactions, such as
achieved for instance by structural modi- macrocycle opening through retro-aldol
fications affecting cellular uptake, in order chemistry or elimination reactions. The ar-
to produce increased transport of the drug omatic side chain would appear compara-
into malignant cells. Conjugation of an ep- tively less problematic for chemical modi-
othilone derivative as a cytotoxic ‘warhead’ fication, but it lacks a suitable functional
to a tumor-targeting moiety may result in group for derivatization; however, a vari-
such preferential accumulation of the con- ety of side-chain-modified epothilones are
jugate into cancer cells. This approach has known,[8]including analogs with additional
been already applied in the past, using a nucleophilicgroupsaspartofthesidechain,
variety of macromolecular and small-mol- and it has been shown that even significant
ecule targeting moieties to prepare conju- alterations of this part of the structure are
*Correspondence: Prof.ꢀDr.ꢀK.-H.ꢀAltmann
SwissꢀFederalꢀInstituteꢀofꢀTechnologyꢀ(ETH)ꢀZürich
InstituteꢀofꢀPharmaceuticalꢀSciences
ETHꢀHönggerberg,ꢀHCIꢀHꢀ405ꢀ
CH-8093ꢀZürich
Tel.:ꢀ+41ꢀ44ꢀ633ꢀ73ꢀ90
Fax:ꢀ+41ꢀ44ꢀ633ꢀ13ꢀ60ꢀ
E-mail:ꢀkarl-heinz.altmann@pharma.ethz.ch
aInstituteꢀofꢀBiochemistryꢀandꢀMolecularꢀMedicine
UniversityꢀofꢀBern,ꢀBern

Laureates: awards and Honors, sCs FaLL Meeting 2009ꢀ
CHIMIAꢀ2010,ꢀ64,ꢀNo.ꢀ3ꢀ 137
often well tolerated in terms of activity.[9]
Therefore, a promising approach for the
development of tumor-targeted epothi-
lones may consist in the utilization of side
chain-modified analogs that incorporate an
additional functionality that is amenable
Fig.ꢀ2.ꢀ
O
OH
NH2
HN
O
N
N
N
HO
HO
N
O
O
O
O
O
O
OH
OH
to derivatization and conjugation. Ana-
logs incorporating a benzimidazole-based
side chain have previously been shown to
exhibit high cytotoxicity in combination
with the epothilone D core macrocycle,[10]
and the benzimidazole moiety offers a site
for further modification at the N(1) posi-
tion. Therefore, we envisioned compounds
1–3 (Fig. 2) as targets for the subsequent
preparation of epothilone conjugates; in
addition to providing the possibility of dif-
ferent conjugation chemistries, the differ-
ent functionalities that decorate the N(1)
appendage would offer further insight into
the SAR of benzimidazole-based epothi-
lone analogues. Should these derivatives
(i.e. 1–3) maintain high cytotoxicity, as we
hoped would be the case, they would open
the way to the easy synthesis of a large ar-
ray of epothilone conjugates with a variety
of targeting moieties.
1
2
OH
N
N
HO
O
O
O
OH
3
Schemeꢀ1.
R
R
N
N
N
N
HO
TBSO
OH
COOH
O
O
O
O
OH
OTBS
1 R = NH2
4a R = NHBoc
4b R = OTBS
2 R = NHCOCH2CH2COOH
3 R = OH
R
TBSO
N
+
I
O
N
O
O
OTBS
OTES
2. Results and Discussion
5
6 R = NHBoc
7 R = OTBS
The synthesis of epothilone analogs
1–3 was based on a strategy previously
applied to the preparation of other side-
chain-modified epothilones,[11] and relied
on the common intermediate 5, to which
the corresponding side-chain moieties 6
and 7 were connected via Suzuki-Miyaura
palladium-mediated coupling (Scheme 1).
Selective deprotection afforded seco-acids
4a–b, then Yamaguchi macrolactoniza-
tion[12] was applied to complete the epothi-
lone structure; full deprotection afforded
1 and 3, while 2 was synthesized through
late-stage functionalization of the protect-
ed lactone precursor to 1 (vide infra).
