Design, synthesis, and biological evaluation of 3,4-diarylmaleimides as angiogenesis inhibitors

Article abstract of DOI:10.1021/jm0580297

The new analogue 2 of combretastatin A-4 was discovered to be an inhibitor of tubulin polymerization with an IC₅₀ of 7.6 μM and reduced angiogenesis in the in vivo chick embryo model. Interestingly, in a series of 2,3-diarylmaleimides closely r

Full text of DOI:10.1021/jm0580297

J. Med. Chem. 2006, 49, 1271-1281
1271
Design, Synthesis, and Biological Evaluation of 3,4-Diarylmaleimides as Angiogenesis Inhibitors
Christian Peifer,| Thomas Stoiber, Eberhard Unger, Frank Totzke,† Christoph Scha¨chtele,† Dieter Marme´,† Ruth Brenk,‡
Gerhard Klebe,‡ Dieter Schollmeyer,§ and Gerd Dannhardt*,|
Department of Pharmacy, Johannes Gutenberg-UniVersity, Staudingerweg 5, D-55099 Mainz, Germany, Department of Single Cell and Single
Molecule Techniques, Institute of Molecular Biotechnology, Beutenbergstraâe 11, D-07745 Jena, Germany, ProQinase GmbH,
Breisacherstrasse 117, D-79106 Freiburg, Germany, Department of Pharmacy, Philipps-UniVersity, Marbacherweg 6, D-35032 Marburg,
Germany, and Department of Organic Chemistry, Johannes Gutenberg-UniVersity, Duesbergweg 10-14, D-55099 Mainz, Germany
ReceiVed May 20, 2005
The new analogue 2 of combretastatin A-4 was discovered to be an inhibitor of tubulin polymerization with
an IC50 of 7.6 µM and reduced angiogenesis in the in vivo chick embryo model. Interestingly, in a series
of 2,3-diarylmaleimides closely related to this lead, no other compound was found to be active in the tubulin
polymerization assay. However, by screening in the in vivo chick embryo assay 10 was identified as a
potent angiogenesis inhibitor indicating an alternative target. Indeed, molecular modeling studies suggest a
reasonable binding mode of 10 at the ATP-binding site of the model kinase CDK2. Motivated by these
results, analogues of 10 were screened for inhibitory activity in a panel of 12 selected protein kinases and
a high affinity of 10 to VEGF-R2 was found showing an IC50 of 2.5 nM. Structure-activity relationships
(SAR) for this compound series with the isolated enzyme and equivalent antiangiogenic activity in the
chick embryo assay are presented herein.
Introduction
Selected for preclinical development was combretastatin A-4
(CA-4) (I, Chart 1). Its prodrug combretastatin A-4 disodium
phosphate causes a virtually complete vascular shutdown in
A number of pathological events such as solid tumor growth
and metastasis but also several nonmalignant diseases such as
macular degeneration, arthritis, or psoriasis in humans have been
found to be associated with angiogenesis, the formation of new
blood vessels from preexisting vasculature.1 Concerning the
neoplastic field, an innovative approach in the treatment of
diseases consists of blocking or delaying the progression of
neovascularization to dysplastic cells to stop the delivery of
nutrition and oxygen. Since physiological angiogenesis in adults
is strictly limited, compounds with antiangiogenic activity may
be useful to treat cancer with potentially less systemic toxicity
than conventional cytotoxic therapeutics.2 Furthermore, tumor
vessels can be distinguished from normal physiological vascu-
lature by their immature development and chaotic architecture
due to a serious disorder of the angiogenic process.3 In contrast
to the antiangiogenic approach, targeting the established vessels
to cause a rapid shutdown of vasculature leading to secondary
tumor cell death is referred to as the antivascular approach.4 A
promising antivascular concept to interfere selectively with
recently built capillaries is to target the microtubule system of
endothelial cells with tubulin binders. Natural tubulin binders
are colchicine, the vinca alkaloids, rhizocin, maytensine, com-
bretastatins, epothilone, and the taxanes.5 These agents generally
exert their effects by microtubule depolymerization or stabiliza-
tion and some belong to one of the most effective classes of
drugs in clinical use for anticancer treatment. One interesting
group among the diverse tubulin polymerization inhibitors are
the combretastatins, natural products isolated from the South
African tree Combretum caffrum. Studies showed combretastatin
stilbene derivatives to be members of the colchicine-like
inhibitors with extraordinary affinity for the tubulin system.6
7
several tumors within minutes after administration. Because
of the relatively simple structure and the strong potency against
a broad spectrum of human cancer cell lines, CA-4 is a very
attractive lead compound. Thus, a large number of modifications
and analogues have been synthesized, and structure-activity
relationships were described.8,9 Since only the cis isomer
possesses strong antitubulin activity but it tends to isomerize
to the inactive trans isomer,10 a ratio was to place the cis double
bond in an additional ring system to arrest the cis conformation
resulting in a noncoplanar orientation of the two aromatic rings
(e.g., in Chart 1, 1,3-imidazole II, 4,5-imidazole III, and oxazole
IV derivatives).11 A similar approach resulted in the antimitotic
(S,S)-isomer of combretadioxolan (V, Chart 1)12 and also
diarylindoles (VI, Chart 1) fulfilling this concept successfully.13
Recently, investigations on pyridine derivatives (VII, Chart 1)
and cis-4,5-diphenylisoxazolines (VIII, Chart 1) of CA-4 have
been reported.14 A number of these structural enhancements
were found to be well tolerated, and compounds with compa-
rable activity to CA-4 were described.
In this study, we report the synthesis and biological evaluation
of the new CA-4 analogue 3-(3,4,5-trimethoxyphenyl)-4-(3-
hydroxy-4-methoxyphenyl)-1H-pyrrole-2,5-dione 2 (Figure 1)
and related compounds bearing the maleimide (1H-pyrrole-2,5-
dione) moiety as a core structure. The maleimide element occurs
as a structural part in several natural compounds, for example,
the antitumor indolocarbazole rebeccamycin15 (IX, Chart 2), a
topoisomerase-I inhibitor. Chemically related to rebeccamycin,
staurosporine (X, Chart 2) represents another indolocarbazole
with broad and potent activity against several kinases at the
ATP site (e.g., cyclin-dependent kinase 1 (CDK1), IC50 ) 4
nM).16 Derived from the unselective structural lead stauro-
sporine, 2,3-diarylmaleimides have been designed and described
as protein kinase inhibitors (e.g., protein kinase C (PKC)
inhibitor bisindolylmaleimide XI, Chart 2, PKC IC50 ) 0.11
µM)17 or glycogen synthase kinase-3 inhibitors18).
* To whom correspondence should be addressed. Tel: +49 6131 39-
25728. Fax: +49 6131 39-23062. E-mail: dannhardt@uni-mainz.de.
| Department of Pharmacy, Johannes Gutenberg-University.
Institute of Molecular Biotechnology.
† ProQinase GmbH.
‡ Philipps-University.
§ Department of Organic Chemistry, Johannes Gutenberg-University.
10.1021/jm0580297 CCC: $33.50 © 2006 American Chemical Society
Published on Web 02/02/2006

1272 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 4
Peifer et al.
Chart 1. Combretastatin A-4 (CA-4, I) and Antimitotic
1,2-Substituted Five-Membered Heterocycle Derivatives
(II-VIII) Showing the Same Aromatic Substitution Pattern
Figure 1. Structure of 3-(3,4,5-trimethoxyphenyl)-4-(3-hydroxy-4-
methoxyphenyl)-1H-pyrrole-2,5-dione (2).
Chart 2. Natural Compounds Rebeccamycin (IX, Bearing the
Maleimide Moiety), Staurosporine (X, 1,5-Dihydro-
pyrrol-2-one), and Related Bisindolylmaleimide (XI)
reaction. The arylacetamides were readily available via amid-
olysis of the commercially available aryl acetyl chlorides. Aryl
glyoxylethyl esters were synthesized by Grignard reaction (320
for preparation of compounds 4 and 9) or obtained via Friedel-
Crafts-type acylation of the desired aryl derivatives with methyl
or ethyl oxalyl chloride (1,21 3-indole derivatives for preparation
of compounds 10, 10a, 12, 12a, 13, 14, 15, 16, 16a, 18, 19,
and 20,22 ethyl-(1-methylindolyl-3)-glyoxylate19b for preparation
of compound 17,23 1H-pyrrole-3-ethylglyoxylate for preparation
of compound 11,24 and indole-2-ethylglyoxylate for preparation
of compound 21,25 Tables 1-3). According to this concept, the
aim was to prepare 3-hydroxy-4-methoxyphenyl methylgly-
oxylate as a key precursor for the synthesis of the CA-4 analogue
3-(3,4,5-trimethoxyphenyl)-4-(3-hydroxy-4-methoxyphenyl)-
maleimide (2). To address this goal, we started from guajacol
(as schown in Scheme 2). 5-Substituted guajacol derivatives
are difficult to prepare since the electrophilic substitution of
guajacol takes place predominantly at the 4-position while the
corresponding mesylate (masking the phenol into a deactivating
group) directs the entry to the 5-position.21 We found out that
it was not necessary to remove the mesylate from 1 prior to the
formation of the maleimide by reaction with 3,4,5-trimethoxy-
phenyl acetamide: when 4 equiv of KOBut were used to remove
the mesyl group in the reaction simultaneously, compound 2
(bearing the free phenolic function) was obtained in moderate
yield (44%). The structure of 2 was proven by X-ray analysis
(see experimental data, Figure 2).
