Highly enantioselective synthesis and cellular evaluation of spirooxindoles inspired by natural products

Article abstract of DOI:10.1038/nchem.730

In biology-oriented synthesis the underlying scaffold classes of natural products selected in evolution are used to define biologically relevant starting points in chemical structure space for the synthesis of compound collections with focused structural diversity. Here we describe a highly enantioselective synthesis of natural-product-inspired 3,3′ -pyrrolidinyl spirooxindoles-which contain an all-carbon quaternary centre and three tertiary stereocentres. This synthesis takes place by means of an asymmetric Lewis acid-catalysed 1,3-dipolar cycloaddition of an azomethine ylide to a substituted 3-methylene-2-oxindole using 1-3 mol% of a chiral catalyst formed from a N,P-ferrocenyl ligand and CuPF₆ (CH₃ CN)₄. Cellular evaluation has identified a molecule that arrests mitosis, induces multiple microtubule organizing centres and multipolar spindles, causes chromosome congression defects during mitosis and inhibits tubulin regrowth in cells. Our findings support the concept that compound collections based on natural-product-inspired scaffolds constructed with complex stereochemistry will be a rich source of compounds with diverse bioactivity.

Full text of DOI:10.1038/nchem.730

ARTICLES
PUBLISHED ONLINE: 11 JULY 2010 | DOI: 10.1038/NCHEM.730
Highly enantioselective synthesis and cellular
evaluation of spirooxindoles inspired by
natural products
1,2
1,2
4
5
¨
Andrey P. Antonchick , Claas Gerding-Reimers , Mario Catarinella , Markus Schurmann ,
Hans Preut5, Slava Ziegler1, Daniel Rauh3 and Herbert Waldmann1,2
*
In biology-oriented synthesis the underlying scaffold classes of natural products selected in evolution are used to define
biologically relevant starting points in chemical structure space for the synthesis of compound collections with focused
structural diversity. Here we describe a highly enantioselective synthesis of natural-product-inspired 3,3′-pyrrolidinyl
spirooxindoles—which contain an all-carbon quaternary centre and three tertiary stereocentres. This synthesis takes place
by means of an asymmetric Lewis acid-catalysed 1,3-dipolar cycloaddition of an azomethine ylide to a substituted
3-methylene-2-oxindole using 1–3 mol% of a chiral catalyst formed from a N,P-ferrocenyl ligand and CuPF6(CH3CN)4.
Cellular evaluation has identified a molecule that arrests mitosis, induces multiple microtubule organizing centres and
multipolar spindles, causes chromosome congression defects during mitosis and inhibits tubulin regrowth in cells. Our
findings support the concept that compound collections based on natural-product-inspired scaffolds constructed with
complex stereochemistry will be a rich source of compounds with diverse bioactivity.
elevance to nature and biological prevalidation are among the of tubulin polymerization. Notably, non-natural spirooxindoles of
most important criteria to be met by compound classes for this class, such as 2 (Fig. 1), inhibit the cell cycle and represent a
R
chemical biology and medicinal chemistry research. Bioactive novel type of potent non-peptidic inhibitor of the p53-MDM2
natural products are a proven and rich source of disease-modulating protein–protein interaction that is crucial for the regulation of the
drugs and of efficient tools and reagents for the study of biological tumour-suppressing activity of the p53 protein11–13
phenomena1–5. Their pronounced bioactivity has been rationalized
The synthesis of spirooxindoles in general remains a significant
.
by the fact that during biosynthesis and when exerting their biologi- challenge in organic chemistry14,15. 3,3′-pyrrolidinyl-spirooxindoles
cal functions they need to interact with multiple proteins as sub- have been the subject of elegant asymmetric synthesis approaches,
strates and targets6. Because the number of structural motifs of in particular multistep diastereoselective transformations using
proteins and natural products is limited6, the underlying scaffold chiral auxiliaries9,10; however, only recently was the first enantiose-
classes of natural products selected in evolution and of compound lective access to this heterocycle class reported through an organo-
classes derived from or inspired by them can be regarded as biologi- catalysed three-component 1,3-dipolar cycloaddition16. Here we
cally prevalidated starting points in chemical structure space for describe the first Lewis acid-catalysed highly enantioselective syn-
compound collection development. Based on this rationale we intro- thesis of 3,3′-pyrrolidinyl-spirooxindoles by means of a 1,3-
duced biology-oriented synthesis (BIOS) as a hypothesis-generating dipolar cycloaddition of an azomethine ylide to a 3-arylidene- or
approach for the design and synthesis of natural-product-inspired alkylideneoxindole to generate an all-carbon quaternary spirocentre
compound collections enriched by molecules with diverse bioactiv- and three tertiary centres in a single reaction step. Initial cellular
ity7,8. In BIOS, biological relevance and prevalidation are used to evaluation of a small compound collection prepared in this way
select the underlying scaffolds for the synthesis of compound identified a library member that arrests the cell cycle at the G2/M
collections with chemical diversity focused around a biologically phase, induces multiple microtubule organizing centres and multi-
selected starting point in chemical space. Given the structural polar spindles, causes chromosome congression defects during
complexity and richness in stereogenic centres of natural products, mitosis and inhibits tubulin regrowth in cells.
