Natural products are biologically validated starting points in structural space for compound library development: Solid-phase synthesis of dysidiolide-derived phosphatase inhibitors

Article abstract of DOI:10.1002/1521-3773(20020118)41:2<307::AID-ANIE307>3.0.CO;2-1

A multistep solid-phase synthesis was used successfully to prepare a series of analogues (see picture) of the protein phosphatase inhibitor dysidiolide. The natural product and its analogues were prepared on a solid support over 8 - 12 linear steps with o

Full text of DOI:10.1002/1521-3773(20020118)41:2<307::AID-ANIE307>3.0.CO;2-1

COMMUNICATIONS
[
[
5] C. A. Bischoff, E. Frˆhlich, Chem. Ber. 1907, 40, 2779 ± 2790.
6] J. T. Adams, C. R. Hauser, J. Am. Chem. Soc. 1944, 66, 1220 ± 1222; R.
Levine, J. A. Conroy, J. T. Adams, C. R. Hauser, J. Am. Chem. Soc.
Natural Products Are Biologically Validated
Starting Points in Structural Space for
Compound Library Development: Solid-Phase
Synthesis of Dysidiolide-Derived Phosphatase
Inhibitors**
1
945, 67, 1510 ± 1512.
[
7] Mononuclear metallacoronates are generated with polyethylene
glycol spacered bis-1,3-diketones. See: Y. Kobuke, Y. Satoh, J. Am.
Chem. Soc. 1992, 114, 789 ± 790.
[
8] For double-stranded intertwined infinite linear silver coordination
networks based on bis-monodentate ligands with polyethylene glycol
spacer, see: B. Schmaltz, A. Jouaiti, M. W. Hosseini, A. De Cain,
Chem. Commun. 2001, 1242 ± 1243.
Dirk Brohm, Susanne Metzger, AjayBhargava,
Oliver M¸ller, Folker Lieb, and Herbert Waldmann*
[
9] Crystal data for K-3: C48
H
51Cu
2
KO16, M
r
ˆ 1050.07; crystal dimensions
The combinatorial synthesis of compound libraries on
polymeric supports is at the heart of protein ligand and
inhibitor discovery, in particular in the development of new
drugs and chemical tools for the studyof biological processes.
To achieve high efficiencyin this process, powerful and
alternative strategies for the design of compound libraries are
of paramount importance. Herein, a structure-based approach
to this fundamental problem is outlined. The keyfeature is to
employthe structural frameworks of biologicallyactive
natural products, which are evolutionarilyselected for binding
to specific protein domains, as a guiding principle for library
development. Furthermore, it is demonstrated that the key
synthetic challenge posed by this approach, the multistep
solid-phase synthesis of natural products and their analogues,
can successfullybe met.
3
0
.25  0.15  0.15 mm ; monoclinic, space group C2/c, a ˆ 984.0(2),
3
b ˆ 2070.8(4), c ˆ 2475.0(5) pm, b ˆ 98.39(3)8, Vˆ 4989.0(17) ä ; Z ˆ
À3
4
; F(000) ˆ 2192, 1calcd ˆ 1.398 gcm . Diffractometer: Nonius Kap-
paCCD, MoKa radiation (l ˆ 0.71073 ä); Tˆ 173(2) K; graphite
monochromator; q range [8] 2.14 < q < 27.50; section of the reciprocal
lattice: À12 ꢀ h ꢀ 12, À26 ꢀ k ꢀ 26, À32 ꢀ l ꢀ 32; of 20042 measured
reflections, 5702 were independent and 4339 with I > 2s(I); linear
À1
absorption coefficient 1.004mm . The structure was solved bydirect
methods using SHELXS-97 and refinement with all data (321
parameters) byfull-matrix least-squares on F using SHELXL-97;
2
[10]
all non-hydrogen atoms were refined anisotropically; R1 ˆ 0.0607 for
À3
I > 2s(I) and wR2 ˆ 0.2074 (all data); largest peak (1.358 eä ) and
À3 [11]
hole (À0.788 eä ).
[
10] G. M. Sheldrick, C. Kr¸ger, P. Goddard, Crystallographic Computing
3
, Oxford UniversityPress, Oxford, 1985, p. 175; G. M. Sheldrick,
SHELXS-97, Program for Crystal Structure Solution, Universit‰t
Gˆttingen, 1997; G. M. Sheldrick, SHELXL-97, Program for Crystal
Structure Refinement, Universit‰t Gˆttingen, 1997.