To prove the concept of combining CA-4 and maleimide,
compound 2 and closely related analogues were synthesized and
evaluated for their biological activity. By screening 2,3-
maleimides for antiangiogenic activity in the in vivo chick
embryo assay, we identified some of the new compounds as
potent angiogenesis inhibitors. In the present paper, structure-
activity relationships (SAR), as well as biological properties,
of selected 2,3-diarylmaleimides are presented.
Results and Discussion
Synthesis of Asymmetrical 3,4-Diarylmaleimide Deriva-
tives. Among the number of synthetic methods for preparing
3,4-diarylmaleimides, the method of Faul et al. can be used
conveniently. According to the procedure, asymmetrically
substituted 3,4-diarylmaleimides were prepared by one-pot
ligand condensation of arylacetamide units with aryl glyoxyl-
ethyl esters using KOBut in THF (Scheme 1). 19 This approach
has been reported to be efficient concerning the toleration of
various functional groups and provides good yields. In the
present study, we used 0.1 M KOBut in BuOH and additionally
4 Å mol sieves to bind the water generated in the chemical
In solution, compound 10 was found to be chemically instable
by generating the corresponding carbazole 10a. The structures
of 10 and 10a were proven by X-ray analysis (see experimental
data, Figure 3).

3,4-Diarylmaleimides as Angiogenesis Inhibitors
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 4 1273
Scheme 1. Synthesis of the Maleimide Moiety by Using the Method of Faul et al.a
a The reaction mechanism is discussed as follows:19 By addition of KOBut into a stirred solution of arylglyoxylylester 1 and arylacetamide 2 in THF, the
amide is acylated yielding the tricarbonyl intermediate A, which undergoes an intramolecular Perkin-type condensation to hydroxy imide B. Using an excess
of KOBut (2-4 equiv) causes dehydration to maleimide C via an E2 or E1CB mechanism. Molecular sieves (4 Å) was used to bind the water generated in
the chemical reaction.
Table 1. Tubulin Inhibitory Activity and Inhibition of Vessel Growth in the in Vivo Chick Embryo Model of 3,4-Diarylmaleimides 2-11 (Series A)
Bearing Core Structure A
a Inhibition of tubulin polymerization. b Inhibition of vessel growth versus control in the chick embryo assay after incubation hour 24. Each experiment
was performed with 15 samples. c Data from ref 6. d Determination not possible due to high toxicity of CA-4; all embryos died within 2 h. e NE ) no effect
was observed at a concentration up to 10 µM.
Carbazole Derivatives. Indolocarbazole alkaloids are a very
interesting class of natural products. The most well-known
members of this family are rebeccamycin and staurosporine
owing to their wide range of pharmacological activities. The
indolocarbazole, the common aglycon moiety of these com-
pounds, is known to retain much of the activity of the parent
molecules.26 Several methods have been reported for the
synthesis of carbazoles, among them an approach involving
oxidative cyclization of diindolylmaleimides27 and 3-aryl-4-
indolylmaleimides.28 Since in these studies several bisaryl-
maleimides have been reported to readily undergo oxidative
cyclization, we supposed an analogous mechanism for 3,4-
diarylmaleimides synthesized in the present approach. However,
carbazole derivatives 10a (structure proven by X-ray analysis),
12a, 16a, and 17a (Table 3) were isolated by chromatography
from the reactions leading to the corresponding 3,4-diaryl-
maleimides. Although no further investigations in terms of
mechanistic aspects were carried out, in analogy to the reported
reaction mechanism,27,28 the formation of carbazoles is discussed
herein as oxidative cyclization of several 3-aryl-4-indolyl-
maleimides as shown in Scheme 3.
Biological Evaluation. To prove the concept of combining
CA-4 and the maleimide moiety, compound 2 was synthesized
and evaluated for antimitotic potency. It was found to be active
in the in vitro tubulin-assembly assay, and an IC50 of 7.6 µM
was determined (Table 1; in comparison, the IC50 for CA-4 was
reported to range from 0.53 to 3.0 µM11). The antiangiogenic
activity of 2 in the in vivo chick embryo model revealed 76.9%
inhibition of vessel area by application of 10 µg per pellet (Table
1). Motivated by these results, a small series of analogues closely
related to this concept was prepared under maintenance of the
3,4-diarylmaleimide moiety (series A with core structure A;
Table 1). Interestingly, none of these compounds was active in
the tubulin assembly assay. However, by screening series A
for antiangiogenic activity, we identified compound 10 with
strong potency against vessel growth in the in vivo chick embryo
assay (81.7% inhibition of vessel area after 24 h incubation by
application of 10 µg per pellet, Table 1). HPLC analysis of the

1274 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 4
Peifer et al.
Table 2. Inhibition of Vessel Growth in the in Vivo Chick Embryo Model of 3-Aryl-4-indolylmaleimides 10-19 (series B) Bearing Core Structure B
and of 20 and 21
a Inhibition of vessel growth versus control in the chick embryo assay after incubation hour 24. Each experiment was performed with 15 samples.
Table 3. Inhibition of Vessel Growth in the in Vivo Chick Embryo
Model of Carbazole Derivatives Generated by Oxidative
Electrocyclization of the Corresponding 3-Aryl-4-indolylmaleimides
Scheme 2. Synthesis of the Key Precursor 1 and Subsequent
Formation of 2a
compd R1
R2
R3
R4
R5
IVGb
(% inh) (µg)
amt
10a
12a
16a
17a
H
H
OCH3
H
H
H
H
OCH3 OCH3 OCH3 10.7
H
OCH3 OCH3 OCH3 22.9
10
H
H
10
CH3 OCH3 OCH3 OCH3
content uniformity in the production process of agarose pellets
(ideal value containing 10 µg of 10) revealed that the calculated
amount per pellet is approximately recovered and that the bulk
is homogeneous but that approximately 10% of compound 10a
can be found (Figure 4). Since 10 was applied with analytical
purity, the carbazole 10a was supposed to be generated during
the production process of the pellets by oxidative cyclization
(as described above, Scheme 3). Investigation of the content of
10 and 10a in pellet samples by HPLC during the antiangiogenic
experiment (Figure 5) indicates only a release of small amounts
in 24 h of 10 (0.26 µg) and 10a (0.02 µg) resulting in the
remarkable in vivo effect. When solely 10a (chemically stable)
was investigated in the chick embryo model, 10.7% inhibition
of vessel growth by application of 10 µg could be determined
in 24 h (Table 3). These findings lead to the conclusion that
the biological effects were mainly due to the activity of
a (i) MsCl, TEA; (ii) methylglyoxalate acid chloride, CH2Cl2; (iii) 4 equiv
of KOBut following the general procedure.
compound 10. But without the expected activity against the
microtubule system, the significant inhibition of vessel growth
suggests an alternative biological target of 10 than that for the
lead compound 2. These results underline the importance of
the modified in vivo chick embryo assay used in our investiga-
tions. The model can be considered suitable for functional
screening of antiangiogenic compounds, as well as to deliver
data of biological relevance. Since compound 10 is related to
staurosporine and 2,3-diarylmaleimides, which are active against
PKC (and other kinases),17 we hypothesized that 10 was also
active against kinases involved in the angiogenic process
resulting in the in vivo activity.

3,4-Diarylmaleimides as Angiogenesis Inhibitors
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 4 1275
Figure 4. Content of compounds 10 and 10a in agarose pellets in
production process. Each point represents the mean ( SD (n ) 9).
Figure 2. ORTEP molecular structure of compound 2 with the
numbering scheme.
by ligands.29 CDK2 in particular was supposed to serve as a
valid example because by X-ray crystallography molecular
details of the ATP pocket had been well characterized.
Furthermore, knowledge of the structure of CDK2 has been a
key element in driving the design and development of a large
number of ATP competitive inhibitors for which several CDK2-
30
inhibitor structures have been determined.