and consequently of the compound libraries inspired by their
In order to establish the enantioselectively catalysed 1,3-dipolar
structure, the development of efficient enantioselective synthesis cycloaddition, (E)-configured benzylidene oxindole (E)-3a and
methods is at the heart of BIOS and related approaches such as glycine ester imine 4a (Fig. 2a and Table 1) were treated with a
diversity-oriented synthesis2,8.
variety of copper catalysts or silver acetate17–20 and chiral ligands
The 3,3′-pyrrolidinyl-spirooxindole scaffold defines the charac- in different solvents in the presence of catalytic amounts of triethyl-
teristic structural core of a large family of alkaloids with pronounced amine (Supplementary Table S1 and Supplementary Fig. S1). From
and diverse bioactivity profiles9,10. For instance, spirotryprostatin B this initial series of experiments the use of 3 mol% of both N,P-fer-
1 (Fig. 1) arrests the cell cycle at the G2/M phase and is an inhibitor rocenyl ligand 7a and CuPF6(CH3CN)4 in tetrahydrofuran (THF)
1Max-Planck-Institut fu¨r Molekulare Physiologie, Abteilung Chemische Biologie, Otto-Hahn-Strasse 11, 44227 Dortmund, Germany, 2Technische Universita¨t
Dortmund, Fakulta¨t Chemie, Chemische Biologie, Otto-Hahn-Strasse 6, 44221 Dortmund, Germany, 3Chemical Genomics Centre of the Max Planck Society,
Otto-Hahn-Strasse 15, 44227 Dortmund, Germany, 4Universita¨t Konstanz, Konstanz Research School Chemical Biology, 78457 Konstanz, Germany,
5Technische Universita¨t Dortmund, Fakulta¨t Chemie, Anorganische Chemie, Otto-Hahn-Strasse 6, 44221 Dortmund, Germany.
e-mail: herbert.waldmann@mpi-dortmund.mpg.de
*
NATURE CHEMISTRY | VOL 2 | SEPTEMBER 2010 | www.nature.com/naturechemistry
© 2010 Macmillan Publishers Limited. All rights reserved.
735

NATURE CHEMISTRY DOI: 10.1038/NCHEM.730
ARTICLES
enantiomeric excess (e.e.) of up to 98%, which was gratifyingly com-
bined with a higher yield (Table 1, entries 3–5). The non-linear
effect may arise because of the formation of an additional
complex; at a 1:1 ratio, intermediate 8 is generated as well as cataly-
tically active Cuþ with no chiral ligands coordinated, but in the
presence of a small excess of ligand the equilibrium between Cuþ
ions and the chiral ligands is changed, and a complex with two
ligand molecules coordinated to one Cuþ may also be generated.
This leads to a high concentration of catalytically active Cuþ ions
without chiral ligands and hence a decrease in the enantioselectivity
of the product is observed. From the complex with two ligands,
intermediate 8 may be formed by exchange of one ligand molecule
for the amino acid ester imine and subsequent deprotonation.
Increasing the ligand/Cuþ ratio to 2:1 would ensure that all the
Cuþ ions were coordinated by two ligands, one of which would
then be exchanged for the imine to subsequently form 8.
Cl
N
N
O
O
H
O
NH
O
O
N
N
Cl
H
H
1
2
Spirotryprostatin B
Figure 1 | Structures of a representative bioactive natural product structure
with 3,3′-pyrrolidinyl-spirooxindole and a non-natural bioactive compound.
Spirotryprostatin B 1 arrests the cell cycle at the G2/M transition and is an
inhibitor of tubulin polymerization, and spirooxindole 2 is an inhibitor of the
p53-MDM2 protein–protein interaction (see Supplementary Information).