R
a)-c) or
d), e)
f)-h) or
i)-m)
H
F
N
N
N
R2
NO2
HOOC
NH2
R1OOC
O
8a R1 = CH3, R2 = NHBoc
8b R1 = H, R2 = OH
9a R = NHBoc
9b R = OTBS
R
R
R
n)
o), p)
q)
N
N
N
N
N
O
N
I
OH
OTES
OTES
Synthesis of the benzimidazole-con-
taining vinyl iodides 6 and 7 started from
4-fluoro-3-nitrobenzoic acid (Scheme 2).
Despite minor differences in the protection
sequence, the synthesis was conducted in
a very similar manner for both derivatives,
with the introduction of the ethylendiamine
or 2-aminoethanol handle at a very early
stage through nucleophilic aromatic sub-
stitution. Subsequent reduction of the nitro
group and cyclization with triethylorthoac-
etate easily afforded the desired function-
alized benzimidazoles in very good yields
(88% in four steps and 69% in three steps
respectively). Interestingly, Swern oxida-
tion afforded aldehyde 9b smoothly from
the corresponding alcohol, while it failed
completely to convert its amino-analog to
9a; the latter was obtained instead in excel-
lent yield (96%) using manganese oxide as
the oxidizing agent.
10a R = NHBoc
10b R = OTBS
11a R = NHBoc
11b R = OTBS
6 R = NHBoc
7 R = OTBS
Schemeꢀ2.ꢀa)ꢀH2SO4,ꢀMeOH,ꢀ65ꢀ°C,ꢀ6ꢀh,ꢀ97%;ꢀb)ꢀBocNHCH2CH2NH2,ꢀDCM,ꢀtriethylamine,ꢀr.t.,ꢀ25ꢀ
h,ꢀ96%;ꢀc)ꢀH2,ꢀPd/C,ꢀMeOH,ꢀr.t.,ꢀ17ꢀh,ꢀ99%;ꢀd)ꢀethanolamine,ꢀMeOH,ꢀr.t.,ꢀ4ꢀh,ꢀ89%;ꢀe)ꢀH2,ꢀPd/C,ꢀ
EtOH,ꢀr.t.,ꢀ40ꢀmin,ꢀ83%;ꢀf)ꢀtriethylorthoacetate,ꢀEtOH,ꢀreflux,ꢀ19.5ꢀh,ꢀ96%;ꢀg)ꢀDIBAL-H,ꢀDCM,ꢀ–78ꢀ
°Cꢀ→ꢀr.t.,ꢀ17ꢀh,ꢀ78%;ꢀh)ꢀMnO2,ꢀDCM,ꢀ40ꢀ°C,ꢀ1ꢀh,ꢀ96%;ꢀi)ꢀtriethylorthoacetate,ꢀEtOH,ꢀreflux,ꢀ2ꢀh,ꢀ
94%;ꢀj)ꢀH2SO4,ꢀMeOH,ꢀ65ꢀ°C,ꢀ27ꢀh,ꢀ98%;ꢀk)ꢀTBSCl,ꢀimidazole,ꢀDMF,ꢀr.t.,ꢀ2.5ꢀh,ꢀ90%;ꢀl)ꢀDIBAL-H,ꢀ
DCM,ꢀ–78°Cꢀ→ꢀr.t.,ꢀ26ꢀh,ꢀ82%;ꢀm)ꢀ(COCl)2,ꢀDMSO,ꢀDCM,ꢀ–78ꢀ°C,ꢀ1ꢀh,ꢀ67%;ꢀn)ꢀi.ꢀ(–)-DIP-Cl,ꢀallyl-
MgBr,ꢀEt2O,ꢀ0ꢀ°C,ꢀ1ꢀh,ꢀthenꢀ–78ꢀ°Cꢀ(solutionꢀA);ꢀii.ꢀ9,ꢀEt2O,ꢀ–100ꢀ°C,ꢀdropwiseꢀadditionꢀofꢀsolutionꢀ
A,ꢀthenꢀ–100ꢀ°C,ꢀ2ꢀh,ꢀ89%ꢀ(10a)ꢀandꢀ95%ꢀ(10b),ꢀeeꢀ94%ꢀ(10a)ꢀandꢀ91%ꢀ(10b)ꢀdeterminedꢀthroughꢀ
Mosherꢀesterꢀanalysis;ꢀo)ꢀTESCl,ꢀimidazole,ꢀDMAP,ꢀDMF,ꢀr.t.,ꢀ4ꢀh,ꢀ98%ꢀ(a)ꢀandꢀ92%ꢀ(b);ꢀp)ꢀOsO4,ꢀ
2,6-lutidine,ꢀNaIO4,ꢀDMF/water,ꢀr.t.,ꢀ23ꢀh,ꢀ74%ꢀ(11a)ꢀandꢀ1.5ꢀh,ꢀ74%ꢀ(11b);ꢀq)ꢀ[Ph3PCH(CH3)I]I,ꢀNa-
HMDS,ꢀTHF,ꢀ–78ꢀ°Cꢀ→ꢀ–30ꢀ°C,ꢀ1ꢀh,ꢀthenꢀ–78ꢀ°C,ꢀ11,ꢀ7ꢀh,ꢀ42%ꢀ(6)ꢀandꢀ4ꢀh,ꢀ37%ꢀ(7).
