Probing the difference between BH3 groove of Mcl-1 and Bcl-2 protein: Implications for dual inhibitors design

Article abstract of DOI:10.1016/j.ejmech.2011.05.062

Based on our previous discovery of a dual inhibitor of Mcl-1 and Bcl-2, 3-thiomorpholin-8-oxo-8H-acenaphtho[1,2-b]pyrrole-9-carbonitrile (1, S1), and guided by structure insight of 1 complex with Mcl-1 and Bcl-2, we exploited the spatial orientation of BH3 groove of the two proteins by a series of analogues of 1. These analogues contain substitutes with various steric hindrance designed to explore the width and length of the p2 pocket. The structure-activity relationships (SARs) studies together with docking studies and cell-based assays proved that the p2 pocket of Mcl-1 is relatively wider and shorter than that of Bcl-2. A novel dual inhibitor 6 was obtained based on these new findings that it exhibited nanomalar affinities toward Mcl-1 and Bcl-2, as well as nanomalar cytotoxicity activity against multiple cancer cell lines.

Full text of DOI:10.1016/j.ejmech.2011.05.062

European Journal of Medicinal Chemistry 46 (2011) 3909e3916
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Probing the difference between BH3 groove of Mcl-1 and Bcl-2 protein:
Implications for dual inhibitors design
,
a
,
,
Zhichao Zhang a , Hongna Yang 1, Guiye Wu a 1, Zhiqiang Li b, Ting Song b, Xiang qian Li a
*
a State Key Laboratory of Fine Chemicals, School of Chemistry, Dalian University of Technology, Dalian 116012, People’s Republic of China
b School of Life Science and Technology, Dalian University of Technology, Dalian 116024, People’s Republic of China
a r t i c l e i n f o
a b s t r a c t
Article history:
Based on our previous discovery of a dual inhibitor of Mcl-1 and Bcl-2, 3-thiomorpholin-8-oxo-8H-
acenaphtho[1,2-b]pyrrole-9-carbonitrile (1, S1), and guided by structure insight of 1 complex with Mcl-1
and Bcl-2, we exploited the spatial orientation of BH3 groove of the two proteins by a series of analogues
of 1. These analogues contain substitutes with various steric hindrance designed to explore the width
and length of the p2 pocket. The structureeactivity relationships (SARs) studies together with docking
studies and cell-based assays proved that the p2 pocket of Mcl-1 is relatively wider and shorter than that
of Bcl-2. A novel dual inhibitor 6 was obtained based on these new findings that it exhibited nanomalar
affinities toward Mcl-1 and Bcl-2, as well as nanomalar cytotoxicity activity against multiple cancer cell
lines.
Received 20 April 2011
Received in revised form
22 May 2011
Accepted 24 May 2011
Available online 31 May 2011
Keywords:
Apoptosis
Mcl-1
Ó 2011 Elsevier Masson SAS. All rights reserved.
Bcl-2
p2 pocket
Dual inhibitor
1. Introduction
that subtly control the orientation of binding could perhaps be
exploited in the design of more potent Mcl-1 inhibitors [8].
Aberrant over-expression of anti-apoptotic B-cell lymphoma 2
(Bcl-2)-like proteins combats tumor cell resistance by programmed
cell death [1,2]. Therefore, strategies that target the functional BH3
binding groove of these Bcl-2-like members may provide a selec-
tive mechanism for killing tumor cells. Since there are more than
one member in Bcl-2-like family, the most versatile cell killer must
be a broad-spectrum “pan” Bcl-2 inhibitor, capable of targeting the
BH3 groove of multiple Bcl-2 family proteins [3].