To the best of our knowledge such a non-linear effect of the
ligand/Cuþ ratio on the enantioselectivity of 1,3-dipolar cycloaddi-
emerged as most advantageous with respect to yield, enantioselec- tions has not been described before. Furthermore, identification of
tivity and diastereoselectivity (Table 1, entry 1). Chiral catalysts this optimized ligand/Cuþ ratio allowed us to reduce the catalyst
comprising a ferrocene ligand and a Cu(I) precursor have been loading to 1% without a negative impact on diastereoselectivity,
used previously in enantioselective azomethine ylide cycloadditions enantioselectivity or yield (Table 1, entries 5–7). Even at 0.1% cata-
with a,b-unsaturated esters20–22, maleimides19, nitro-olefines23,
lyst loading, product 5a is formed with an e.e. of 91% (Table 1,
vinyl sulfones18 or fullerene17 as dipolarophiles. Notably, if instead entries 7–10). The absolute configuration of the major stereoisomer
of ligand 7a, which contains an NH2 group, ligands 7b or 7c were was unambiguously determined by crystal structure analysis
used, in which the NH2 group is replaced by an NR2-group, the (Supplementary Fig. S2).
In order to rationalize the steric course of the reaction we
propose the formation of intermediate complex 8 in which (1) the
direction of the enantioselection is reversed (Supplementary
Table S1 and Supplementary Fig. S1). Further variation of the reac-
tion conditions to improve the method surprisingly reveal a non- metal atom is simultaneously coordinated by the ligand 7a and
linear dependence of the enantioselectivity on the ligand/Cuþ the glycine ester imine 4a (Fig. 2b); (2) unfavourable steric inter-
ratio. Moderate increase of the ligand/Cuþ ratio from 1:1 to 1.1:1 actions are minimized; and (3) the dipolarophile attacks from the
led to reduced diastereoselectivity and enantioselectivity (Table 1, least hindered face. Deprotonation of the chiral complex 8 leads
entries 1 and 2); a further increase up to a ratio of 2:1 restored to the azomethine ylide, which undergoes the 1,3-dipolar cyclo-
the diastereoselectivity and resulted in a significant increase in addition. The benzylidene oxindole 3a attacks the azomethine
a
NH2
CO2Me
N
PPh2
CO2Me
NH
CO2Me
NH
Fe
7a
+
+
Cu(CH3CN)4PF6,
Et3N
Br
Br
O
O
O
N
N
N
H
H
H
Br
(E)-3a
4a
5a
6a
c
Br
b
–
+
PF6
Br
HN
Ar
OMe
CO2Me
O
N
H
NH2
OMe
N
Et3N
N
N
Cu
H
P
Cu
Fe
OMe
P
O
Ph
Ph
O
O
Fe
O
O
Ph
Ph
N
H
8
9
Figure 2 | Representative enantioselective 1,3-dipolar cycloaddition between an arylidene oxindole and an amino acid ester imine in the presence of a
chiral catalyst, and proposed rational for the observed stereocontrol. a, Enantioselective 1,3-dipolar cycloaddition between dipolarophile (E)-3a and the
1,3-dipole formed from Schiff base 4a in the presence of a chiral Lewis acid catalyst formed in situ from Cu(CH3CN)4PF6 and a ferrocenyl P-N ligand 7a.
b, Proposed structure of the tetrahedral complex 8 formed by coordination of Cuþ with the ligand 7a and imine 4. c, The Michael adduct 9 most likely
formed in the reaction of (Z)-3-(3,4,5-trimethoxybenzylidene)indolin-2-one with 4a.
736
NATURE CHEMISTRY | VOL 2 | SEPTEMBER 2010 | www.nature.com/naturechemistry
© 2010 Macmillan Publishers Limited. All rights reserved.

NATURE CHEMISTRY DOI: 10.1038/NCHEM.730
ARTICLES
conditions with 90% e.e. and in moderate yield (Table 2, entry 9).
Spirooxindole 5g (Table 2, entry 6) provides a particularly remark-
able case as it contains a penta-substituted aromatic ring, and is
therefore a significant challenge for an enantioselectively catalysed
reaction. This example demonstrates that it is not necessary to use
pure (E)-configured dipolarophiles in the synthesis sequence. The
corresponding benzylidene-indolinone was used as a 2:1 mixture
of the (E)- and (Z)-isomers, which equilibrates under the reaction
conditions24,25 and from which the (E)-isomer reacts more
rapidly. The corresponding product 5g was obtained with 84%
yield and 97% e.e. However, if the benzylidene-indolinone is
equipped with electron-donating substituents, as is the case for
the 3,4,5-trimethoxyphenyl-substituted indolinone, the equili-
bration of the (Z)- and the (E)-isomer is slowed down24,25 and
both isomers can be separated and investigated independently. As
expected, the (E)-isomer reacts rapidly with p-bromobenzylidene
glycine ester under the optimized conditions to yield the cyclo-
adduct 5k with an e.e. of 97% (Table 2, entry 10). But, under the
same conditions the corresponding (Z)-isomer gave only trace
amounts of the expected cycloadducts. Instead, with silver acetate
as the catalyst, the diastereomeric pyrrolidinyl-indolinone 6k was
obtained as a racemate (Table 2, entry 11). We assume that the
complex formed between the imino ester, the Cuþ and the chiral
ligand cannot undergo a productive interaction with the (Z)-config-
ured olefin because of unfavourable steric interactions between the
ligand and the bulky trimethoxyphenyl group. In the presence of
the silver catalyst, however, product formation most likely occurs
by means of a stepwise sequence including formation of an
intermediate Michael adduct 9 (Fig. 2c) followed by a fast
Mannich reaction in analogy to the reaction of azomethine ylides
Table 1 | Evaluation of reaction conditions for the catalytic
1,3-dipolar cycloaddition.