The crystal
31
structure of staurosporine in complex with CDK2 provides
insight into the interactions responsible for high-affinity binding
to a variety of kinases. Many of these interactions are also
observed in other CDK2-inhibitor complexes. Namely, stauro-
sporine forms a hydrogen bond between the lactam nitrogen
and the carbonyl oxygen of Glu81. Further the carbonyl oxygen
of the lactam moiety is a hydrogen bond acceptor for the
backbone amine of Leu83. The extended indolocarbazole ring
system is accommodated in the hydrophobic cleft formed by
the amino acids Ile10, Val18, Val64, Phe80, Leu134, and
Ala144. The glycosyl moiety forms a salt bridge to the
carboxylate group of Asp86 via the protonated N-methyl amino
group and a charge-assisted hydrogen bond to the carbonyl
oxygen of Gln131 in the ribose-binding site (Figures 6 and 7).
Docking 10 and 10a into the binding site with the program
FlexX 32 suggests a reasonable binding mode in the ATP-binding
pocket of CDK2 (Figures 6 and 7). This binding mode is very
similar to the staurosporine binding mode. Likewise, hydrogen
bonds to the backbone carbonyl of Glu81 and to the backbone
amine of Leu83 are formed. The indole ring and the trimethoxy-
phenyl moiety are located in the hydrophobic cleft. Although
sharing the most important interactions of staurosporine, 10 or
10a does not address the sugar pocket. This is also the case for
numerous other ligands. Even then, they achieve affinities in
the micromolar and submicromolar range. 33 Therefore without
addressing Asp86 and Gln131 in the sugar pocket, 10 and 10a
are likely to inhibit this kinase. This prompted us to synthesize
a series of analogues of 10 (series B, 3-aryl-4-indolyl-deriva-
tives, core structure B, Table 2) and 10a (Table 3) and
subsequently to screen these compounds for inhibitory activity
in a panel of selected protein kinases.
Figure 3. ORTEP molecular structures of compounds 10 and 10a
(containing 1 equiv of DMSO H-bonded to the imide) with the
numbering scheme.
Molecular Modeling. To confirm our hypothesis, molecular
modeling studies were performed at the ATP binding pocket
of CDK2 as model kinase. Even if all members of the kinase
family bind the same cofactor, there are slight differences in
this pocket and the adjacent subpockets, which can be explored
Kinase Assays. To profile the compounds, 12 kinases were
selected for an initial screening. Members of the CDK family
(CDK2/cyclinE, CDK1/cyclinB, CDK4/cyclinD1, CDK6/
cyclinD1) were chosen because of the molecular modeling data.
Scheme 3. 1,6-π-Electrocyclization of 10 and Subsequent Oxidation of the Intermediate to the Carbazole Derivative 10a

1276 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 4
Peifer et al.
kinase panel are summarized in Table 4 (for the profile at 100
µM, see Supporting Information). The testing showed a dif-
ferential inhibitory profile. Compounds 6-9, 15, 19, and 21
have no significant inhibitory potency against any of the 12
protein kinases at 1 × 10-4 M. None of the compounds inhibited
all kinases to nearly full extent at 1 × 10-4 M. Compound 18
inhibits CDK4 (87%) and CDK6 (89%), while 20 inhibits
especially CDK4 (95%), CDK6 (92%), and PKC-R (93%).
Compounds 10 and 10a, used in the molecular modeling studies
at CDK2, were indeed found at 10 µM to inhibit this enzyme
to 42% and 41%, respectively. These results support the modeled
binding mode. Remarkable inhibitory activity for compounds
10 (100%), 14 (100%), 17 (95%), and 17a (98%) especially at
the VEGF-R2 kinase (with selectivity over the other kinases)
can be detected. The inhibition of VEGF-R2 of 10 and 14
matches the in vivo inhibition of angiogenesis in the chick
embryo model mostly (Table 2). However, VEGF-R2 was
reported to be involved in the early growth of blood vessels in
the avian embryo,34 triggering a progressive change in the
branching pattern in the subsequent development stages resulting
Figure 5. Content of compounds 10 and 10a in agarose pellets time
dependently across 48 h in the chick embryo model. Each point
represents the mean ( SD (n ) 9). In the first 24 h of the experiment,
0.26 µg of 10 (0.69 nmol) and 0.02 µg of 10a (0.05 nmol) are released
and target the vasculature.
35
in a stable vascular architecture. VEGFs are key mediators
of angiogenesis and interact with cell surface receptors, in
humans KDR (kinase domain-containing receptor or VEGFR-
2) and Flt-1 (fms-like tyrosine kinase or VEGFR-1), expressed
36
almost exclusively on vascular endothelial cells.
A third
homologous receptor, VEGF-R3 (Flt-4), is constitutively ex-
pressed on adult lymphatic endothelial cells and supports
37
lymphangiogenesis. By binding to VEGFR-2/KDR, VEGFs
mainly exert their mitogenic, chemotactic, and vascular per-
38
meabilizing effects on vasculature. Concerning the role of
VEGF-R2/KDR as an important factor in the process of
angiogenesis,39 it is possible that the inhibition of vessel growth
in the in vivo model is caused by these compounds.
Figure 6. Crystallographically determined binding mode of stauro-
sporine (blue) at the ATP-binding site of CDK2 and modeled binding
mode of compound 10 (yellow). Key residues and hydrogen bonds are
labeled.
To evaluate the full potency against KDR tyrosine kinase
and to study SAR, we decided to determine the IC50 values of
seven selected compounds structurally related to 10 in a KDR
in vitro kinase assay (Table 5).
Compound 10 was found to possess remarkable activity
against this kinase showing an IC50 of 2.5 nM. Based on the
data, SAR can be deduced by comparing 10 to the analogues.
Compound 14 bearing only two methoxy groups at the phenyl
system is 102-fold less potent than 10 indicating that all three
methoxy groups contribute to the full activity. Since the pyrrole
system (11) instead of the indole and both 5′-methoxy (16) and
1′-methyl substitution (17) of the indole system decrease activity,
the unsubstituted indole moiety can be considered optimal.
Particulary with regard to the indole nitrogen, investigations of
Davis et al. on PKC activity of a series of similar 3-phenyl-4-
indolylmaleimides derived from the structural lead staurosporine
can be discussed: SAR in their work revealed that methylation
at the indole nitrogen increased activity.17 Furthermore, they
found that any substituent different from hydrogen at position
4 of the phenyl system decreases activity. However, with regard
to the different architecture of the ATP-binding pocket of KDR
and PKC, slight structural differences of the compounds can
be essential for activity and selectivity. To understand the
molecular interactions, determination of the X-ray structure of
10 in complex with KDR and evaluation of the binding mode
would be of interest.
Figure 7. Crystallographically determined binding mode of stauro-
sporine (blue) at the ATP-binding site of CDK2 and modeled binding
mode of compound 10a (yellow). Key residues and hydrogen bonds
are labeled.
Since staurosporine-related diarylmaleimides have been reported
to show inhibitory activity to PKC,17 several isozymes of the
PKC family (PKC-R, -γ, -ꢀ, and -ι) were also included.
Furthermore, members of growth factor receptor tyrosine kinases
(EGF-R, ERBB2, TIE2, and VEGF-R2) were added to the panel
because in particular compound 10 was a strong inhibitor of in
vivo angiogenesis in the chick embryo model. The results of
screening all compounds at 10 µM for inhibitory activity in the
Focusing on the indole in 10 adherent at the 3′-position, the
analogue 21 bearing the indole moiety at the 2′-position can be
compared. Interestingly, 21 showed no inhibitory activity at 10
µM in the kinase assay (Table 4) but was determined to be
significantly antiangiogenic in the chick embryo assay (71.1%

3,4-Diarylmaleimides as Angiogenesis Inhibitors
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 4 1277
Table 4. Results of All Compounds (at 10 µM) in All Kinase Assaysa
CDK2/
cycE
CDK1/
cycB
CDK4/
cycD1
CDK6/
cycD1
compd
PKC-R
PKC-γ
PKC-ꢀ
PKC-ι
EGFR
ERBB2
TIE2
VEGF-R2
2
4
5
6
7
8
9
10
10a
11
12
12a
13
14
15
16
16a
17
17a
18
19
20
21
80
83
76
87
87
79
90
58
59
88
75
80
101
93
103
88
94
53
42
31
90
43
96
103
100
102
99
101
113
100
108
112
88
126
54
41
80
65
75
76
105
115
121
106
37
37
13
126
5
114
70
79
69
89
81
53
96
34
37
80
32
73
47
77
92
96
92
26
25
11
95
8
61
90
79
85
86
82
94
29
54
62
63
90
95
65
94
79
104
28
83
39
97
7
76
88
83
93
89
89
98
55
75
99
73
106
83
80
105
99
114
65
94
65
73
108
94
105
127
98
98
33
85
62
94
107
83
77
98
82
111
41
101
39
100
21
103
97
99
94
102
97
95
81
78
92
95
98
77
79
84
85
85
93
80
78
81
77
83
99
109
108
103
122
106
106
83
98
105
80
86
99
99
78
92
87
85
100
88
94
102
91
64
101
93
81
101
97
84
89
60
92
86
101
108
103
100
101
99
109
14
81
116
51
136
93
65
108
79
61
51
32
45
119
71
103
19
45
55
86
94
81
100
0
16
17
21
45
17
0
87
100
108
86
80
99
92
87
101
96
102
105
96
77
43
48
96
90
107
105
97
93
67
95
72
102
87
83
4
31
5
2
16
97
26
89
97
45
101
30
97
77
88
95
81
83
a The data are presented as residual activity- in percent related to the 100% controls. Spots with residual activity <10 % are indicated with bold typeface.