Proteins can be regarded as modularlybuilt biomolecules
assembled from individual domains as building blocks. Since
the total number of all available protein domains appears to
[
11] Crystallographic data (excluding structure factors) for the structures
reported in this paper have been deposited with the Cambridge
Crystallographic Data Centre as supplementary publication nos.
[1]
CCDC-168265 (3, M ˆ K) and CCDC-168266 [(4)
n
]. Copies of the
be fairlylimited,
it has to be expected that in newly
data can be obtained free of charge on application to CCDC, 12 Union
Road, Cambridge CB21EZ, UK (fax: (‡44)1223-336-033; e-mail:
deposit@ccdc.cam.ac.uk).
discovered proteins with widelyvar yi ng function and activity
the same modules (i.e. domains) or close relatives will be
found repeatedlyin var yi ng combinations and arrangements
as structure- and function-determining entities. Thus, a keyto
the efficient discoveryof new ligands and inhibitors for known
and, in particular, for newlydiscovered proteins is to identify
compound classes alreadybiologicallyvalidated as being
[
12] L. Plasseraud, H. Maid, F. Hampel, R. W. Saalfrank, Chem. Eur. J.
2
001, 7, 4007 ± 4011; R. W. Saalfrank, H. Maid, F. Hampel, K. Peters,
Eur. J. Inorg. Chem. 1999, 1859 ± 1867.
[
[
13] R. W. Saalfrank, I. Bernt, F. Hampel, Chem. Eur. J. 2001, 7, 2770 ±
2
774.
14] M. Albrecht, S. Kotila, Angew. Chem. 1996, 108, 1299 ± 1300; Angew.
Chem. Int. Ed. Engl. 1996, 35, 1208 ± 1210; D. L. Caulder, K. N.
Raymond, J. Chem. Soc. Dalton Trans. 1999, 1185 ± 1200; R. W.
Saalfrank, V. Seitz, F. W. Heinemann, C. Gˆbel, R. Herbst-Irmer, J.
Chem. Soc. Dalton Trans. 2001, 599 ± 603
[
*] Prof. Dr. H. Waldmann, Dipl.-Chem. D. Brohm
Max-Planck-Institut f¸r Molekulare Physiologie
Abteilung Chemische Biologie
[
15] Crystal data for (4 ¥ 3HOMe)
n
: C51
H
60Cs
2
Ni
2
O
21, M
r
ˆ 1392.24; crystal
3
Otto-Hahn-Stra˚e 11, 44227 Dortmund (Germany)
Fax : (‡49)231-133-2499
dimensions 0.40 Â 0.35 Â 0.35 mm ; monoclinic, space group P2(1)/n,
a ˆ 1767.67(5), b ˆ 1459.88(2), c ˆ 2271.77(3) pm, b ˆ 103.9790(10)8,
3
À3
E-mail: herbert.waldmann@mpi-dortmund.mpg.de
and
Vˆ 5688.88(13) ä ; Z ˆ 4; F(000) ˆ 2788, 1calcd ˆ 1.626 gcm . Dif-
fractometer: Nonius KappaCCD, MoKa radiation (l ˆ 0.71073 ä);
Tˆ 173(2) K; graphite monochromator; q range [8] 1.32 < q < 27.48;
section of the reciprocal lattice: À22 ꢀ h ꢀ 22, À18 ꢀ k ꢀ 14, À29 ꢀ
Universit‰t Dortmund, Fb. 3, Organische Chemie
Dipl.-Chem. D. Brohm
Semaia Pharmaceuticals GmbH & Co KG
Emil-Figge-Strasse 76 ± 80, 44227 Dortmund (Germany)
l ꢀ 29; of 19152 measured reflections, 12962 were independent and
À1
9
364 with I > 2s(I); linear absorption coefficient 1.997mm . The
structure was solved bydirect methods using SHELXS-97 and
Dr. S. Metzger
refinement with all data (685 parameters) byfull-matrix least-squares
Bayer AG, Pharma Forschung
PH-R LSC-NP, Geb. 6200, 40789 Monheim (Germany)
on F 2 using SHELXL-97;
[10]
all non-hydrogen atoms were refined
anisotropically; R1 ˆ 0.0446 for I > 2s(I) and wR2 ˆ 0.1396 (all data);
Dr. A. Bhargava
Bayer Corporation
À3
À3 [11]
largest peak (1.304 eä ) and hole (À1.607 eä ).