Entry* [Cu1]
(mol%)
[7a]
(mol%)
Time
(h)
d.r.†
5a/5b
Yield
(%)‡
e.e. of
5a (%)§
1
3
3
3
3
3
2
1
0.5
0.3
0.1
3
3.3
4
5
6
4
2
1
0.6
0.2
1
1
1
1
1
1
1
3
5
12
15/1
10/1
10/1
10/1
15/1
15/1
15/1
12/1
12/1
10/1
86
85
77
84
92
92
91
75
54
33
90
72
93
98
98
98
98
95
93
91
2
3
4
5
6
7
8
9
10
*Reaction conditions: ligand 7a, Cu(CH3CN)4PF6, iminoester 4a (1.3 equivalents), Et3N (20 mol%)
and dipolarophile 3a (1 equivalent) in THF (1 ml) at ambient temperature. †See Supplementary
Information for details; further diastereomers were formed in only very minor amounts and not
characterized. ‡Isolated yields of the major diastereomer 5a after column chromatography.
§Determined by chiral HPLC analysis.
ylide from the least hindered face, that is, from behind (as indicated
by the arrow), thereby avoiding unfavourable steric interaction with
the bulky PPh2 group that points to the front. The carbonyl group of
the oxindole 3a may form a hydrogen bond with the amino group of
the ligand 7a and thereby stabilize the transition state. In ligands 7b
and 7c the amino group is replaced by a dialkylamino group
(Supplementary Fig. S1), and the hydrogen bond cannot be
formed. Instead the alkyl groups will cause unfavourable steric inter-
actions. Thereby the attack of the oxindole is directed to the front
and the direction of the enantioselection is reversed
(Supplementary Table S1). According to this model, protic additives
should compete with hydrogen-bond formation in intermediate 8
and consequently lead to reduced enantioselectivity. Therefore, we
performed cycloadditions of 4a to 3a in the presence of different
alcohols and acetic acid (Supplementary Table S2). The cyclo-
addition products were indeed formed with reduced enantioselec-
tivity and the degree of reduction correlated with the
concentration and pKa of the additives. In addition, replacement
of the glycine ester imine 4a with derivatives of b-alanine did not
lead to the formation of the corresponding cycloadducts, presum-
ably because complexes analogous to 8 cannot be formed and
deprotonated to yield dipoles under these reaction conditions.
This model is in agreement with previous studies on the use of
with nitroalkenes23,26
.
In a second series of experiments the substitution pattern in the
glycine ester imines was varied. In these cases the enantioselective
catalysis results in high levels of enantioselectivity and fast reactions
for monosubstituted benzylideneglycine esters bearing electron-
withdrawing and -donating substituents and for multiply substi-
tuted benzylideneglycine esters (Table 2, entries 12–18). Notably
the transformation proceeds smoothly in the presence of acidic
groups; for example, the phenol embedded in 5q (Table 2, entry
17). An imine derived from an aliphatic aldehyde was not reactive
even at high catalyst loading (Table 2, entry 18).
In light of the finding that 3,3-pyrrolidino-spirooxindoles induce
mitotic arrest by interfering with the p53-MDM2 interaction13 we
investigated whether the cycloadducts obtained as described above
share this bioactivity. To this end, the compound collection was
screened at a concentration of 30 mM for phenotypic changes
associated with mitotic arrest in BSC-1 cells. From the 39 com-
pounds tested only cycloadduct 6k induced phenotypic changes,
such as accumulation of round-shaped cells with condensed
DNA, which are indicative of mitotic arrest (Supplementary
Fig. S3). Compound 6k differs from the other screened compounds
by its relative configuration. In order to determine whether absolute
configuration is also important for bioactivity, both enantiomers of
6k were subjected to fluorescence activated cell sorting (FACS)
analysis. This revealed that (2)-6k but not (1)-6k caused accumu-
lation of cells in the G2/M phase in BSC-1, HCT116 p53 þ/þ and
p532/2, and HeLa cells (Fig. 3a, Supplementary Table S5). At a
10 mM concentration virtually all cells were arrested in G2/M
phase. In HeLa cells an increase of cells in the G2/M phase was
observed at concentrations as low as 2 mM.
ferrocenyl ligands in asymmetric 1,3-dipolar cycloadditions20,21
.