Table 5. VEGF-R2 Kinase Activity of Selected Maleimides
the in vivo chick embryo model. Furthermore, molecular
modeling studies suggest a plausible binding mode of 10 at the
ATP-binding site of the model kinase CDK2. Motivated by these
results, a small library of 2,3-diarylmaleimides was tested in a
panel of 12 selected protein kinases. This profile revealed highly
potent VEGF-R2 tyrosine kinase inhibitors without the expected
antitubulin activity. In particular, compound 10 possesses
remarkable potency against VEGF-R2/KDR showing an IC50
of 2.5 nM. SAR for the isolated enzyme and equivalent anti-
angiogenic activity are presented. Considering the high in vitro
potency and also being bioavailable in the in vivo chick embryo
assay with marked activity, certain 2,3-diarylmaleimides may
have potential for clinical development as antiangiogenic drugs.
VEGF-R2
IC50 [µM]
compd
10
10a
11
14
16
2.5 × 10 -9
6.2 × 10 -8
2.5 × 10 -6
4.3 × 10 -7
9.3 × 10 -7
6.0 × 10 -7
6.4 × 10 -7
17
17a
inhibition of vessel growth, Table 2). These results point to
another biological target, possibly the inhibition of another
kinase involved in the angiogenic process not investigated in
the profile. Advanced studies of selectivity with a broad
spectrum of kinases are under work. Furthermore, since related
bisindolylmaleimides have been reported to be promiscuous
inhibitors,40 investigations on diverse non-kinase enzymes will
answer the question of specificity in particular of compound
10.
The carbazole derivative 10a is about 10-fold less potent than
the analogue 3,4-diarylmaleimide 10. This fact can possibly be
explained by the rigid formation caused by the carbazole. Due
to the flexibility of 10, the slightly rotated orientation of the
aromatic systems maybe better fit to the binding pocket as
shown in the molecular modeling studies (Figures 6 and 7).
Interestingly, although 10a was found to be active at KDR with
an IC50 of 62 nM, no significant antiangiogenic effect can be
detected in the chick embryo assay (Table 3). This was
correlated to the slight solubility in the in vivo system (as
discussed above). In contrast, compound 10 was shown to be
bioavailable in the chick embryo model and subsequently reach
the molecular target in the developing vasculature. Furthermore,
these findings punctuate the physiological relevance of the chick
embryo assay as a useful angiogenesis model.
Experimental Section
Chemistry. Infrared spectra were recorded on a Perkin-Elmer
1310 infrared spectrophotometer. 1H (300 MHz, digital resolution
0.3768 Hz) and 13C (75 MHz, digital resolution 1.1299 Hz) NMR
were recorded on a Bruker AC 300; the data are reported as
follows: chemical shift in ppm from Me4Si as external standard,
multiplicity, and coupling constant (Hz). Electron impact (EI) mass
spectra were recorded on a Varian MAT 311A (70 eV) and field-
desorption (FD) mass spectra on a MAT-95 (Finnigan). For clarity,
only the highest measured signal is given for FD mass spectra.
Elemental analyses were performed on an elemental analyzer Carlo
Erba Strumentazione model 1106. Combustion analyses agreed with
the calculated data within (0.4 unless otherwise stated. Melting
points/decomposition temperatures were determined on a Bu¨chi
apparatus according to Dr. Tottoli and are uncorrected. X-ray
structure determination was performed on a CAD4-Enraf-Nonius
using Cu KR radiation with graphite monochromator. Where
appropriate, column chromatography was performed for crude
precursors with Merck silica gel 60 (0.063-0.200 mm) or Acros
organics silica gel (0.060-0.200 mm; pore diameter ca. 60 nm).
Column chromatography for test compounds was performed using
a MPLC system B-680 (Bu¨chi) with Merck silica gel 60 (0.015-
0.040 mm). The progress of the reactions was monitored by thin-
layer chromatography (TLC) performed with Merck silica gel 60
F-245 plates. Where necessary, reactions were carried out in a
nitrogen atmosphere using 4 Å molecular sieves. All reagents and
solvents were obtained from commercial sources and used as
received (THF was used after distillation over potassium/benzo-
phenone). Reagents were purchased from Sigma-Aldrich Chemie,
Conclusions
We have synthesized a new analogue of combretastatin A-4
bearing 2,3-diarylmaleimide as a core structure (compound 2).
This compound exhibits an IC50 of 7.6 µM in the in vitro tubulin
polymerization assay and significantly reduced angiogenesis in

1278 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 4
Peifer et al.
Steinheim, Germany, Lancaster Synthesis, Mu¨hlheim, Germany,
or Acros, Nidderau, Germany.
3-(4-Methoxyphenyl)-4-phenyl-1H-pyrrole-2,5-dione (7). The
general procedure was followed using phenylethylglyoxylate (2.25
g, 12.63 mmol), 4-methoxyphenylacetamide (1.65 g, 10.0 mmol),
and 1.0 M KOBut (30 mL, 30.0 mmol). Purification was achieved
by column chromatography (ethyl acetate/hexanes 3/7) to yield 7
(1.58 g, 57%) as bright yellow crystals. Mp ) 215 °C. EI-MS (m/
z): 279.5 (M+). Anal. (C17H13NO3) C, H, N.
3,4-Diphenyl-1H-pyrrole-2,5-dione (8). The general procedure
was followed using phenylethylglyoxylate (3.60 g, 20.20 mmol),
phenylacetamide (1.35 g, 10.0 mmol), and 1.0 M KOBut (30 mL,
30.0 mmol). Purification was achieved by column chromatography
(ethyl acetate/hexanes 8.5/1.5) to yield 8 (2.13 g, 86%) as bright
yellow crystals. Mp ) 213 °C. EI-MS (m/z): 249.5 (M+). Anal.
(C16H11NO2) C, H, N.
3,4-Di-(4-methoxyphenyl)-1H-pyrrole-2,5-dione (9). The gen-
eral procedure was followed using 3 (4.0 g, 19.20 mmol),
4-methoxyphenylacetamide (1.65 g, 10.0 mmol), and 1.0 M KOBut
(25 mL, 25.0 mmol). Purification was achieved by column
chromatography (ethyl acetate/hexanes 7/3) to yield 9 (1.75 g, 57%)
as yellow crystals. Mp ) 240 °C. EI-MS (m/z): 309.2 (M+). Anal.
(C18H15NO4) C, H, N.
3-(Indole-3-yl)-4-(3,4,5-trimethoxyphenyl)-1H-pyrrole-2,5-
dione (10). The general procedure was followed using indole-3-
ethylglyoxylate (1.6 g, 7.38 mmol), 3,4,5-trimethoxyphenylacet-
amide (1.2 g, 5.33 mmol), and 1.0 M KOBut (20 mL, 20.0 mmol).
Purification was achieved by column chromatography (ethyl acetate/
hexanes 1/1) to yield 10 (1.08 g, 54%) as yellow crystals. Mp )
227 °C. EI-MS (m/z): 378.8 (M+). Anal. (C21H18N2O5) C, H, N.
5,6,7-Trimethoxy-1,2,3,8-tetrahydro-benzo[a]pyrrolo[3,4-c]-
carbazol-1,3-dione (10a). The product was isolated by column
chromatography (ethyl acetate/hexanes 4/6) from reaction mixture
10 as bright yellow crystals (151 mg, 8%). Mp ) 272 °C. EI-MS
(m/z): 376.7 (M+ ). Anal. (C21H16N2O5) C, H, N.
3-Methylsulfonyl-4-methoxyphenylmethylglyoxalate (1)Friedel-
Crafts-type Acylation. To a stirred mixture of 7.0 g (52 mmol) of
AlCl3 in CH2Cl2 at 0 °C, 7.0 g (57 mmol) of methylglyoxalate
acid chloride was added dropwise. After the mixture had been stirred
for 30 min, 10.0 g (50 mmol) of 1-methoxy-2-methylsulfonyloxy-
benzol (mesylguajacol) was added slowly. The mixture was stirred
overnight at room temperature. After being quenched with ice, the
organic layer was washed with water and brine. The organic phase
was dried over Na2SO4 and concentrated to give the crude product
as a brown oil. The oil was purified by silica gel column
chromatography using ethyl acetate/hexanes 1/1, yielding the
product as a light yellow oil, which was dried to give 4.43 g (31%)
of yellowish crystals of 1 (slow crystallization). Mp ) 70 °C.