[
[
16] See: S. Y. Lai, T. W. Lin, Y. H. Chen, C. C. Wang, G. H. Lee, M. H.
Yang, M. K. Leung, S. M. Peng, J. Am. Chem. Soc. 1999, 121, 250 ± 251;
H. C. Chang, J. T. Li, C. C. Wang, T. W. Lin, H. C. Lee, G. H. Lee,
S. M. Peng, Eur. J. Inorg. Chem. 1999, 1243 ± 1251.
4
00 Morgan Lane, West Haven, CT 06525 (USA)
Dr. O. M¸ller
Max-Planck-Institut f¸r Molekulare Physiologie Dortmund
Abteilung Strukturelle Biologie
Otto-Hahn-Strasse 11, 44227 Dortmund (Germany)
17] D. J. Eichorst, D. A. Payne, S. R. Wilson, K. E. Howard, Inorg. Chem.
1
990, 29, 1458 ± 1459; R. Fuchs, N. Habermann, P. Kl¸fers, Angew.
Dr. F. Lieb
Chem. 1993, 105, 895 ± 897; Angew. Chem. Int. Ed. Engl. 1993, 32,
Bayer AG, Zentrale Forschung und Entwicklung
ZF-LSC-SH, Geb. Q 18, 51368 Leverkusen (Germany)
8
1
52 ± 854; S. I. Troyanov, O. Yu. Gorbenko, A. A. Bosak, Polyhedron
999, 18, 3505 ± 3509.
[
18] The microanalytical data deviate from theory due to crystal solvents
and are not reported here.
[**] This research was supported bythe Fonds der Chemischen Industrie
and the Bayer AG.
Angew. Chem. Int. Ed. 2002, 41, No. 2
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
1433-7851/02/4102-0307 $ 17.50+.50/0
307

COMMUNICATIONS
capable of binding to specific protein domains. These
compound classes can then be employed as starting points
in structural space for librarydevelopment. Libraries de-
signed and synthesized around the basic structure of such
compounds should yield modulators of protein activity with
high hit rates. Furthermore theyshould yi eld modulators of
activityfor proteins with differing activit y, function and origin
which bear the same or verysimilar domains. Within a group
of closelyrelated protein domains usuallythe individual
amino acid sequence and the detailed structure associated
with it displaysubstantial variation, which often provides the
basis for selective binding and inhibition (e.g. in the ATP-
binding sites of protein kinases). Consequently, a central and
important aspect of this concept is that it is not necessaryto
build up a given structure in everydetail. Rather, the basic
underlying structural framework has to be conserved for the
individual librarymembers, and its stereochemistryand
functional group pattern have to be varied to achieve
selectivitybetween related proteins.
(Scheme 1). The g-hydroxybutenolide moiety should be
obtainable bythe addition of 3-lithiofuran to a corresponding
aldehyde, and subsequent oxidation of the heterocycle with
singlet oxygen. The bicyclic core structure of 1 could be
generated bymeans of a Diels ± Alder route previously
O
ring-closing
metathesis linker
Diels–Alder
15
11
7
6
H
HO
HO
O
4
2
5
OH
OH
O
1
furan addition
O
O
1
2
Biologicallyactive natural products can be regarded as
chemical entities that were evolutionarilyselected and
validated for binding to particular protein domains. There-
fore, theyare alreadybiologicallyvalidated, and the under-
lying structural architectures of such natural products may
provide powerful guiding principles for librarydevelopment.
This structure-based concept differs in its fundamental
reasoning from related approaches, which focus on the
creation of chemical diversity. However, our concept does not
neglect this issue, and builds on the diversitycreated by
Nature itself.