The regioselectivity observed in the enantioselective 1,3-dipolar
cycloaddition reaction catalysed by the copper catalysts used here
is opposite to the selectivity observed in analogous enantioselective
proton-acid-catalysed 1,3-dipolar cycloadditions to nitrogen-pro-
tected methyleneindolinones16. This difference has been rational-
ized by the involvement of a zwitterionic azomethine ylide
resonance form that is not usually present in 1,3-dipolar cycloaddi-
tions. Reversed regioselectivity is thought to result from the stabiliz-
ation stemming from the favourable p–p stacking interaction
between the oxo-indole ring and the conjugated esters, on the
basis of theoretical calculations of the transition state16.
To further explore the scope of the reaction the structures of the
dipolarophile and the 1,3-dipole were varied. Benzylidene indoli-
nones with varying substituents and a heteroaromatic analogue
yielded the desired cycloadducts in rapid reactions that proceeded
in high yield and with excellent diastereoselectivity and enantios-
electivity (Table 2, entries 1–10; diastereomeric ratio (d.r.) . 10–
15:1). Replacement of the aromatic substituent on the double
bond of the dipolarophile with an ester led to moderate decrease
of enantioselectivity (Table 2, entry 8); however, after a single recrys-
tallization the cycloadduct was obtained nearly enantiomerically
pure. Alkylideneindolinone 5j was formed under optimized
Unexpectedly, both enantiomers of 6k were inactive in an
enzyme-linked immunosorbent assay monitoring the formation of
the p53-MDM2 complex, which indicates that the observed bioac-
tivity does not correlate with inhibition of the p53-MDM2 inter-
action (Supplementary Fig. S4). This finding is supported by the
fact that arrest in the G2/M phase by (2)-6k is not dependant on
NATURE CHEMISTRY | VOL 2 | SEPTEMBER 2010 | www.nature.com/naturechemistry
737
© 2010 Macmillan Publishers Limited. All rights reserved.

NATURE CHEMISTRY DOI: 10.1038/NCHEM.730
ARTICLES
Table 2 | Lewis acid catalysed synthesis of 3,3′-pyrrolidinyl-spirooxindoles.
Entry
Product
Time (h)
Yield (%)*
e.e. (%)†
Entry
Product
Time (h)
Yield (%)*
e.e. (%)†
97
CO Me
2
MeO
MeO
1
4
80
96
10
2
85
CO Me
2
MeO
NH
NH
Br
MeO
O
Br
N
H
5b
O
N
H
5k
CO Me
2
MeO
}
F
2
3
3
2
82
89
98
98
11
4
41
–
CO Me
2
MeO
NH
O
N
H
5c
NH
MeO
Br
Br
O
N
H
6k
F
CO Me
2
12
1
95
94
CO Me
2
NH
NH
O
CF
3
O
Br
N
H
N
H
5l
5d
F
CO Me
2
CO2Me
NH
4
5
6
2
81
98
13
14
15
2
94
84
89
96
NH
O
CN
Br
O
N
H
N
H
5m
5e
CO Me
2
CO Me
2
91ꢀ
Br
1
65
84
97
0.5
NH
NH
O
Cl
Br
Cl
O
N
H
5f
N
H
5n
CO Me
2
3
97‡
4
84ꢀ
F
N
F
NH
O
F
F
CO Me
2
NH
Br
F
N
H
Br
O
5o
N
H
5g
CO Me
2
CO Me
2
7
8
9
3
82
97
50
95
16
17
18
0.5
93
84ꢀ
Cl
NH
NH
Cl
O
Br
O
O
N
H
N
H
5h
5p
CO Me
2
CO Me
2
1
85 (99§)
18
58
96ꢀ
Br
NH
NH
OH
OMe
Me OC
2
Br
O
O
N
H
N
H
5i
5q
CO Me
2
CO Me
2
90ꢀ
36
n.d.