Myeloid cell leukemia sequence 1 (Mcl-1) is one of the Bcl-2-like
proteins which has recently been revealed to play a critical and
distinct role in the regulation of apoptosis [4]. Furthermore, it has
been well established that a “pan-Bcl-2 inhibitor” should be
a compound that may not bind to all of the Bcl-2-like proteins but
that can bind to at least Mcl-1 and Bcl-2 to disarm the apoptosis
capacities of these key targets [5]. Unfortunately, a few of small
molecules described to date are dual inhibitors that bind to Mcl-
1and Bcl-2. The most authentic BH3 mimetic ABT737, for
example, is not a pan-Bcl-2 inhibitor [6,7]. However, a recent
publication reported that the structural features in the antagonists
Recently, we have reported a small molecule 3-thiomorpholin-
8-oxo-8H-acenaphtho[1,2-b]pyrrole-9-carbonitrile (1, S1) as a dual
nanomolar inhibitor of Mcl-1 and Bcl-2 (Ki ¼ 58 and 310 nM,
respectively) (Fig. 1) [9]. Further structureeactivity relationship
studies revealed that 1 lies in the BH3 groove of Mcl-1 and Bcl-2,
with the thiomorpholine extended into the p2 binding pocket of
Mcl-1 and Bcl-2 [10].
To guide the design of Mcl-1/Bcl-2 dual inhibitors, especially
predict the occupation of their p2 pockets, we further probed the
difference between the p2 pocket of Mcl-1 and Bcl-2 in this study.
Based on two series potent pan-active inhibitors of structurally
related compounds with different binding profiles against Mcl-1
and Bcl-2, we not only established the molecular determinants
governing the specificity of ligand engaging into the p2 of Mcl-1
and Bcl-2, but also got a potent compound 6, binding to Mcl-1
and Bcl-2 proteins with Ki values of 24 and 158 nM, respectively,
which provide some fundamental insights into the future design
and development of Mcl-1 and Bcl-2 inhibitors.
2. Chemistry
* Corresponding author. Tel./fax: þ86 411 84986032.
E-mail address: zczhang@dlut.edu.cn (Z. Zhang).
These authors contributed equally to this work.
The 8-oxo-8H-acenaphtho[1,2-b]pyrrole-9-carbonitrile deriva-
tives 2aec, 3aec and 6 were synthesized by the procedure reported
1
0223-5234/$ e see front matter Ó 2011 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ejmech.2011.05.062

3910
Z. Zhang et al. / European Journal of Medicinal Chemistry 46 (2011) 3909e3916
Since accommodation in p2 pocket of both Mcl-1 and Bcl-2
could determine whether a molecule is a dual inhibitor of Mcl-1/
Bcl-2, we aimed to further probe the character of this pocket in
the two proteins. Because the X-ray evidence of the BH3 groove of
Bcl-2 protein was published very recently [15], at the beginning of
this study we had to compare Mcl-1 with Bcl-xL, whose three-
dimensional architecture is quite similar with that of Bcl-2 [16a].
Either Bcl-2 or Bcl-xL composes of eight a-helices with a hydro-
phobic groove on the surface. The overall backbone RMSD (root
ꢀ
mean square deviation) between them is only about 1.85 A,
excluding the loop between
Fig. 1. Structures of compounds 1, 2aec, 3aec and 4e6.
a1 and a2 [16b]. Research showed helix
a3 is well formed in the Mcl-1 but poorly so in the Bcl-xL, which
results in the protein backbones of the p1 and p2 pockets are less
contiguous in the Mcl-1 than Bcl-xL (Fig. 2a) [17,18]. Additionally,
the BH3 groove of Bcl-xL is narrow than that of Mcl-1 [18]. When
crystallized Bcl-2/Bak complex was reported, we compared Mcl-
1with Bcl-2 directly by using AutoDock tools (ADT). Result identi-
fied the BH3 groove of Bcl-2 is narrow than that of Mcl-1 (Fig. 2b).
Here, we focused synthetic two series analogues of 1 specially
designed to probe the spatial orientation of the p2 pocket that
thiomorpholine of 1 was replaced with substituent of various bulk
and steric hindrance.
in Scheme 1. Compound 8 (5-bromoacenaphthenequinone) was
readily accessible via the reaction between acenaphthenequinone
and bromine under reflux condition [11]. Compounds 9aeg were
obtained in high yields (85e95%) by the reaction between 8 and
corresponding phenol with K2CO3 in DMF. The key step in the
synthesis of target structures is the Knoevenagel condensation
reaction [12,13] between substituted acenaphthenequinones
(9aeg) and malononitrile, using silica gel in room temperature to
give monoadducts (10aeg) with yields of 40e60% (Scheme 1). The
target compounds 2aec, 3aec and 6 were prepared by a “one-pot”
base-catalyzed cyclization reaction of 10aeg, respectively [14]. All
steps were in good yields and under mild conditions.