Wittig
reaction
H
O
O
O
5
CO2Me
[2]
O
3
4
Efficient and reliable methods and multistep sequences for
the total synthesis of natural products and their analogues on
polymeric supports are paramount to the success of this
approach. The corresponding transformations must proceed
with a degree of selectivityand robustness t yp ical of related
classical transformations in solution, irrespective of the
stringencies and differing demands imposed bythe presence
of and the anchoring to the polymeric support. In a few cases,
the structural variation of natural products bymeans of solid-
phase methodologies has been achieved,[ mostlyin the late
steps of the syntheses and through the modification of a core-
structure that had been pre-synthesized in solution. But the
total synthesis of a natural product and its analogues in long
Scheme 1. Retrosynthetic analysis of dysidiolide and delineation of a solid-
phase synthesis strategy with an olefin-metathesis linker.
investigated in our laboratories[7] (2 ) 3) and successfully
employed in the total syntheses of natural dysidiolide in
solution.[
16b±e]
The knowledge gleaned from these studies
suggested that this cycloaddition leads primarily to the
6-epimer of the dysidiolide framework. However, we chose
to accept this deviation from the goal of synthesizing the
parent natural product itself, since in the concept delineated
above this is not necessarilyrequired to identifynew bio-
logicallyactive compounds. It was planned to s yn thesize the
diene for the Diels ± Alder reaction in solution from com-
merciallyavailable chiral ketoester 4. The diene would then
be attached to the linker resin 5 bymeans of a Wittig reaction
in a convergent strategy( 3 ) 4 ‡ 5).
3]
multistep sequences (i.e. ten steps and more) on polymeric
supports has onlybeen successful in one case. [
3, 4]
Herein we
describe the solid-phase synthesis of analogues of the
phosphatase inhibitor dysidiolide, which rapidly yielded
compounds with significantlyimproved biological activity.
The sesquiterpenoid dysidiolide (1) is the first naturally
occurring inhibitor of the dual-specificitycdc25 protein
phosphatase family, which play a crucial role in the regulation
of the cell cycle.[ Because of this propertyand the resulting
antitumor activity, dysidiolide has been of intense interest to
chemists, biologists, and pharmacologists.[ For the planned
solid-phase synthesis, we intended to attach the olefinic side
chain to a robust linker that would provide the terminal
alkene structure after cleavage from the solid support in a
mild and traceless manner. The new olefin metathesis linker
incorporated in 2 was thought to fulfill these requirements
Starting from dihydropyranone 6, linker 8 was synthesized
in the reaction with diallylzinc (generated in situ) in the
presence of trimethylsilyl chloride (TMSCl), subsequent
reduction of lactone 7 with lithium aluminum hydride
(LAH), and monoprotection of the resulting diol (Sche-
me 2A). The protection of one hydroxy group with dihydro-
pyran was necessary to prevent crosslinking in the subsequent
attachment of the linker to the Merrifield resin byether
formation. After deprotection and subsequent oxidation of
the hydroxy group, the active linker resin 5 was obtained in
high yield (90%, 3 steps in solid phase) with loadings of up to
5]
6]
À1
1.1 mmolg .
308
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
1433-7851/02/4102-0308 $ 17.50+.50/0
Angew. Chem. Int. Ed. 2002, 41, No. 2

COMMUNICATIONS
A)
O
O
a
b,c
H
O
O
a
O
O
O
I
6
7
5
11
H
3
d–f
O
O
HO
OTHP
O
5
8
O
b,c
B)
OTBDPS
12
CO2Me
O
O
a–c
O
O
4
9
d,e
d,e
I
OTBDPS
H
H
O
H
O
H
f
13
f,g
14
endo:exo = 91:9
endo:endo' = 95:5
d.r. = 87:4:8.9:0.1
11
10
Scheme 2. A) Synthesis of the linker and attachment to the resin.
Reagents and conditions: a) allylmagnesium bromide, ZnBr , TMSCl,
2
THF, À788C, 6 h, 52%; b) LAH, THF, RT, 1 h, 97%; c) 3,4-dihydropyran,
Dowex50WX2, toluene, RT, 3 h, 95%; d) NaH, nBu
4
NI, Merrifield-Cl
O
À1
resin (1.1 mmolg ), DMF, RT, 18 h; e) pyridinium-p-toluene sulfonate,
EtOH, dichloroethane, D, 18 h; f) o-iodoxybenzoic acid, THF/DMSO (1:1),
RT, 8 h; 90% (three steps). B) Synthesis of the diene: a) LAH, THF, 08C,
g,h
overall yield : 14%
6
3
0 min, 100%; b) TBDPSCl, dimethylaminopyridine, Et
thane, RT, 17 h, 80%; c) PCC, dichloromethane, RT, 4 h, 94%; d) vinyl-
magnesium bromide, THF, RT, 5 h; e) BF ¥ OEt , benzene/THF (4:1), D,
3% (d ± e); f) nBu NF, THF, RT, 3 h, 95%; g) iodine, imidazole, PPh
dichloromethane, RT, 1 h, 85%.