–
#
4
NH
NH
Br
O
O
N
H
N
H
5j
5r
For reaction conditions see Supplementary Information. *Isolated yields of the major diastereomer after column chromatography. †Determined by chiral HPLC analysis. ‡After single crystallization. §A mixture of
stereoisomeric oxindoles (E/Z ¼ 2/1) was used. ꢀ3 mol% of catalyst and 6 mol% of ligand 7a were used. }The reaction was performed with the (Z)-isomer of the arylidene-oxindole using AgOAc as catalyst.
#5 mol% of catalyst and 10 mol% of ligand 7a were used. n.d. – not detected.
738
NATURE CHEMISTRY | VOL 2 | SEPTEMBER 2010 | www.nature.com/naturechemistry
© 2010 Macmillan Publishers Limited. All rights reserved.

NATURE CHEMISTRY DOI: 10.1038/NCHEM.730
ARTICLES
a
(–)-6k
b
DMSO
Nocodazole
200
150
100
50
DMSO
63.0%
300
200
100
0
28.3%
BSC-1
0
1
2
3
4
5
1
2
3
4
5
10 10 10 10 10
PI-A
10 10 10 10 10
PI-A
10µM (–)-6k
2µM (–)-6k
42.2%
400
300
200
100
0
300
200
100
0
60.0%
HeLa-L
1
2
3
4
5
1
2
3
4
5
10 10 10 10 10
PI-A
10 10 10 10 10
PI-A
c
3min
5min
10min
15min
DMSO
(–)-6k
Figure 3 | Biological evaluation of cycloadduct (2)-6k. a, FACS analysis of HeLa cells treated with (2)-6k. Cells were incubated for 20 h with 2, 10
(2)-6k or 1 M nocodazole and DMSO as controls prior to staining with propidium iodide. Numbers show the percentage of cells in the G2/M phase. PI-A,
propidium iodide fluorescence intensity. b, Influence of (2)-6k on BSC-1 and HeLa-L cells. Cells were treated for 20 h with 10 M (2)-6k or DMSO as a
control. Cells were then fixed and stained with an fluorescein isothiocyanate-coupled anti-a-tubulin antibody and DAPI/Hoechst for visualizing the DNA.
Scale bar: 20 m. c, Microtubule regrowth assay after cold treatment in BSC-1 cells. Cells were incubated for 20 h with 10 M (2)-6k or DMSO as a control.
Microtubule repolymerization was detected at the given time points after reheating, fixation and staining with an anti-a-tubulin antibody and Alexa Fluor
488-coupled secondary antibody, and 4′,6-diamidino-2-phenylindole (DAPI) for visualizing the DNA. Scale bar: 20
m.
mM
m
m
m
m
m
the presence of wild type p53 as shown for HCT116 p53 þ/þ and Fig. S7). Furthermore, a considerable number of BSC-1 cells were
p532/2 cell lines.
polynucleated, which is indicative of incomplete cell division.
Unlike the findings reported for related 3,3-spirooxindoles13, our
Further microscopic analysis in BSC-1 cells revealed that (2)-6k
influences microtubules and results in a diffuse tubulin distribution results demonstrate that (2)-6k does not act by inhibition of the
(Fig. 3b). Interestingly, this phenotype might not be caused by a direct p53-MDM2 interaction, but rather via an interference with microtu-
interference with the tubulin cytoskeleton because (2)-6k at concen- bule polymerization. Ultimately this impairs the formation of the
trations as high as 50 mM did not significantly affect the in vitro mitotic spindle leading to multipolar spindles, lagging chromosomes,
polymerization of tubulin under the experimental conditions used. mitotic arrest or incomplete mitosis. Notably, this difference in bioac-
Furthermore, (2)-6k inhibited microtubule regrowth in BSC-1 cells tivity may be linked to the very different spatial arrangement of func-
after depolymerization by cold treatment at 10 mM (Fig. 3c and tional groups for stereoisomer 6k as compared with the diastereomeric
Supplementary Fig. S6), but (1)-6k did not. In cells treated with spirooxindole-based p53-MDM2 inhibtors11 (Supplementary Fig. S8).
dimethylsulfoxide (DMSO), microtubule organizing centres were
The mode of action of compound (2)-6k also differs from the
detected just 1 min after reheating and mictrotubule polymerization biological function of the related 3,3′-pyrrolidinyl-spirotryprostatin
was completed after 15 min. (2)-6k caused a significant delay in the 2. This spiro compound arrests the cell cycle at the G2/M phase but
microtubule repolymerization and more than one microtubule organiz- unlike cycloadduct (2)-6k it directly inhibits tubulin polymeriz-
ing centre per cell appeared 2 min after reheating. Microtubules failed ation. In addition, formation of multipolar spindles and more
to repolymerize after 5 min, but after 10 min the formation of poorly than one microtubule organizing centre per cell has not been
organized microtubules was detected. Consistent with the presence described for the spirotryprostatins27,28
of more than one microtubule organizing centre per cell, treatment
.