The next step was the crucial introduc- tion with aldehyde 9a; stereoselectivity,
tion of the stereogenic center at the future however, proved to be rather disappointing
position C(15) of the epothilone. Our ini- with a diastereomeric ratio of only 2:1.
tial approach made use of Oppolzer’s bor- We therefore turned our attention towards
nane sultam auxiliary[13] in an aldol reac- Brown allylation[14] of 9 to install the de-

138ꢀ CHIMIAꢀ2010,ꢀ64,ꢀNo.ꢀ3ꢀ
Laureates: awards and Honors, sCs FaLL Meeting 2009
NHBoc
OTBS
N
N
TBSO
a), b)
N
a)
5
TBSO
TBSO
N
COOCH3
OTBS
OH
OTES
O
COOH
OTBS
COOCH3
OTBS
O
O
5
4a
OTBS
NHBoc
N
N
N
b)
c)-e)
c)
d)
3
1
2
TBSO
N
TBSO
OTES
COOH
OTBS
O
O
e)-g)
O
O
OTBS
13
12
Schemeꢀ3.ꢀa)ꢀi.ꢀ9-BBN,ꢀTHF,ꢀr.t.,ꢀ3ꢀhꢀ(solutionꢀA);ꢀii.ꢀAsPh3,ꢀCsCO3,ꢀ
[PdCl2(dppf)]⋅DCM,ꢀDMF,ꢀwater,ꢀ6,ꢀsolutionꢀA,ꢀ–5ꢀ°Cꢀ→ꢀr.t.,ꢀ12ꢀh,ꢀ83%;ꢀb)ꢀ
i.ꢀLiOH·H2O,ꢀdioxane/water,ꢀ60ꢀ°C,ꢀ11.5ꢀh;ꢀii.ꢀDCM,ꢀwater,ꢀHCl,ꢀpHꢀ2,ꢀr.t.,ꢀ
6ꢀh,ꢀ85%;ꢀc)ꢀi.ꢀ2,4,6-trichlorobenzoylchloride,ꢀtriethylamine,ꢀTHF,ꢀ–10ꢀ°Cꢀ
→ 0ꢀ°C,ꢀ1ꢀh;ꢀii.ꢀdilutionꢀwithꢀtolueneꢀ(solutionꢀA);ꢀiii.ꢀDMAP,ꢀtoluene,ꢀslowꢀ
additionꢀofꢀsolutionꢀAꢀoverꢀ3ꢀh,ꢀr.t.,ꢀthenꢀr.t.,ꢀ2ꢀh,ꢀ57%;ꢀd)ꢀTFA,ꢀDCM,ꢀ0ꢀ
°Cꢀ→ꢀr.t.,ꢀ2ꢀh,ꢀ44%ꢀafterꢀHPLCꢀpurification;ꢀe)ꢀZnBr2,ꢀDCM,ꢀ0ꢀ°Cꢀtoꢀ4ꢀ°C,ꢀ
72ꢀh,ꢀ57%;ꢀf)ꢀsuccinicꢀanhydride,ꢀdiisopropylethylamine,ꢀDMF,ꢀr.t.,ꢀ2ꢀh,ꢀ
92%;ꢀg)ꢀTFA,ꢀDCM,ꢀ0ꢀ°Cꢀ→ꢀr.t.,ꢀ5ꢀh,ꢀ23%ꢀafterꢀHPLCꢀpurification.