3
N, dichlorome-
H
HO
H
(
11 steps)
3
2
4
OH
O
4
4
3
,
OH
average yield:
84% per step
4
O
O
16
1
5
Scheme 3. Solid-phase synthesis of 16. Reagents and conditions: a) EtP-
Ph I, nBuLi, 11, THF, RT, 16 h, then nBuLi, 08C, 2 h; b) trimethylsilyltri-
fluoromethanesulfonate, dichloromethane, À788C, 7 h; c) p-toluenesul-
fonic acid, acetone, dichloroethane, H O, D, 20 h; d) Ph PCH OMeCl,
KOtBu, THF, RT, 4 h; e) pyridinium p-toluene sulfonate, THF, H O (1%),
D, 16 h; f) 3-bromofuran, nBuLi, THF, À788C, 5 h; g) O , EtiPr N, rose
bengal, hn, dichloromethane, À788C, 5 h, then RT, 10 min; h) (PCy ) Cl -
The diene was synthesized in solution starting from
commerciallyavailable chiral ketoester 4. After reduction
with LAH to form the diol, the primaryh yd roxygroup was
protected with tert-butyldiphenylsilyl chloride (TBDPSCl)
and the secondaryalcohol was oxidized byusing p yr idinium
chlorochromate (PCC) to obtain ketone 9 in high yield
3
2
3
2
2
2
2
3
2
2
RuˆCHPh (2  10 Mol%), dichloromethane, RT, 16 h.
(
Scheme 2B). Vinylmagnesium chloride was added to the
carbonyl group and the resulting tertiary alcohol was con-
verted into diene 10 byelimination of water under Lewis acid
catalysis. After cleavage of the protecting group, the alcohol
was finallytransformed into alk yl iodide 11, the precursor for
the subsequent Wittig reaction.
diastereoselectivity( endo/exo 91:9; ratio of endo isomers
95:5; endo/endo'/exo/exo' 87:4:8.9:0.1; the values were deter-
mined bymeans of H NMR spectroscopic analysis after
1
hydrolysis of the acetals and release of the resulting aldehydes
from the solid support byolefin metathesis). Structure and
configuration were determined bycomparison with literature
data.[
In a one-pot two-step procedure, alkyl iodide 11 was treated
with the ylide that was obtained by the deprotonation of
ethyltriphenylphosphonium iodide to form a secondary
phosphonium salt in a nucleophilic attack. This resulting
phosphonium salt was then deprotonated with a second
equivalent of base, and the aldehyde resin 5 was finallyadded
to yield diene resin 3 (Scheme 3). Subsequently, 3 was treated
with the chiral unsaturated acetal 12[ at À788C in a Lewis
acid catalyzed asymmetric Diels ± Alder reaction to generate
the bicyclic core structure of the target molecule with high
6d, 7]
After hydrolysis of the acetal, the carbon chain of aldehyde
13 was elongated bymeans of a Wittig reaction with
methoxymethyltriphenylphosphonium chloride and subse-
quent hydrolysis of the resulting ether to yield aldehyde 14.
Nucleophilic addition of 3-lithiofuran to the aldehyde resulted
in a 2:1 mixture of the epimeric alcohols 15 (determined by
means of GC-MS). The furan unit was oxidized with singlet
8]
Angew. Chem. Int. Ed. 2002, 41, No. 2
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
1433-7851/02/4102-0309 $ 17.50+.50/0
309

COMMUNICATIONS
oxygen in the presence of H¸nig×s base to form the g-
hydroxybutenolide moiety. Finally, the products were re-
leased from the polymeric carrier under very mild conditions
The ketones 18 and 19 were obtained byoxidation of
secondaryalcohols 16 and 17, respectively, with o-iodoxyben-
zoic acid, after compounds 16 and 17 were released from the
polymeric support. The olefinic analogues 20 and 21 were
synthesized by Wittig reaction of the aldehydes 13 and 14,
respectively, with the ylide obtained from furylmethyltriphe-
nylphosphonium bromide and subsequent oxidation with
singlet oxygen. Derivatives 22 and 23 were obtained by
addition of furylmethylmagnesium bromide to the carbonyl
group, followed byoxidation of the aromatic ring. Com-
pounds 17 ± 23 were obtained in overall yields of 6 ± 27%.