In conclusion, we developed a highly enantioselective Lewis acid-
of BSC-1 or HeLa-L cells with (2)-6k resulted in the formation of catalysed 1,3-dipolar cycloaddition for the synthesis of biologically
multipolar spindles and chromosome congression defects in relevant natural-product-inspired 3,3′-pyrrolidinyl-spirooxindoles
mitotic cells (Fig. 3b; for higher magnification see Supplementary from simple and readily available starting materials. For the first
NATURE CHEMISTRY | VOL 2 | SEPTEMBER 2010 | www.nature.com/naturechemistry
© 2010 Macmillan Publishers Limited. All rights reserved.
739

NATURE CHEMISTRY DOI: 10.1038/NCHEM.730
ARTICLES
time we discovered a non-linear effect in the ligand/Cuþ ratio in the
asymmetric 1,3-dipolar cycloaddition. This enantioselective syn-
thesis efficiently gives access to spirocycles with an all-carbon qua-
ternary spirocentre and three tertiary centres in one step using low
catalyst loading. Our findings support the concept that compound
collections based on natural-product-inspired scaffolds constructed
with complex stereochemistry and decorated with assorted substitu-
ents will be a rich source of compounds with diverse bioactivity.
17. Filippone, S., Maroto, E. E., Martin-Domenech, A., Suarez, M. & Martin, N. An
efficient approach to chiral fullerene derivatives by catalytic enantioselective
1,3-dipolar cycloadditions. Nature Chem. 1, 578–582 (2009).
18. Lopez-Perez, A., Adrio, J. & Carretero, J. C. Bis-sulfonyl ethylene as masked
acetylene equivalent in catalytic asymmetric [3 þ 2] cycloaddition of
azomethine ylides. J. Am. Chem. Soc. 130, 10084–10085 (2008).
19. Cabrera, S., Arrayas, R. G. & Carretero, J. C. Highly enantioselective copper(I)-
fesulphos-catalyzed 1,3-dipolar cycloaddition of azomethine ylides. J. Am.
Chem. Soc. 127, 16394–16395 (2005).
20. Cabrera, S., Arrayas, R. G., Martin-Matute, B., Cossio, F. P. & Carretero, J. C.
Cu-I-fesulphos complexes: efficient chiral catalysts for asymmetric 1,3-dipolar
cycloaddition of azomethine ylides. Tetrahedron 67, 6587–6602 (2007).
21. Zeng, W., Chen, G. Y., Zhou, Y. G. & Li, Y. X. Hydrogen-bonding directed
reversal of enantioselectivity. J. Am. Chem. Soc. 129, 750–751 (2007).
22. Wang, C. J., Liang, G., Xue, Z. Y. & Gao, F. Highly enantioselective 1,3-dipolar
cycloaddition of azomethine ylides catalyzed by copper(I)/TF-BiphamPhos
complexes. J. Am. Chem. Soc. 130, 17250–17251 (2008).
23. Yan, X. X. et al. A highly enantio- and diastereoselective Cu-catalyzed
1,3-dipolar cycloaddition of azomethine ylides with nitroalkenes. Angew.
Chem. Int. Ed. 45, 1979–1983 (2006).
24. Pandit, B. et al. Structure-activity-relationship studies of conformationally
restricted analogs of combretastatin A-4 derived from SU5416. Bioorg. Med.
Chem. 14, 6492–6501 (2006).
25. Sun, L. et al. Synthesis and biological evaluations of 3-substituted indolin-2-
ones: A novel class of tyrosine kinase inhibitors that exhibit selectivity toward
particular receptor tyrosine kinases. J. Med. Chem. 41, 2588–2603 (1998).
26. Vivanco, S. et al. Origins of the loss of concertedness in pericyclic reactions:
Theoretical prediction and direct observation of stepwise mechanisms in [3 þ 2]
thermal cycloadditions. J. Am. Chem. Soc. 122, 6078–6092 (2000).
27. Cui, C. B., Kakeya, H. & Osada, H. Novel mammalian cell cycle inhibitors,
spirotryprostatins A and B, produced by Aspergillus fumigatus, which inhibit
mammalian cell cycle at G2/M phase. Tetrahedron 52, 12651–12666 (1996).
28. Sebahar, P. R., Osada, H., Usui, T. & Williams, R. M. Asymmetric,
stereocontrolled total synthesis of (þ) and (2)-spirotryprostatin B via a
diastereoselective azomethine ylide [1,3]-dipolar cycloaddition reaction.