General Procedure for the Preparation of Diarylmaleimides
(Modified from ref 19). A stirred solution of the arylacetamide (1
equiv) and arylmethyl- or arylethylglyoxylate (2-4 equiv) in dry
THF containing 15 g of molecular sieves (4 Å) was cooled to 0 °C
t
under nitrogen. At this temperature, 1.0 M KOBut in BuOH (3.0
equiv) was added via septum, and the mixture was allowed to warm
to room temperature. After being stirred overnight, the reaction was
again cooled to 0 °C and quenched with saturated NH4Cl solution.
The residue was filtered and extracted with ethyl acetate, and the
organic phase was dried over Na2SO4, then concentrated and
purified by column chromatography.
3-(3,4,5-Trimethoxyphenyl)-4-(3-hydroxy-4-methoxyphenyl)-
1H-pyrrole-2,5-dione (2). The general procedure was followed
using 1 (1.50 g, 5.2 mmol), 3,4,5-trimethoxyphenylacetamide (1.13
g, 5.02 mmol), and 1.0 M KOBut (20 mL, 20.0 mmol). Purification
was achieved by column chromatography (ethyl acetate/hexanes
1/1) to yield 2 (0.85 g, 44%) as a yellow solid. Mp ) 234 °C.
EI-MS (m/z): 385.1 (M+). Anal. (C20H19NO7) C, H, N.
Structure of 2 was proven by X-ray analysis showing the
compound to have the desired 3′-hydroxy and 4′-methoxy substitu-
tion pattern, as does the natural template CA-4 (Figure 2). X-ray
analysis: CAD4 Enraf Nonius, Cu KR, SIR-92, SHELXL-97.
Further details of the crystal structure analysis are available in ref
41.
4-Methoxyphenylethylglyoxylate (3). For Grignard reagent, 0.49
g (20 mmol) of magnesium was stirred in diethyl ether, and 3.74
g (20 mmol) of 4-bromanisol was added. Diethyloxalate (7.5 g, 50
mmol) in diethyl ether was cooled to -78 °C and stirred
mechanically while the Grignard-reagent was added slowly (color
changes to yellow-brown). The reaction was allowed to warm to
room temperature, NH4Cl solution was added, and the organic layer
was separated and dried over Na2SO4. The organic phase was
concentrated under vacuum to afford a red oil, which was purified
by column chromatography (ethyl acetate/hexanes 3/7) to yield 3
(2.5 g, 60%) as a yellow oil. EI-MS (m/z): 208.7 (M+).
3-(4-Methoxyphenyl)-4-(3,4,5-trimethoxyphenyl)-1H-pyrrole-
2,5-dione (4). The general procedure was followed using 3 (0.80
g, 3.84 mmol), 3,4,5-trimethoxyphenylacetamide (0.68 g, 3.02
mmol), and 1.0 M KOBut (8 mL, 8.0 mmol). Purification was
achieved by column chromatography (ethyl acetate/hexanes 1/1)
to yield 4 (0.79 g, 71%) as yellow crystals. Mp ) 201 °C. FD-MS
(m/z): 370.3 (M+ + 1). Anal. (C20H19NO6) C, H, N.
3-(3,4,5-Trimethoxyphenyl)-4-phenyl-1H-pyrrole-2,5-dione (5).
The general procedure was followed using phenylethylglyoxylate
(1.5 g, 8.42 mmol), 3,4,5-trimethoxyphenylacetamide (1.5 g, 6.66
mmol), and 1.0 M KOBut (20 mL, 20.0 mmol). Purification was
achieved by column chromatography (ethyl acetate/hexanes 3/7)
to yield 5 (0.72 g, 32%) as yellow crystals. Mp ) 212 °C. EI-MS
(m/z): 338.4 (M+ - 1). Anal. (C19H17NO5) C, H, N.
X-ray Crystal Structure Determination of 10 and 10a.
Structures of 1042 and 10a43 have been proven by X-ray analysis
(Figure 3). X-ray analysis: CAD4 Enraf Nonius, Cu KR, SIR-92,
SHELXL-97. Further details of the crystal structure analysis are
available in the references.
3-(1H-Pyrrole-3-yl)-4-(3,4,5-trimethoxyphenyl)-1H-pyrrole-
2,5-dione (11). The general procedure was followed using 1H-
pyrrole-3-ethylglyoxylate (0.5 g, 2.99 mmol), 3,4,5-trimethoxy-
phenylacetamide (0.58 g, 2.58 mmol), and 1.0 M KOBut (10 mL,
10.0 mmol). Purification was achieved by column chromatography
(ethyl acetate/hexanes 1/1) to yield 11 (1.08 g, 54%) as red crystals.
Mp ) 223 °C. EI-MS (m/z): 328.1 (M+). Anal. (C17H16N2O5) calcd
C 62.21, H 4.88, N 8.53; found C 61.82, H 4.32, N 8.10.
3-(Indole-3-yl)-4-phenyl-1H-pyrrole-2,5-dione (12). The gen-
eral procedure was followed using indole-3-ethylglyoxylate (1.9
g, 10.26 mmol), phenylacetamide (3.6 g, 20.2 mmol), and 1.0 M
KOBut (10 mL, 10.0 mmol). Purification was achieved by column
chromatography (ethyl acetate/hexanes 2/8) to yield 12 (0.73 g,
25%) as yellow crystals. Mp ) 240 °C. FD-MS (m/z): 289.8 (M+
+ 1). Anal. (C18H12N2O2) calcd C 74.99, H 4.20, N 9.72; found C
74.38, H 4.47, N 9.71.
1,2,3,8-Tetrahydro-benzo[a]pyrrolo[3,4-c]carbazole-1,3-
dione (12a). The product was isolated by column chromatography
(ethyl acetate/hexanes 2/8) from reaction mixture 12 as bright
yellow crystals (86 mg, 3%). Mp > 280 °C. FD-MS (m/z): 387.6
(M+ + 1). Anal. (C18H10N2O2) calcd C 75.32, H 3.52, N 9.79; found
C 75.84, H 3.63, N 9.86.
3-(Indole-3-yl)-4-(4-methoxyphenyl)-1H-pyrrole-2,5-dione (13).
The general procedure was followed using indole-3-ethylglyoxylate
(1.17 g, 5.37 mmol), 4-methoxyphenylacetamide (0.85 g, 5.11
mmol), and 1.0 M KOBut (10 mL, 10.0 mmol). Purification was
achieved by column chromatography (ethyl acetate/hexanes 4/6)
to yield 13 (0.59 g, 36%) as yellow crystals. Mp ) 220 °C. EI-MS
(m/z): 318.3 (M+). Anal. (C18H12N2O2) C, H, N.
3-(3,4-Dimethoxyphenyl)-4-phenyl-1H-pyrrole-2,5-dione (6).
The general procedure was followed using phenylethylglyoxylate
(1.0 g, 5.61 mmol), 3,4-dimethoxyphenylacetamide (0.5 g, 2.56
mmol), and 1.0 M KOBut (10 mL, 10.0 mmol). Purification was
achieved by column chromatography (ethyl acetate/hexanes 3/7)
to yield 6 (0.41 g, 52%) as yellow crystals. Mp ) 202 °C. EI-MS
(m/z): 309.2 (M+). Anal. (C18H15NO4) C, H, N.
3-(Indole-3-yl)-4-(3,4-dimethoxyphenyl)-1H-pyrrole-2,5-
dione (14). The general procedure was followed using indole-3-
ethylglyoxylate (0.86 g, 3.94 mmol), 3,4-dimethoxyphenylacetamide
(0.70 g, 3.59 mmol), and 1.0 M KOBut (10 mL, 10.0 mmol).

3,4-Diarylmaleimides as Angiogenesis Inhibitors
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 4 1279
44
Purification was achieved by column chromatography (ethyl acetate/
hexanes 1/1) to yield 14 (0.51 g, 41%) as yellow crystals. Mp )
210 °C. EI-MS (m/z): 348.0 (M+). Anal. (C20H16N2O4) C, H, N.
3-(Indole-3-yl)-4-(4-chlorophenyl)-1H-pyrrole-2,5-dione (15).
The general procedure was followed using indole-3-ethylglyoxylate
(1.07 g, 4.92 mmol), 4-chlorophenylacetamide (1.90 g, 5.31 mmol),
and 1.0 M KOBut (10 mL, 10.0 mmol). Purification was achieved
by column chromatography (ethyl acetate/hexanes 2/8) to yield 15
(0.76 g, 45%) as yellow crystals. Mp ) 275 °C. EI-MS (m/z): 323.5
(M+). Anal. (C18H11ClN2O2) C, H, N.
the structures SYBYL 6.8 was used. The Connolly surface was
calculated using the MOLCAD module in Sybyl and colored
according to the lipophilicity (from very hydrophobic to hydrophilic
corresponds to brown over green to blue). Compound 10 was
45
docked into the active site of CDK2 using the FlexX docking
32
46
program (1.102).