These results demonstrate that the multistep total synthesis
of natural products and analogues on the solid phase is
feasible. The transformations applied in the syntheses de-
scribed above include a varietyof verydifferent reaction types
that are used widely in organic synthesis: an asymmetric
cycloaddition with a removable chiral auxiliary, different
organometallic transformations, olefination reactions, differ-
ent oxidation reactions, acidic hydrolysis of acetals and enol
ethers, and a nucleophilic substitution.
in an olefin-metathesis reaction mediated byGrubbs catal ys t
9]
(
Scheme 3).[
6-epi-Dysidiolide and four other diastereomers could be
purified byflash chromatographyand separated bysemi-
preparative HPLC on SiO (98% n-hexane/2% 2-propanol).
2
1
1
1
The analytical data ( H NMR, H, H-COSY, IR, HR-MS) of
the new analogues were in full agreement with the structure,
and similar to the data reported for the natural product. The
synthesis sequence from the attachment of 8 to the solid phase
to the release of 16 into solution proceeds through a total of
eleven chemical steps and makes the desired natural product
analogue available with an overall yield of 14%, that is, with
an average yield of 84% per step. By using this procedure,
4
8 mg of the desired product can be obtained from one gram
À1
of resin with a loading of 1.1 mmolg .
Based on the developed synthesis strategy, a small library of
seven analogues was synthesized from the intermediate
aldehydes 13 and 14. Thus analogue 17 (Scheme 4) with a
shorter carbon chain was synthesized by addition of 3-lith-
iofuran to aldehyde 13 followed byoxidation of the furan ring.
To determine if the solid-state synthesis delivered bio-
logicallyactive natural product analogues with a high
frequency, we investigated compounds 16 ± 23 as inhibitors
of the protein phosphatase cdc25C and in cellular cytotoxicity
assays. From the cdc25 phosphatase family, the cdc25C
protein was chosen because 6-epi-dysidiolide 16 has previ-
ouslybeen investigated as an inhibitor of cdc25A and
cdc25B,[ therebyallowing for comparison of data.
10]
The results obtained in the phosphatase assays (Table 1)
demonstrate that all dysidiolide analogues inhibit cdc25C in
the low micromolar range with IC50 values varying by a factor
of 20. The IC50 of 5.1 mm determined for inhibition of cdc25C
H
R
R =
HO
Table 1. Results of the inhibition of cdc25C and the cytotoxicity tests performed
on the colon cancer cell lines SW480 and HCT116, the prostrate cell line PC3,
and the breast cell line MDA-MB231.
HO
26%
steps
8
O
O
Compound cdc25C[a] SW480[b] HCT116[c] PC3[c]
IC50 [mm] IC50 [mm] IC50 [mm]
MDA-MB231[c]
17
IC50 [mm] IC50 [mm]
O
1
1
1
19
20
6
7
8
5.1
16
4
1
> 33
20
4
2
> 33
> 33
1.2
1
1.6
HO
O
1
1%
21%
O
HO
0.8
1.5
6.8
2.4
6.1
9
15
11
> 20
13
> 10
> 10
1
1 steps
8 steps
on solid
support
O
on solid
support
O
O
18
19
21
22
23
[
(
a] For the phosphatase assay, the compound (5 mL) was dissolved in DMSO
100%) and added to a solution of recombinant cdc25C (0.2 mg) in assaybuffer
(85 mL; 50 mm TRIS-HCl, pH 8.0, 100 mm NaCl, 1 mm DTT, 1 mm EDTA and
0% DMSO). After incubation for 30 min at 308C, the substrate fluorescein
diphosphate (FDP) was added to give a final concentration of 1 mm. Plates were
read after a reaction time of 30 min at 485/535 nm (ex./em.). o-Vanadate (IC50
.1 mm) was used as reference compound. [b] The cells were incubated with the
HO
O
1
8%
27%
10 steps
HO
8
steps
1
O
O
O
20
21
ˆ
0
compounds at concentrations between 1.2 and 100 mm for three days. The
viabilityof the cells was measured with MTT, which is a slightly ey llow
tetrazolium salt. Living cells reduce the tetrazolium moietybya mitochondrial
dehydrogenase to a dark violet dye which is detected at 570 nm. [c] The assays
were performed by using the CytoTox 96 Cytotoxicity Assay Kit from Promega
Corporation, USA. 3000 cells were plated per well in a 96-well flat-bottomed
plate, incubated with the compounds at concentrations ranging from 0.033 mm to
HO
HO
O
OH
7
%
6%
10 steps
8
steps
HO
O
O
O
22
23
1
0 mm for three days. Cells were lysed, and the cellular lactate dehydrogenase
Scheme 4. Dysidiolide analogues obtained in solid-phase syntheses.