3.2. Structureeactivity relationship
We substituted the thiomorpholine using oxygen atoms as
a linker, which renders flexibility to compounds, allowing them to
engage well into the p2 pocket. Specifically, we examined the effect
of various steric hindrance by using the methyl at the ortho-, meta-,
and para-position of phenyl, yielding analogues 3-(o-tolyloxy)-
8-oxo-8H-acenaphtho[1,2-b]pyrrole-9-carbonitrile (2a), 3-(m-tol-
3. Results and discussion
3.1. Rationale
Our previous studies identified 1 lies along a BH3 hydrophobic
binding pocket of Mcl-1 and Bcl-2. A hydrogen bonding network
could be formed between carbonyl group of 1 and R263 of Mcl-1
and R146 of Bcl-2, respectively [10]. The similar positioning of
R263 of Mcl-1 and R146 of Bcl-2 in their three-dimensional struc-
ture allows 1 to bind in similar orientations within their BH3
binding grooves. As such, thiomorpholine extends to the p2 pocket
when 1 binds to the two proteins. Further, we revealed that occu-
pation of the two protein’s p2 pocket is crucial for a Mcl-1/Bcl-2
inhibitor.
yloxy)-8-oxo-8H-acenaphtho[1,2-b]pyrrole-9-carbonitrile
(2b),
3-(p-tolyloxy)-8-oxo-8H-acenaphtho[1,2-b]pyrrole-9-carbonitrile
(2c), respectively (Fig. 1). The binding affinities Ki of the compounds
were evaluated using fluorescence polarization assays (FPAs) that
measure their abilities to competitively displace a Bid-derived
peptide from Mcl-1 and Bcl-2 as described in the biological assay.
The competitive binding curves of these compounds to Mcl-1 and
Bcl-2 are outlined in Fig. 3a and b, respectively. The para-methyl
substituted compound 2c exhibited potent inhibition against both
Scheme 1. Synthesis of compounds 2aec, 3aec and 6. Reagents and conditions: (a) Br2, 60 ꢀC; (b) RePhOH, K2CO3, DMF, 60 ꢀC; (c) CH2(CN)2, C2HCl2, silica gel; (d) K2CO3, CH3CN,
reflux.

Z. Zhang et al. / European Journal of Medicinal Chemistry 46 (2011) 3909e3916
3911
Fig. 2. Overlay of the crystal structures of (a) Mcl-1:Bcl-xL and (b) Mcl-1:Bcl-2 by using AutoDock Tools (ADT). Mcl-1, Bcl-xL, and Bcl-2 are shown in white, pink and blue,
respectively. The significant difference in 3 of Mcl-1 and Bcl-xL is circled and the p1 and p2 pockets are labeled in (a). The dotted line in (b) represents the width of BH3 groove
a
(vertical to the axis of the BH3 groove) formed largely by the
web version of this article.)
a3, a4, and a5 helices (For interpretation of the references to color in this figure legend, the reader is referred to the
Mcl-1 and Bcl-2 (174 nM and 445 nM, respectively). Interestingly,
when shifting the methyl group from para-position to ortho/meta-
position, different specificities for Mcl-1 and Bcl-2 were found. For
Mcl-1, the Ki of ortho-substituted 2a (213 nM) is similar with that of
2c, and nearly a 3-fold decrease was found for meta-substituted 2b
(445 nM). A much more significant decrease of inhibition was
suggested that the position change of the larger methoxy group
leads to more remarkable influence on affinity toward Bcl-2 than
that of methyl. When the para-methoxy substituted phenyl moiety
can be accommodated by both Mcl-1 and Bcl-2, the meta-methoxy
substituted phenyl can only be accepted by Mcl-1 and almost
completely rejected by Bcl-2. This strongly supported that the p2
pocket of Bcl-2 is narrow than that of Mcl-1.
found for Bcl-2. Compound 2a (1.43
mM) lost nanomolar affinity,
while 2b (6.03 M) even showed a remarkable 12-fold decrease.
m
To further investigate the effect of position of methyl/methoxy
phenol moiety, we analyzed the predicted structures of the meta-
methoxy substituted 3b and para-methoxy substituted 3c in
complex with human Mcl-1 and Bcl-2 proteins respectively, with
the help of computational modeling studies by using AutoDock 4.0.