The FlexX scoring function
was applied
during the placement and construction phase of the ligands and
DrugScore for the final ranking. 47 The 3D coordinates of the CDK2
catalytic core in complex with staurosporine were taken from the
48
Brookhaven Protein Databank (PDB code: 1AQ1).
3-(5-Methoxyindole-3-yl)-4-(3,4,5-trimethoxyphenyl)-1H-pyr-
role-2,5-dione (16). The general procedure was followed using
5-methoxyindole-3-ethylglyoxylate (1.5 g, 6.07 mmol), 3,4,5-
trimethoxyphenylacetamide (1.2 g, 5.33 mmol), and 1.0 M KOBut
(10 mL, 10.0 mmol). Purification was achieved by column
chromatography (ethyl acetate/hexanes 4/6) to yield 16 (1.72 g,
Chick Embryo Assay as in Vivo Angiogenesis Model. The
assay was basically performed according to the procedure in ref
49. Test compounds were dissolved in DMSO and used as stock
solution. For the preparation of agarose pellets, the stock solution
was mixed with 2% agarose solution at 80 °C, and 10 µL pellets
were produced (final content 5-100 µg).
70%) as red crystals. Mp ) 268 °C. FD-MS (m/z): 410.1 (M+
+
Fertilized chicken eggs (White leghorn, Freddy’s Hu¨hnerhof
GmbH & Co. KG, Mainz, Germany) were incubated at 37.5 °C
and 62% relative humidity in an automatically turning egg incubator
(Ehret). After 70 h of incubation and removal of 8 mL of albumen,
a window was opened in the shell to uncover the developing
germination layer. The window was sealed with transparent tape;
the eggs were returned to the incubator and stored at 90% relative
humidity without turning. At 75-h incubation time, the angiogenic
experiment was initiated (start time 0 h) and 15 agarose pellets
were placed on the area vasculosa preparations. After 24 h, images
of the control and sample-treated areas were captured. Images were
digitized via Leica QWinTM-Software, and the angiogenic response
was measured using the automatic image analysis system pro-
grammed in a routine. Captured and digitized images were analyzed
for total, major (capillaries with a diameter of more than 50 µm),
and minor blood vessel areas using the customized image analysis
software Leica QWinTM (Leica Microsystems Bensheim, Ger-
many). Calculated data were sent automatically via dynamic data
exchange to an Excel table and processed statistically. An anti-
angiogenic response due to agent-treated samples was considered
positive if the mean of minor blood vessel area showed >20%
inhibition compared to control samples. Differences between treated
and control samples were evaluated using Student’s t-test.
Tubulin Polymerization Assay. The antimitotic activity of the
compounds was determined using a slightly modified version of
the turbidimetric tubulin assembly in vitro assay described by
Gaskin et al.50
In this assay, the assembly of microtubule protein isolated from
porcine brain (method according to Shelanski et al.50a is monitored
in vitro in the presence of effectors. The principle is based on the
fact that microtubules increase the turbidity of the test solution
compared to unpolymerized tubulin dimers. The degree of assembly
is measured by detecting the change in absorbance of the test
solution compared to a control.
The measurements were performed at 360 nm with a Uvikon
930 UV spectrophotometer (Kontron Instruments) equipped with
a multicell holder connected to a thermostat. The influence of the
test substances on tubulin assembly (protein concentration 1 mg/
mL) was determined by comparing the change of absorption to a
control without effectors. Assembly was performed for 20 min in
a PIPES (piperazine-1,4-bis-(2-ethanesulfonic acid)) buffer contain-
ing 10 mM GTP at pH 6.8 and a temperature of 37 °C. To exclude
protein denaturation and to check reversibilty of assembly, the
measurement was continued for additional 20 min while decreasing
the temperature to 4 °C, which should shift turbidity to the pre-
assembly level.
1), 408.1 (M+). Anal. (C22H20N2O6) C, H, N.
11-Methoxy-5,6,7-trimethoxy-1,2,3,8-tetrahydro-benzo[a]-
pyrrolo[3,4-c]carbazol-1,3-dione (16a). The product was isolated
by column chromatography (ethyl acetate/hexanes 4/6) from
reaction mixture 16 as red crystals (440 mg, 18%). Mp ) 290 °C.
FD-MS (m/z): 406.0 (M+). Anal. (C22H18N2O6) C, H, N.
3-(1-Methylindole-3-yl)-4-(3,4,5-trimethoxyphenyl)-1H-pyr-
role-2,5-dione (17). The general procedure was followed using
1-methylindole-3-methylglyoxylate (0.65 g, 3.0 mmol), 3,4,5-
trimethoxyphenylacetamide (0.69 g, 3.06 mmol), and 1.0 M KOBut
(5 mL, 5.0 mmol). Purification was achieved by column chroma-
tography (ethyl acetate/hexanes/CH2Cl2 3.5/6/0.5) to yield 17 (0.74
g, 62%) as red crystals. Mp ) 187 °C. EI-MS (m/z): 392.2 (M+).
Anal. (C22H20N2O6) calcd C 67.34, H 5.14, N 7.14; found C 66.94,
H 5.56, N 7.01.
5,6,7-Trimethoxy-8-methyl-1,2,3,8-tetrahydro-benzo[a]pyrrolo-
[3,4-c]carbazol-1,3-dione (17a). The product was isolated by
column chromatography (ethyl acetate/hexanes 4/6) from reaction
mixture 17 as pale red crystals (60 mg, 5%). Mp ) 185 °C. FD-
MS (m/z): 390.2 (M+). Anal. (C22H18N2O5) C, H, N.
3-(2-Methylindole-3-yl)-4-(3,4,5-trimethoxyphenyl)-1H-pyr-
role-2,5-dione (18). The general procedure was followed using
2-methylindole-3-ethylglyoxylate (1.60 g, 6.9 mmol), 3,4,5-tri-
methoxyphenylacetamide (1.2 g, 5.33 mmol), and 1.0 M KOBut
(20 mL, 20.0 mmol). Purification was achieved by column
chromatography (ethyl acetate/hexanes 1/1) to yield 18 (0.91 g,
44%) as red crystals. Mp ) 266 °C. EI-MS (m/z): 393.3 (M+
+
1). Anal. (C22H20N2O5) calcd C 67.34, H 5.14, N 7.14; found C
67.01, H 5.57, N 6.98.
3-(2-Phenylindole-3-yl)-4-(3,4,5-trimethoxyphenyl)-1H-pyrrole-
2,5-dione (19). The general procedure was followed using 2-phenyl-
indole-3-ethylglyoxylate (1.25 g, 4.26 mmol), 3,4,5-trimethoxy-
phenylacetamide (0.82 g, 3.06 mmol), and 1.0 M KOBut (10 mL,
10.0 mmol). Purification was achieved by column chromatography
(ethyl acetate/hexanes 1/1) to yield 19 (0.82 g, 56%) as red crystals.
Mp ) 304 °C. EI-MS (m/z): 455.2 (M+). Anal. (C27H22N2O5) C,
H, N.
3-(Indole-3-yl)-4-(1-naphthyl)-1H-pyrrole-2,5-dione (20). The
general procedure was followed using indole-3-ethylglyoxylate
(1.61 g, 7.40 mmol), 1-naphthylacetamide (1.0 g, 5.4 mmol), and
1.0 M KOBut (10 mL, 10.0 mmol). Purification was achieved by
column chromatography (ethyl acetate/hexanes 4/6) to yield 20 (0.44
g, 25%) as yellow crystals. Mp ) 212 °C. FD-MS (m/z): 339.8
(M+ + 1). Anal. (C20H17N3O5) calcd C 78.09, H 4.17, N 8.28; found
C 76.78, H 4.49, N 7.89.
3-(Indole-2-yl)-4-(3,4,5-trimethoxyphenyl)-1H-pyrrole-2,5-
dione (21). The general procedure was followed using indole-2-
ethylglyoxylate (1.0 g, 4.61 mmol), 3,4,5-trimethoxyphenylacet-
amide (0.8 g, 3.0 mmol), and 1.0 M KOBut (10 mL, 10.0 mmol).
Purification was achieved by column chromatography (ethyl acetate/
hexanes 1/1) to yield 21 (1.12 g, 65%) as red crystals. Mp ) 252
°C. EI-MS (m/z): 378.7 (M+). Anal. (C21H18N2O5) C, H, N.
Molecular Modeling. All modeling was performed on a Silicon
Graphics O2 (R10.000) workstation. For visualization and building
Selectivity Profiling of Compounds in Two Concentrations
Using 12 Protein Kinases. Recombinant Protein Kinases. The
inhibitory profile of compounds was determined using the follow-
ing 12 protein kinases: CDK1/cyclinB, CDK2/cyclinE, CDK4/
cyclinD1, CDK6/cyclinD1, PKC-R, PKC-γ, PKC-ꢀ, PKC-ι, EGF-
R, ERBB2, TIE2, and VEGF-R2.