activitywas measured which quantitativelyreflects the number of cells.
310
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
1433-7851/02/4102-0310 $ 17.50+.50/0
Angew. Chem. Int. Ed. 2002, 41, No. 2

COMMUNICATIONS
by6- epi-dysidiolide 16 is considerablylower than the values
recorded for the inhibition of cdc25A (13 mm) and cdc25B
[5] G. Draetta, J. Eckstein, Biochim. Biophys. Acta 1997, 1332, M53 ±
M63.
[
6] a) E. J. Corey, B. E. Roberts, J. Am. Chem. Soc. 1997, 119, 12425 ±
2431; b) S. R. Magnuson, L. Sepp-Lorenzino, N. Rosen, S. J. Dani-
(
18 mm). Furthermore, the most active compound in this
1
enzyme-assay, ketone 18, exhibited an IC50 value in the high
nanomolar range (800 nm) and was 6.4 times more active than
shefsky, J. Am. Chem. Soc. 1998, 120, 1615 ± 1616; c) J. Boukouvalas,
Y.-X. Cheng, J. Robichaud, J. Org. Chem. 1998, 63, 228 ± 229; d) M.
Takahashi, K. Dodo, Y. Hashimoto, R. Shirai, Tetrahedron Lett. 2000,
16. These results indicate that dysidiolide analogues and their
4
2
1, 2111 ± 2114; e) M. Jung, N. Nishimura, Org. Lett. 2001, 3, 2113 ±
115; f) E. Piers, S. Caill e¬ , G. Chen, Org. Lett. 2000, 2, 2483 ± 2486;
derivatives can differentiate selectivelybetween different
types of phosphatases and conceivably among the three cdc25
familymembers. The data also indicate that substantial
variation of the precise structural details of the natural
product itself is tolerated and leads to inhibitors with
significantlyenhanced potenc.y Thus, replacement of the
hydroxyethyl bridge between the annelated core ring system
and the hydroxybutenolide present in compound 16 byan
unsaturated three-carbon unit (see 21) or introduction of a
keto group (see 18 and 19) leads to more potent cdc25C
inhibitors.
g) D. Demeke, C. J. Forsyth, Org. Lett. 2000, 2, 3177 ± 3179; h) H.
Miyaoka, Y. Kajiwara, Y. Yamada, Tetrahedron Lett. 2000, 41, 911 ±
914; i) J. W. Eckstein, Invest. New Drugs 2000, 18, 149 ± 156.
[7] D. Brohm, H. Waldmann, Tetrahedron Lett. 1998, 39, 3995 ± 3998.
[8] T. Sammakia, M. A. Berliner, J. Org. Chem. 1994, 59, 6890 ± 6891.
[9] P. Schwab, M. B. France, J. W. Ziller, R. H. Grubbs, Angew. Chem.
1
995, 107, 2197 ± 2181; Angew. Chem. Int. Ed. Engl. 1995, 34, 2039 ±
2041.
[
10] M. Takahashi, K. Dodo, Y. Sugimoto, Y. Aoyagi, Y. Yamada, Y.
Hashimoto, R. Shirai, Bioorg. Med. Chem. Lett. 2000, 10, 2571 ± 2574.
11] T. Mosman, J. Immunol. Methods 1983, 65, 55 ± 63.