As shown in Fig. 4a and c, compounds 3b and 3c lie along
a hydrophobic binding pocket of Mcl-1 similar with the docking
result of 1 [10]. It suggested whether methoxy group is attached
para or meta-position, it could be accommodated by the p2 pocket
of Mcl-1. By contrast, the p2 pocket of Bcl-2 could not endure the
meta-methoxy substituted phenyl group of 3b so well as the para-
methoxy substituted phenyl group of 3c (Fig. 4b and d). This further
identified the width (vertical to the long axis of BH3 groove) of the
p2 pocket of Bcl-2 is narrow than that of Mcl-1. As such, the group
at meta-position is fatal for Bcl-2, which suggested that the para-
substituent of phenyl was optimal for occupying the p2 of both Mcl-
1 and Bcl-2.
The data above indicated that the position of the methyl group is
crucial for occupying Bcl-2 over Mcl-1. This supported the differ-
ence between BH3 groove of Mcl-1 and Bcl-2, especially in the p2
pocket.
For further test the difference between p2 pocket of Mcl-1 and
Bcl-2, we replaced the methyl with a larger methoxy group at the
ortho-, meta- and para-position of phenyl, yielding another series
compounds, 3-(2-methoxyphenoxy)-8-oxo-8H-acenaphtho[1,2-b]
pyrrole-9-carbonitrile (3a), 3-(3-methoxyphenoxy)-8-oxo-8H-ace-
naphtho[1,2-b]pyrrole-9-carbonitrile (3b), 3-(4-methoxyphenoxy)-
8-oxo-8H-acenaphtho[1,2-b]pyrrole-9-carbonitrile (3c), respec-
tively (Fig. 1). Excitingly, the same trend was found for these three
compounds that the position of the methyl group is crucial for
occupying Bcl-2 over Mcl-1. For Mcl-1, the Ki value of para-
substituted 3c was 122 nM, which is similar with ortho-substituted
3a (149 nM) and a little enhanced than meta-substituted 3b
(375 nM). For Bcl-2, an about 3-fold lower affinity was found for 3a
(886 nM, Fig. 3b) than 3c (304 nM), while a 50 times decrease of 3b
Subsequently, we calculated the width of the substituted phenyl
group of different compounds by ChemBioDraw and AutoDock
tools at the minimize energy of the structures. As shown in Table 1,
(16.6 mM) revealed most of the affinities have been lost. These data
Fig. 3. Competitive binding curves of designed small-molecule inhibitors 2aec and 3aec to (a) Mcl-l and (b) Bcl-2 as determined by FPAs.

3912
Z. Zhang et al. / European Journal of Medicinal Chemistry 46 (2011) 3909e3916
Fig. 4. Predicted binding models of 3c in complex with (a) Mcl-1, (b) Bcl-2, and 3b in complex with (c) Mcl-1, (d) Bcl-2. Mcl-1 and Bcl-2 are shown in a surface representation, and
the carbon, oxygen, nitrogen, and sulfur atoms are shown in gray, red, blue, and yellow, respectively. The p1 and p2 pockets are labeled (For interpretation of the references to color
in this figure legend, the reader is referred to the web version of this article.)
compounds 2c and 3c, which are most potent toward both Mcl-1
(Ki ¼ 1281 nM) toward Mcl-1, in which the isopropyl group was
replaced with a longer phenyl group. Here, we tested whether 5
could still occupy the p2 pocket of Bcl-2. FPA showed 5 exhibited
a potent affinity toward Bcl-2 with a Ki of 308 nM (Fig. 5b). This
confirmed our hypothesis that in the orientation along the long axis
of BH3 groove Mcl-1 is more constricted than Bcl-2.
Consequently, we aimed to design a more potent dual inhibitor
based on our observation. We supposed that replacing phenyl
group of 5 with a shorter sec-butyl may occupy the p2 pocket of
both Mcl-1 and Bcl-2 and then enhance the affinity to them.