All protein kinases were expressed in Sf9 insect cells as
recombinat GST fusion proteins or His-tagged proteins by means
of the baculovirus expression system. Except for PKC-R (mouse),

1280 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 4
Peifer et al.
(10) Gaukroger, K.; Hadfield, J. A.; Hepworth, L. A.; Lawrence, N. J.;
McGown, A. T. Novel syntheses of cis and trans isomers of
combretastatin A-4. J. Org. Chem. 2001, 66, 8135-8138.
(11) Wang, L.; Woods, K. W.; Li, Q.; Barr, K. J.; McCroskey, R. W.;
Hannick, S. M.; Gherke, L.; Credo, B. R.; Hui, Y. H.; Marsh, K.;
Warner, R.; Lee, J. Y.; Zielinski-Mozng, N.; Frost, D.; Rosenberg,
S. H.; Sham, H. L. Potent, orally active Heterocycle-based CA-4
analogues. Synthesis, Structure-Activity Relationship, Pharmaco-
kinetics, and in vivo antitumor activity evaluation. J. Med. Chem.
2002, 45, 1697-1711.
(12) Shirai, R.; Takayama, H.; Nishikawa, A.; Koiso, Y.; Hashimoto, Y.
Asymmetric synthesis of antimitotic combretadioxolane with potent
antitumor activity against multi-drug resistant cells. Bioorg. Med.
Chem. Lett. 1998, 8, 1997-2000.
PKC-ꢀ (mouse), and PKC-γ (rat), human cDNAs were used for
the expression of the protein kinases. Kinases were purified by
affinity chromatography using either GSH-agarose (Sigma) or Ni-
NTH-agarose (Qiagen). The purity and identity of each kinase was
checked by SDS-PAGE/silver staining and by Western blot
analysis with specific antibodies.
Protein Kinase Assay. A proprietary protein kinase assay
(33PanQinase Activity Assay) was used for measuring the kinase
activity of the 12 protein kinases. All kinase assays were performed
in 96-well FlashPlates from Perkin-Elmer/NEN (Boston, MA) in a
50 µL reaction volume.
The assay for all enzymes (except for the PKC assay, see below)
contained 60 mM HEPES-NaOH, pH 7.5, 3 mM MgCl2, 3mM
MnCl2, 3 µM sodium orthovanadate, 1.2 mM DTT, 50 µg/mL
PEG20000, 1 µM [γ-33P]-ATP (approximately 5 × 105 cpm per well),
recombinant protein kinase (50-400 ng), and depending on the
kinase, the following substrate proteins: Rb-CTF (CDK1, CDK4,
CDK6), poly(Glu,Tyr)4:1 (EGF-R, ERBB2, TIE2, VEGF-R2), and
histone H1 (CDK2).
The PKC assay contained 60 mM HEPES-NaOH, pH 7.5, 1
mM EDTA, 1.25 mM EGTA, 5 mM MgCl2, 1.32 mM CaCl2, 5
µg/mL phosphatidylserine, 1µg/mL 1,2-dioleyl glycerol, 1.2 mM
DTT, 50 µg/mL PEG20000, 1 µM [γ-33P]-ATP (approximately 5 ×
105 cpm per well), recombinant protein kinase (20-100 ng), and
histone H1 as substrate.
(13) Medarde, M.; Ramos, A. C.; Pelaez, R.; Lo´pez, J. L.; Gra´valos, D.
G.; San Feliciano, A. Synthesis and pharmacological activity of
diarylindole derivatives. Cytotoxic agents based on combretastatins.
Bioorg. Med. Chem. Lett. 1999, 9, 2303.
(14) Simoni, D.; Grisolia, G.; Giannini, G.; Roberti, M.; Rondanin, R.;
Piccagli, L.; Baruchello, R.; Rossi, M.; Romagnoli, R.; Invidiata, F.
P.; Grimaudo, S.; Jung, M. K.; Hamel, E.; Gebbia, N.; Crosta, L.;
Abbadessa, V.; Di Cristina, A.; Dusonchet, L.; Meli, M.; Tolomeo,
M. Heterocyclic and Phenyl Double-Bond-Locked Combretastatin
Analogues Possessing Potent Apoptosis-Inducing Activity in HL60
and in MDR Cell Lines. J. Med. Chem. 2005, 48 (3), 723-736.
(15) Moreau, P.; Anizon, F.; Sancelme, M.; Prudhomme, M.; Bailly, C.;
Carrasco, C.; Ollier, M.; Severe, D.; Riou, J. F.; Fabbro, D.; Meyer,
T.; Aubertin, A. M. Syntheses and biological evaluation of indolo-
carbazoles, analogues of rebeccamycin, modified at the imide
heterocycle. J. Med. Chem. 1998, 7, 41 (10), 1631-1640.
(16) Gray, N.; Detivaud, L.; Doering, C.; Meijer, L. ATP-site directed
inhibitors of cyclin-dependent kinases. Curr. Med. Chem. 1999, 6
(9), 859-875.
(17) Davis, P. D.; Hill, C. H.; Lawton, G.; Nixon, J. S.; Wilkinson, S. E.;
Hurst, S. A.; Keech, E.; Turner, S. E. Inhibitors of protein kinase C.
1. 2,3-Bisarylmaleimides. J. Med. Chem. 1992, 35 (1), 177-184.
(18) Zhang, H.-C.; Ye, H.; Conway, B. C.; Derian, C. K.; Addo, M. F.;
Kuo, G. H.; Hecker, L. R.; Croll, J. L.; Westover, L.; Xu, J. Z.;
Look, R.; Demarest, K. T.; Andrade-Gordon, P.; Damiano, B. P.;
Maryanoff, B. E. 3-(7-Azaindolyl)-4-arylmaleimides as potent, selec-
tive inhibitors of glycogen synthase kinase-3. Bioorg. Med. Chem.
Lett. 2004, 14, 3245-3250.
The compounds were tested at concentrations of 10 and 100 µM
in singlicate. The final DMSO concentration in the assay was 1%.
The reaction cocktails were incubated at 30 °C for 80 min. The
reaction was stopped with 50 µL of 2% (v/v) H3PO4, and plates
were aspirated and washed two times with 200 µL of 0.9% (w/v)
NaCl. Incorporation of 33Pi was determined with a microplate
scintillation counter (Microbeta Trilux, Wallac). All assays were
performed with a BeckmanCoulter/Sagian robotic system.
For each compound, inhibition was calculated as percentage
relative to control values without test compound.
(19) (a) Faul, M. M.; Winneroski, L. L.; Krumrich, C. A. A new one
Step Synthesis of maleimides by condensation of glyoxylate esters
with acetamides. Tetrahedron Lett. 1999, 40, 1109-1112. (b) Faul,
M. M.; Winneroski, L. L.; Krumrich, C. A. A new, efficient method
for the synthesis of bisindolylmaleimides. J. Org. Chem. 1998, 63,
6053-6058.
Acknowledgment. Financial support by the Ministerium fu¨r
Umwelt und Forsten/Mainz and the Fonds der Chemischen
Industrie, Germany, is gratefully acknowledged.
(20) Creary, X.; Mehrsheikh-Mohammadi, M. E. Captodative rate en-
hancements in the methylenecyclopropane rearrangement. J. Org.
Chem. 1986, 51, 2664-2668.
Supporting Information Available: MS, NMR, IR-data, and
selectivity profiling for compounds at 100 µM using 12 kinases.
This material is available free of charge via the Internet at http://
pubs.acs.org.
(21) Bensel, N.; Pevere, V.; Desmurs, J. R.; Wagner, A.; Mioskowski,
C. Mesyl guajacol: a versatile intermediate for the synthesis of
5-aminomethyl guajacol and related compounds. Tetrahedron Lett.
2002, 43, 4281-4283.
(22) (a) Okauchi, T.; Itonaga, M.; Minami, T.; Owa, T.; Kitoh, K.;
Yoshino, H. A general method for acylation of indoles at the
3-position with acyl chlorides in the presence of dialkylaluminium
chloride. Org. Lett. 2000, 2 (10), 1485-1487. (b) Ottoni, O.; Neder,
A. de V. F.; Dias, A. K. B.; Cruz, R. P. A.; Aquino, L. B. Acylation
of indole under Friedel-Crafts conditions- an improved method to
obtain 3-Acylindoles regioselectively. Org. Lett. 2001, 3 (7), 1005-
1007.
(23) Zhang, Z.; Yang, Z.; Wong, H.; Zhu, J.; Meanwell, N. A.; Kadow,
J. F.; Wang, T. An effective procedure for the acylation of azaindoles
at C-3. J. Org. Chem. 2002, 67, 6226-6227.
(24) Bray, B. L.; Mathies, P. H.; Naef, R.; Solas, D. R.; Tidwell, T. T.;
Artis, D. R.; Muchowski, J. M. N-(Triisopropylsilyl)pyrrole. A
progenitor “par excellence” of 3-substituted pyrroles. J. Org. Chem.