Schemeꢀ4.ꢀa)ꢀi.ꢀ9-BBN,ꢀTHF,ꢀr.t.,ꢀ3ꢀhꢀ(solutionꢀ
A);ꢀii.ꢀAsPh3,ꢀCsCO3,ꢀ[PdCl2(dppf)]⋅DCM,ꢀ
DMF,ꢀwater,ꢀ7,ꢀsolutionꢀA,ꢀ–10ꢀ°Cꢀ→ꢀr.t.,ꢀ2.5ꢀ
h,ꢀ77%;ꢀb)ꢀLiOH·H2O,ꢀisopropanol/water,ꢀ
r.t.,ꢀ72ꢀh,ꢀ26%;ꢀc)ꢀDCM/isopropanol/1Mꢀ
HClꢀ2:1:1,ꢀr.t,ꢀ2.5ꢀh,ꢀnotꢀpurified;ꢀd)ꢀi.ꢀ
2,4,6-trichlorobenzoylchloride,ꢀtriethylamine,ꢀ
THF,ꢀ–10ꢀ°Cꢀ→ꢀ0ꢀ°C,ꢀ1ꢀh;ꢀii.ꢀdilutionꢀwithꢀ
tolueneꢀ(solutionꢀA);ꢀiii.ꢀDMAP,ꢀtoluene,ꢀslowꢀ
additionꢀofꢀsolutionꢀAꢀoverꢀ1.5ꢀh,ꢀr.t.,ꢀ54%ꢀoverꢀ
twoꢀsteps;ꢀe)ꢀHF·Py,ꢀTHF,ꢀr.t.,ꢀ26ꢀh,ꢀ84%.
sired chirality and we obtained a much achieved smoothly in 92% yield; deprotec- these mixtures through flash chromatog-
more satisfying enantiomeric excess for tion under the same conditions used for 1 raphy, and sufficient material was eventu-
both intermediates 10a and 10b. Oxida- afforded 2 in a moderate but still acceptable ally obtained to complete the synthesis.
tion of the latter with osmium tetroxide/ yield of 23% after HPLC purification.
Selective cleavage of the TES ether in 13
periodate followed by Wittig reaction of The synthesis of 3 was planned along was achieved under acidic conditions; the
the resulting aldehydes with iodoethyltri- the same sequence of reactions that led to resulting seco-acid was cyclized success-
phenylphosphonium iodide[15a] success- 1 (Scheme 4); after a successful palladium- fully underYamaguchi conditions, then de-
fully completed the synthesis of 6 and 7. mediated coupling of 5 and 7, however, the protection with hydrofluoric acid/pyridine
The Wittig step gave the desired Z product deprotection step proved to be unselective, afforded the desired analog 3.
exclusively, albeit in modest yields. This with silyl ether cleavage occurring at both
Compounds 1–3 were evaluated in vitro
is consistent with a number of literature position 15 and the ethanolamino moiety for their effect on tubulin polymerization
reports on this type of reaction, and there during the basic hydrolysis of the methyl and on human cancer cell growth against
is some indication that it may be inherently ester. Despite our attempts to optimize the several cell lines (Table 1). While 1 and 2
limited to about 50% yield.[15b] Despite its reaction conditions, the result was always induce tubulin polymerization with simi-
moderate yield, the reaction is valuable a mixture of derivatives with a modest lar potency and in the same range as Epo
for the present synthesis because it affords yield of the desired product; under milder A, their antiproliferative activity is signifi-
straightforward access to the Z-configured saponification conditions, one of the side cantly different. Compound 1 has IC50 only
methyl vinyliodide motif, which is other- products even presented a free alcohol at about three times higher than that of Epo
wise is rather cumbersome to access.
the ethanolamino moiety, while both the A, while for 2 this value is about 20 times
With 6 in hand, the backbone of 1 TES ether and the methyl ester were still higher than for Epo A. This discrepancy
could be assembled via Suzuki-Miyaura intact. However, 13 could be isolated from between the effects displayed on tubulin
coupling with 5 (Scheme 3); subsequent
basic hydrolysis of the methyl ester pro-
ceeded smoothly and also allowed selec-
tive cleavage of the silyl ether at position
15 by simply prolonging the acidic workup
step. The resulting seco-acid was cyclized
in a Yamaguchi reaction[12] and acidic de-
protection of all remaining groups in one
step afforded 1 in a satisfactory 25% yield
over two steps after HPLC purification.