[
The synthetic dysidiolide analogues also displayed consid-
erable and differing biological activityin a ct yo toxicity
assay[ of the colon cancer cell line SW480 (Table 1). Four
of the eight compounds investigated showed IC50 values in the
verylow micromolar range and pronounced antitumor
activity. In this cellular assay, alcohol 17 with a shortened
carbon chain was the most active compound, whereas
inhibitors 18 and 19 which had shown the lowest IC50 values
and compounds 22 and 23 in which the alcohol is positioned
differently between the hydroxybutenolide and the core
structure of dysidiolide were considerably less active. This
trend also became apparent when ketones 18 and 19 as well as
epi-dysidiolide 16 were subjected to cytotoxicity assays of the
colon cell line HCT116, the prostate cancer cell line PC3, and
the breast cancer cell line MDA-MB231. Ketones 18 and 19
again were substantiallyless active than epi-dysidiolide 16,
which inhibits cell proliferation in all three cases with IC50
values in the verylow micromolar range (Table 1).
11]
Dinitroxide Carbenes, A New Class of
Carbenes with Autoumpolung Character:
Preparation in Solution and Stabilization in
Transition Metal Complexes**
Robert Weiss* and Norbert Kraut
With the synthesis of 1 we recentlyreported the protot yp e
of a carbene with autoumpolung character.[ As the general
structure 2 shows, in such a system a singlet carbene center is
conjugated with the termini of a two-step p-redox system in
the reduced state (RED).[ The electrons in this reservoir can
be used in the sense of a redox umpolung to respond to
electronic requirements of the coordination partner at the
carbene center.[ To extend this concept we report here the
first synthesis of the dinitroxide carbene 3 in solution and its
stabilization in transition metal complexes.
1]
2]
Thus, the data indicate that the small libraryof natural
product analogues alreadycontains potent compounds with
significantlydifferent biological activities both in vitro and in
vivo. The observation that the order of IC values determined
3]
5
0
in an enzyme assay does not necessarily parallel the outcome
of cellular assays is not uncommon.
π-Redox
(RED)
In conclusion, we have demonstrated that the synthesis of
natural product derived libraries in long multistep sequences
executed on a polymeric support and employing a variety of
widelydiffering s yn thetic transformations is feasible and that
it can deliver potent biologicallyactive compounds with high
frequency.
N
N
N
N
C
1
3
2
π-Redox
SEM)
π-Redox
(SEM)
(
Received: August 31, 2001 [Z17833]
O
O
C
[
1] Yu. I. Wolf, N. V. Grishin, E. V. Koonin, J. Mol. Biol. 2000, 299, 897 ±
05.
4
9
[
[
2] S. L. Schreiber, Science 2000, 287, 1964 ± 1969.
3] For an up-to-date review, see: D. G. Hall, S. Manku, F. Wang, J. Comb.
Chem. 2001, 3, 125 ± 150.
[
*] Prof. Dr. R. Weiss, Dipl.-Chem. N. Kraut
Institut f¸r Organische Chemie
Universit‰t Erlangen-N¸rnberg
[
4] a) K. C. Nicolaou, N. Winssinger, J. Pastor, S. Ninkovic, F. Sarabia, Y.
He, D. Vourloumis, Z. Yang, T. Li, P. Giannakakou, E. Hamel, Nature
Henkestrasse 42, 91054 Erlangen (Germany)
Fax : (‡49)9131-85-25876
1
997, 387, 268 ± 272; b) K. C. Nicolaou, D. Vourloumis, T. Li, J. Pastor,
E-mail: robert.weiss@chemie.uni-erlangen.de
N. Winssinger, Y. He, S. Ninkovic, F. Sarabia, H. Vallberg, F.
Roschangar, N. P. King, M. R. V. Finlay, P. Giannakakou, P. Verdier-
pinard, E. Hamel, Angew. Chem. 1997, 109, 2181 ± 2187; Angew. Chem.
Int. Ed. Engl. 1997, 36, 2097 ± 2103.
[**] We thank Prof. P. Audebert, ENS Cachan (Paris), for undertaking the
cyclic voltammetric measurements and Dr. F. Hampel, Universit‰t
Erlangen-N¸rnberg, for performing the crystal-structure analysis.
Angew. Chem. Int. Ed. 2002, 41, No. 2
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
1433-7851/02/4102-0311 $ 17.50+.50/0
311