1990, 55, 6317-6328.
References
(1) Carmeliet, P.; Jain, R. K. Angiogenesis in cancer and other diseases.
Nature 2000, 407, 249-257.
(2) Fan, T. D.; Brem, S. In Angiosuppression. The search for new anti-
cancer drugs; Waring, M. J., Ponder, B., Eds.; Kluwer Academic
Press: Dordrecht, The Netherlands, 1992; pp 185-229.
(3) Yancopoulus, G. D.; Davis, S.; Gale, N. W.; Rudge, J. S.; Wiegand,
S. J.; Holash, J. Vascular-specific growth factors and blood vessel
formation. Nature 2000, 407, 242-248.
(4) Tozer, G. M.; Prise, V. E.; Wilson, J.; Locke, R. J.; Vojnovic, B.;
Stratford, M. R. L.; Dennis, M. F.; Chaplin, D. J. Combretastatin
A-4 phosphate as a tumor vascular-targeting agent: early effects in
tumors and normal tissues. Cancer Res. 1999, 59, 1626-1634.
(5) Jordan, A.; Hadfield, J. A.; Lawrence, N. J.; McGown, A. T. Tubulin
as a target for anticancer drugs: agents which interact with the mitotic
spindle. Med. Res. ReV. 1998, 18, 259-296.
(6) Flynn, B. L.; Hamel, E.; Jung, M. K. One-pot synthesis of benzo-
[b]furan and indole inhibitors of tubulin polymerization. J. Med.
Chem. 2002, 45, 2670-2673.
(7) Bibby, M. C. Combretastatin anticancer drugs. Drugs Future 2002,
27 (5), 475-480.
(8) Bailly, C.; Bal, C.; Barbier, P.; Combes, P.; Finet, J.-P.; Hildebrand,
M.-P.; Peyrot, V.; Wattez, N. Synthesis and biological evaluation of
4-Arylcoumarin analogues of combretastatins. J. Med. Chem. 2003,
46 (25), 5437-5444.
(9) Woods, J. A.; Hadfield, J. A.; Pettit, G. R.; Fox, B. W.; McGown,
A. T. The interaction with tubulin of a series of stilbenes based on
combretastatin A-4. Br. J. Cancer 1995, 71, 705-711.
(25) Tholander, J.; Bergman, J. Syntheses of 6,12-disubstituted 5,11-
dihydroindolo[3,2-b]carbazoles, including 5,11-dihydroindolo[3,2-
b]carbazole-6,12-dicarbaldehyde, an extremely efficient ligand for
the TCDD (Ah) receptor. Tetrahedron 1999, 55 (43), 12577-12594.
(26) Pindur, U.; Kim, Y.-S.; Mehrabani, F. Advances in indolo[2,3-a]-
carbazole chemistry: design and synthesis of Protein kinase C and
topoisomerase-I-inhibitors. Curr. Med. Chem. 1999, 6, 29-69.
(27) Reddy, G. M.; Chen, S.-Y.; Uang, B.-J. A facile synthesis of indolo-
[2,3-a]pyrrolo[3,4-c]carbazoles via oxidative photocyclization of
bisindolylmaleimides. Synthesis 2003, 4, 497-500.
(28) Harris, W.; Hill, C. H.; Keech, E.; Malsher, P. Oxidative cyclisations
with palladium acetate. A short synthesis of staurosporine aglykon.
Tetrahedron Lett. 1993, 34 (51), 8361-8364.

3,4-Diarylmaleimides as Angiogenesis Inhibitors
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 4 1281
(29) Noble, M. E.; Endicott, J. A.; Johnsen, L. N. Science 2004, 303,
1800-1805.
(41) Schollmeyer, D.; Peifer, C.; Dannhardt, G. 3-(3,4,5-Trimethoxy-
phenyl)-4-(3-hydroxy-4-methoxyphenyl)-1H-pyrrole-2,5-dione. Acta
Crystallogr. 2005, E61, 604-606.
(42) Peifer, C.; Schollmeyer, D.; Dannhardt, G. 3-(Indole-3-yl)-4-(3,4,5-
trimethoxyphenyl)-1H-pyrrole-2,5-dione. Acta Crystallogr. 2005,
E61, 721-723.
(43) Peifer, C.; Schollmeyer, D.; Dannhardt, G. 5,6,7-Trimethoxy-1,2,3,8-
tetrahydro-benzo[a]pyrrolo[3,4-c]carbazole-1,3-dione containing one
equivalent of DMSO. Acta Crystallogr. 2005, E61, 724-725.
(44) SYBYL Molecular Modeling Software, 6.8 ed., Tripos Inc.: St. Louis,
MO, 1992.
(45) Waldherr-Teschner, M.; Henn, Ch.; Vollhardt, H.; Reiling, S.;
Brickmann, J. Texture Mapping: a new tool for molecular graphics.
J. Mol. Graphics 1994, 12 (2), 89-105.
(46) Bo¨hm, H. J. The development of a simple empirical scoring function
to estimate the binding constant for a protein-ligand complex of
known three-dimensional structure. J. Comput.-Aided Mol. Des. 1994,
8 (3), 243-256.
(47) Gohlke, H.; Hendlich, M.; Klebe, G. Knowledge-based scoring
function to predict protein-ligand interactions. J. Mol. Biol. 2000,
295 (2), 337-356.
(48) Lawrie, A. M.; Noble, M. E.; Tunnah, P.; Brown, M. R.; Johnson,
L. N.; Endicott, J. A. Nat. Struct. Biol. 1997, 4, 796-801.
(49) Peifer, C.; Dannhardt, G. A novel quantitative chick embryo assay
as an angiogenesis model using digital image analysis. Anticancer
Res. 2004, 24, 3A, 1545-1552.
(50) (a) Gaskin, F.; Cantor, C. R.; Shelanski, M. L. Turbidimetric studies
of the in vitro assembly and disassembly of porcine neurotubules. J.
Mol. Biol. 1974, 89 (4), 737-755. (b) Shelanski, M. L.; Gaskin, F.;
Cantor, C. R. Microtubule assembly in the absence of added
nucleotides. Proc. Natl. Acad. Sci. U.S.A. 1973, 70, 765-768.
(30) Davies, T. G.; Pratt, D. J.; Endicott, J. A.; Johnson, L. N.; Noble,
M. E. M. Structure-based design of cyclin-dependent kinase inhibi-
tors. Pharmacol. Ther. 2002, 93, 125-133.
(31) Lawrie, A. M.; Noble, M. E.; Tunnah, P.; Brown, N. R.; Johnson,
L. N.; Endicott, J. A. Nat. Struct. Biol. 1997, 4, 796-801.
(32) Rarey, M.; Kramer, B.; Lengauer, T.; Klebe, G. A fast flexible
docking method using an incremental constructing algorithm. J. Mol.
Biol. 1996, 261 (3), 470-489.
(33) Sielecki, T. M.; Boylan, J. F.; Benfield, P. A.; Trainor, G. L. Cyclin-
dependent kinase inhibitors: useful targets in cell cycle regulation.
J. Med. Chem. 2000, 43, 1-18.
(34) Aitkenhead, M.; Christ, B.; Eichmann, A.; Feucht, M.; Wilson, D.
J.; Wilting, J. Paracrine and autocrine regulation of vascular
endothelial growth factor during tissue differentiation in the quail.
DeV. Dyn.1998, 212, 1-13.
(35) Vico, P. G.; Kyriacos, S.; Heymans, O.; Louryan, S.; Cartilier, L.
Dynamic study of the extraembryonic vascular network of the chick
embryo by fractal analysis. J. Ther. Biol. 1998, 195, 525-532.
(36) Neufeld, G.; Cohen, T.; Gengrinovitch, S.; Poltorak, Z. Vascular
endothelial growth factor (VEGF) and its receptors. FASEB J. 1999,
13, 9-22.
(37) Korpelainen, E. I.; Alitalo, K. Signaling angiogenesis and lymphan-
giogenesis. Curr. Opin. Cell Biol. 1998, 10, 159-164.
(38) Gille, H.; Kowalski, J.; Li, B.; LeCouter, J.; Moffat, B.; Zioncheck,
T. F.; Pelletier, L.; Ferrara, N. Analysis of biological effects and
signaling properties of Flt-1 (VEGF-R1) and KDR (VEGF-R2). J.
Biol. Chem. 2001, 276, 3222-3230.
(39) Dvorak, H. F. Vascular permeability factor/vascular endothelial
growth factor: a critical cytokine in tumor angiogenesis and a
potential target for diagnosis and therapy. J. Clin. Oncol. 2002, 20
(21), 4368-4380.
(40) McGovern, S. L.; Shoichet, B. K. Kinase inhibitors: Not just for
kinases anymore. J. Med. Chem. 2003, 46, 1478-1483.
JM0580297