Succinic acid derivative 2 was synthe-
sized from 12 through selective removal of
the BOC-group with zinc bromide. While
partial loss of TBS groups also occured,
sufficiently diluted reaction conditions pro-
vided roughly 60% of the desired TBS-pro-
tected, free amine product. Acylation of the
amino group with succinic anhydride was
Tableꢀ1.ꢀTubulin-polymerizingꢀandꢀantiproliferativeꢀactivityꢀofꢀepothilonesꢀ1,ꢀ2ꢀandꢀ3
IC50ꢀ[nM]b
EC50ꢀtubulinꢀpolymerization
Compound
[µM]a
MCF-7
A549
HCT116ꢀ
n.ꢀd.c
1
4.3ꢀ ꢀ0.8d
10.5ꢀ ꢀ3.0d
13.0ꢀ ꢀ4.8d
2
4.1ꢀ ꢀ0.5d
n.ꢀd.c
65ꢀ ꢀ12ꢀd
108ꢀ ꢀ14ꢀd
0.35ꢀ ꢀ0.019
5.0ꢀ ꢀ1.4ꢀ
n.ꢀd.c
3
0.52ꢀ ꢀ0.018
2.9ꢀ ꢀ0.3d
0.38ꢀ ꢀ0.029
2.8ꢀ ꢀ0.4d
0.16ꢀ ꢀ0.01d
EpothiloneꢀA
3.9ꢀ ꢀ0.6d
EpothiloneꢀB 3.0ꢀ ꢀ0.3d
0.33ꢀ ꢀ0.01d
0.34ꢀ ꢀ0.03d
aConcentrationꢀrequiredꢀtoꢀinduceꢀ50%ꢀofꢀtheꢀmaximumꢀtubulinꢀpolymerizationꢀachievableꢀwithꢀ
theꢀrespectiveꢀcompoundꢀ(10ꢀµMꢀofꢀporcineꢀbrainꢀtubulin).ꢀbIC50-valuesꢀforꢀhumanꢀcancerꢀcellꢀ
growthꢀinhibition.ꢀMCF-7:ꢀbreast;ꢀA549:ꢀlung;ꢀHCT116:ꢀcolon.ꢀcn.ꢀd.ꢀ=ꢀnotꢀdetermined.ꢀdSeeꢀref.ꢀ
[16].

Laureates: awards and Honors, sCs FaLL Meeting 2009ꢀ
CHIMIAꢀ2010,ꢀ64,ꢀNo.ꢀ3ꢀ 139
and on cells is not unprecedented,[8b,16]
and may reflect differences in the cellular
uptake of these compounds, but its precise
reasons remain unclear. Strikingly, deriva-
tive 3 is significantly more active than 1,
with an IC50 value 20 to 40 times lower,
coming closer to the values of the more
potent Epothilone B. Although the chemi-
cal structures of 1 and 3 would seem very
similar, it appears that a hydroxyl group in
this particular position is much more ad-
vantageous than an amine.
200021-107876 and 205320-117594) and ETH
Zürich.
Received: January 27, 2010
[1] D. M. Bollag, P. A. McQueney, J. Zhou, O.
Hensens, L. Koupal, J. Liesch, M. Goez, E.
Lazarides, C. M. Woods, Cancer Res. 1995, 55,
2325.
[2] a) R. J. Kowalski, P. Giannakakou, E. Hamel, J.
Biol. Chem. 1997, 272, 2534; b) K.-H. Altmann,
M. Wartmann, T. O’Reilly, Biochim. Biophys.
Acta Rev. Cancer 2000, 1470, M79.
[3] S. Goodin, M. P. Kane, E. H. Rubin, J. Clin.
Onc. 2004, 22, 2015.
The reduction in antiproliferative activ-
ity of compounds 1 and 2 in comparison
with natural epothilones is somewhat more
pronounced than we had anticipated, based
on the nanomolar cytotoxicity displayed
by the dimethylbenzimidazole analog of
epothilone D.[10] The increased bulkiness
of these derivatives might play a role, al-
though literature precedents would indi-
cate a good tolerance to large substituents
in this region of the molecule.[9] Besides,
comparison of 1 with 3 indicates that steric
hindrance is very unlikely to be the only
factor. In addition to their size, the acid-
base properties of the newly introduced
functional groups should also be consid-
ered, as they may have a significant influ-
ence on the transport of these analogs into
and inside cells. This may be particularly
important in the case of 1 and 2, which
possess ionizable groups in addition to
the benzimidazole nitrogen, whose precise
pKa is unknown in these compounds.