Therefore, we synthesized 3-(4-sec-butylphenoxy)-8-oxo-8H-ace-
naphtho[1,2-b]pyrrole-9-carbonitrile (6) (Fig. 1). As expected, we
detected a significant improved affinity for 6 toward Mcl-1 and Bcl-
2 (Ki ¼ 24 nM and 158 nM, respectively, Fig. 5a). About 2e3 times
enhanced affinity than its parent 1 (58 and 310 nM toward Mcl-1
and Bcl-2, respectively) was found. These data supported our
hypothesis that a proper spatial bulk, including both width and
length, is a determination for a molecule to occupy the p2 pocket of
both Mcl-1 and Bcl-2.
Furthermore, we assessed the cytotoxicity of compounds 3b, 3c
and 6 against MCF-7, K562 and HL-60 cell lines by using MTT
method. The results are shown in Fig. 6. When 3b and 3c showed
comparable mediated cytotoxicity on MCF-7 and K562 cells, much
higher cytotoxicity was found for 6. Of note, on HL-60 cells whose
Bcl-2 level is much higher than that in MCF-7 and K562 cells, 3b
exhibited much weaker killing ability than compound 3c. This
could be explained that 3b could not overcome the protection of
high level of Bcl-2 protein due to its poor affinity toward Bcl-2
ꢀ
and Bcl-2, exhibit 4.32 A in width. By comparison, 3b, whose
ꢀ
width is 7.16 A, was completely excluded by the p2 pocket of Bcl-2,
while it still can be accommodated by that of Mcl-1.
So far, we identified that the p2 pocket of Bcl-2 is indeed narrow
than that of Mcl-1. Surprisingly, it seems this result is not consistent
with the result of our previous study. In that study we found the
molecule containing a t-amyl group at the para-position of phenyl
could enter into the p2 of Bcl-2 but failed to enter into that of Mcl-1,
which suggested that the p2 of Mcl-1 is more challenging than that
of Bcl-2. We hypothesized this is largely due to the discontinuous of
the p1 and p2 pocket in Mcl-1. In another word, there is a hindrance
between the p1 and p2 pocket in Mcl-1, which is a challenge to the
length of an inhibitor (along the long axis of BH3 groove). Next,
exploration of the proper length of group at the para-position of
phenyl was carried out, which may engage well in the p2 pocket of
Bcl-2 without adversely affecting the occupation that of Mcl-1. We
used to identify that compound 4 containing isopropyl group at the
para-position of phenyl group is a dual nanomolar inhibitor of Mcl-
1 and Bcl-2 (Ki ¼ 99 nM and 232 nM, respectively). However,
a more significant decrease of inhibition was found for compound 5
Table 1
The width of the substituted phenyl group of different compounds was calculated by
ChemBioDraw and AutoDock tools.
Compounds
Width/A
2a
5.273
2b
5.903
2c
4.320
3a
6.230
3b
7.159
3c
4.317
ꢀ

Z. Zhang et al. / European Journal of Medicinal Chemistry 46 (2011) 3909e3916
3913
a
b
100
80
60
40
20
0
100
80
60
40
20
0
-2
-1
0
1
2
3
4
-2
-1
0
1
2
3
4
inhibitor (Log[µM])
inhibitor (Log[µM])
Fig. 5. Competitive binding curves of designed compounds 4e6 to (a) Mcl-l and (b) Bcl-2 as determined by FPAs.
Taken together, we have successfully discovered a new potent
and dual, nanomolar small-molecule inhibitor of Mcl-1 and Bcl-2
based on the structural insight into the difference between the p2
pocket of Mcl-1 and Bcl-2.
5. Experimental protocols
5.1. Chemistry
5.1.1. Materials and methods
All commercial reagents were purchased and used without
further purification or distillation unless otherwise stated. Melting
points were measured on a Griffin apparatus and are uncorrected.
NMR (1H, 13C) was obtained with Bruker AV-400 spectrometer with
chemical shifts reported as parts per million (in CDCl3, TMS as
internal standard). The following abbreviations are used for
multiplicity of NMR signals: s ¼ singlet, d ¼ doublet, t ¼ triplet,
q ¼ quartet, m ¼ multiplet. High-resolution mass spectra (HRMS)
were obtained on HPLC-Q-Tof MS (Micro) spectrometer. Column
chromatography was performed on silica gel 200e300 mesh. Purity
of all final products was determined by analytical HPLC to be ꢁ95%.