Tetrahedron 58, 6311–6322 (2002).
Received 14 December 2009; accepted 18 May 2010;
published online 11 July 2010
References
1. Li, J. W. H. & Vederas, J. C. Drug discovery and natural products: end of an era
or an endless frontier? Science 325, 161–165 (2009).
2. Kumar, K. & Waldmann, H. Synthesis of natural product inspired compound
collections. Angew. Chem. Int. Ed. 48, 3224–3242 (2009).
3. Harvey, A. L. Natural products in drug discovery. Drug Discovery Today 13,
894–901 (2008).
4. Ganesan, A. The impact of natural products upon modern drug discovery. Curr.
Opin. Chem. Biol. 12, 306–317 (2008).
5. Newman, D. J. & Cragg, G. M. Natural products as sources of new drugs over the
last 25 years. J. Nat. Prod. 70, 461–477 (2007).
6. Koch, M. A. et al. Charting biologically relevant chemical space: A structural
classification of natural products (SCONP). Proc. Natl Acad. Sci. USA 102,
17272–17277 (2005).
7. No¨ren-Mu¨ller, A. et al. Discovery of protein phosphatase inhibitor classes by
biology-oriented synthesis. Proc. Natl Acad. Sci. USA 103, 10606–10611 (2006).
8. Kaiser, M., Wetzel, S., Kumar, K. & Waldmann, H. Biology-inspired synthesis of
compound libraries. Cell. Mol. Life Sci. 65, 1186–1201 (2008).
9. Galliford, C. V. & Scheidt, K. A. Pyrrolidinyl-spirooxindole natural products as
inspirations for the development of potential therapeutic agents. Angew. Chem.
Int. Ed. 46, 8748–8758 (2007).
10. Marti, C. & Carreira, E. M. Construction of spiro[pyrrolidine-3,3′-oxindoles] -
recent applications to the synthesis of oxindole alkaloids. Eur. J. Org. Chem.
2209–2219 (2003).
11. Ding, K. et al. Structure-based design of potent non-peptide MDM2 inhibitors.
J. Am. Chem. Soc. 127, 10130–10131 (2005).
12. Ding, K. et al. Structure-based design of spiro-oxindoles as potent, specific
small-molecule inhibitors of the MDM2-p53 interaction. J. Med. Chem. 49,
3432–3435 (2006).
Acknowledgements
The authors thank T. U. Mayer for helpful discussions and B. Vogelstein for the HCT116
p53þ/þ and p53 2/2 cell lines. This work was supported by the German Federal
Ministry for Education and Research through the German National Genome Research
Network-Plus (NGFN-Plus) (Grant No. BMBF 01GS08102).
Author contributions
A.P.A. designed and carried out chemical experiments. C.G-R. and M.C. performed
biological and biochemical experiments. M.S. and H.P. carried out the X-ray
crystallographic analysis. H.W., D.R. and S.Z. designed experiments and supervised the
project. All authors discussed the results and commented on the manuscript. H.W., D.R.,
S.Z. and A.P.A. wrote the manuscript.
13. Shangary, S. et al. Temporal activation of p53 by a specific MDM2 inhibitor is
selectively toxic to tumors and leads to complete tumor growth inhibition. Proc.
Natl Acad. Sci. USA 105, 3933–3938 (2008).
14. Ashimori, A. & Overman, L. E. Catalytic asymmetric-synthesis of quarternary
carbon centers - palladium-catalyzed formation of either enantiomer of
spirooxindoles and related spirocyclics using a single enantiomer of a chiral
diphosphine ligand. J. Org. Chem. 57, 4571–4572 (1992).
Additional information
15. Trost, B. M. & Brennan, M. K. Asymmetric syntheses of oxindole and indole
spirocyclic alkaloid natural products. Synthesis 3003–3025 (2009).
16. Chen, X. H., Wei, Q., Luo, S. W., Xiao, H. & Gong, L. Z. Organocatalytic
synthesis of spiro[pyrrolidin-3,3′-oxindoles] with high enantiopurity and
structural diversity. J. Am. Chem. Soc. 131, 13819–13825 (2009).
The authors declare no competing financial interests. Supplementary information and
chemical compound information accompany this paper at www.nature.com/
naturechemistry. Reprints and permission information is available online at http://npg.nature.
com/reprintsandpermissions/. Correspondence and requests for materials should be
addressed to H.W.
740
NATURE CHEMISTRY | VOL 2 | SEPTEMBER 2010 | www.nature.com/naturechemistry
© 2010 Macmillan Publishers Limited. All rights reserved.