Overall, however, compounds 1–3 all
display potent antiproliferative activity,
and, therefore, are interesting candidates
for the development of drug conjugates for
tumor targeting.
[4] For a review on the epothilone SAR see
for example: K.-H. Altmann, B. Pfeiffer, S.
Arseniyadis, B. A. Pratt, K. C. Nicolaou,
ChemMedChem 2007, 2, 396.
[5] E. Kaminskas, X. Jiang, R. Aziz, J. Bullock, R.
Kasliwal, R. Harapanhalli, S. Pope, R. Sridhara,
J. Leighton, B. Booth, R. Dagher, R. Justice, R.
Pazdur, Clin. Cancer Res. 2008, 14, 4378.
[6] For a recent review on epothilone clinical trials:
M. Harrison, C. Swanton, Exp. Opin. Invest.
Drugs 2008, 17, 523.
[7] For a review on this topic see for example:
a) S. Jaracz, J. Chen, L. V. Kuznetsova, I.
Ojima, Bioorg. Med. Chem. 2005, 13, 5043
and references cited therein; for more recent
examples see, for instance: b) P. S. Low, W. A.
Henne, D. D. Doorneweerd, Acc. Chem. Res.
2008, 41, 120; c) G. Pasut, F. M. Veronese,
Prog. Polym. Sci. 2007, 32, 933.
[8] a) See, for example: K. C. Nicolaou, R.
Scarpelli, B. Bollbuck, B. Werschkun, M. M.
A. Pereira, M. Wartmann, K.-H. Altmann, D.
Zaharevitz, R. Gussio, P. Giannakakou, Chem.
Biol. 2000, 7, 593; b) G. Bold, S. Wojeik, G.
Caravatti, R. Lindauer, C. Stierlin, J. Gertsch,
M. Wartmann, K.-H. Altmann, ChemMedChem
2006, 1, 37.
[9] K. C. Nicolaou, B. A. Pratt, S. Arseniyadis, M.
Wartmann, A. O’Brate, ChemMedChem 2006,
1, 41.
[10] K.-H. Altmann, G. Bold, G. Caravatti, A.
Flörsheimer, V. Guagnano, M. Wartmann,
Bioorg. Med. Chem. 2000, 10, 2765.
[11] a) K.-H. Altmann, G. Bold, G. Caravatti, D.
Denni, A. Flörsheimer, A. Schmidt, G. Rihs, M.
Wartmann, Helv. Chim. Acta 2002, 85, 4086;
b) F. Cachoux, T. Isarno, M. Wartmann, K.-H.
Altmann, ChemBioChem 2006, 7, 54.
3. Conclusions
[12] J. Inanaga, K. Hirata, H. Saeki, T. Katsuki, M.
Yamaguchi, Bull. Chem. Soc. Jpn. 1979, 52,
1989.
[13] W. Oppolzer, Pure Appl. Chem. 1990, 62, 1241.
[14] H. C. Brown, P. K. Jadhav, J. Am. Chem. Soc.
1983, 105, 2092.
[15] a) J. Chen, T. Wang, K. Zhao, Tetrahedron Lett.
1994, 35, 2827; b) H. Arimoto, M. D. Kaufman,
K. Kobayashi, Y. Qiu, A. B. Smith III, Synlett
1998, 765.
[16] S. Anthoine Dietrich, R. Lindauer, C. Stierlin, J.
Gertsch, R. Matesanz. S. Notararigo, J. F. Dìaz,
K.-H. Altmann, Chem. Eur. J. 2009, 15, 10144.
We have synthesized three novel side-
chain-modified Epo D analogs and we
have established their tubulin-polymeriz-
ing and antiproliferative activity in vitro.
Due to the additional functional group
they bear, these analogs may now be used
to prepare conjugates with appropriately
functionalized tumor-targeting molecules.
The different functional groups displayed
by compounds 1–3 as an appendix to their
benzimidazole side chain offer a signifi-
cant degree of flexibility in terms of the
conjugation chemistry that may be chosen.
The synthesis of tumor-targeted conjugates
based on 1–3 is currently in progress in our
laboratory, and the results of these efforts
will be disclosed in future publications.
Acknowledgments
The excellent technical assistance by Kurt
Hauenstein is gratefully acknowledged. This
work was supported by the Swiss National
Science Foundiation (S.A.D.; project nos