HPLC purity of compounds was measured with a reverse phase
Fig. 6. Cytotoxicity evaluation of compounds 3b, 3c and 6 against tumor cell lines HL-
60, K562, and MCF-7 (a). Histograms represent the IC50 values (
triplicate analyses). The levels of Mcl-1 and Bcl-2 expression were analyzed by Western
m
M, means ꢃ SD of
HPLC (XBridge C18, 4.6 ꢂ 150 mm, 5
mm) with two diverse wave-
blot in the three cell lines (b). The levels of
loading.
b-actin were determined to assure equal
length detection systems.
5.1.2. General procedure for the preparation of 9aeg
mixture of 5-bromoacenaphthenequinone (8) (260 mg,
protein. By contrast, over-expression of either Mcl-1 or Bcl-2 could
not protect cells from compound 6. These data were in agreement
with our discovery in FP assay and further identify the strategy
represented in this study for design of dual inhibitor.
A
1.0 mmol), corresponding phenol (1.2 mmol) and K2CO3 (165 mg,
1.2 mmol) was stirred in DMF (10 mL) for 2 h at 50e60 ꢀC. The
solution was poured into a saturated NaCl aqueous solution
(20 mL), followed by extraction with ethyl acetate (3 ꢂ 20 mL). The
organic layer was collected and dried with dry MgSO4. After
evaporation under reduced pressure, the product was purified by
column chromatography on silica gel using dichloromethane/
petroleum ether from 1:1 to 2:1.
4. Conclusions
In summary, we substituted the 3-position of compound 1 to
further probe the spatial orientation of the p2 pocket of Mcl-1 and
Bcl-2. SAR studies together with docking studies revealed on the
one hand that the p2 pocket of Bcl-2 could not endure moiety at the
ortho or meta-position so well as the para-position, which to
a much extent was different from Mcl-1. On the other hand,
although the para-substituted could be accommodated well by
both Mcl-1 and Bcl-2, the p2 pocket of Mcl-1 cannot endure relative
longer group than Bcl-2. It suggested that the p2 pocket of Mcl-1
prefers wide and short moieties while that of Bcl-2 is apt to
narrow and long group.
5.1.2.1. 5-(p-Tolyloxy)acenaphthenequinone (9c). Yield: 0.270 g,
94%. 1H NMR (400 MHz, CDCl3):
J ¼ 8.4 Hz, 1H), 8.18 (d, J ¼ 7.6 Hz, 1H), 7.86 (t, J ¼ 7.6 Hz, 1H), 7.45 (d,
J ¼ 8.4 Hz, 2H), 7.25 (d, J ¼ 8.4 Hz, 2H), 7.08 (d, J ¼ 7.6 Hz, 1H), 2.43
(s, 3H). TOF MS (EIþ): C19H12O3, calcd 288.0786, found 288.0780.
d
8.48 (d, J ¼ 8.4 Hz, 1H), 8.33 (d,
5.1.2.2. 5-(4-Methoxyphenoxy)acenaphthenequinone
(9f). Yield:
8.44 (d, J ¼ 8.4 Hz, 1H),
0.276 mg, 91%. 1H NMR (400 MHz, CDCl3):
d
With this regard, we synthesized compound 6 containing a sec-
butyl group at para-position of phenyl. FP-based studies and cell-
based studies identified it is a dual inhibitor with much improved
activities than compound 1.
8.35 (d, J ¼ 9.2 Hz, 2H), 8.18 (d, J ¼ 8.0 Hz, 1H), 8.11 (d, J ¼ 8.0 Hz,
1H), 7.89 (t, J ¼ 8.0 Hz, 1H), 7.65 (t, J ¼ 9.2 Hz, 1H), 7.30 (t, J ¼ 9.2 Hz,
2H), 7.25 (d, J ¼ 8.4 Hz, 1H). TOF MS (EIþ): C19H12O4, calcd 304.0736,
found 304.0735.

3914
Z. Zhang et al. / European Journal of Medicinal Chemistry 46 (2011) 3909e3916
5.1.3. General procedure for the preparation of 10aeg
To a solution was added the corresponding compounds 9aeg
(1 mmol) and malononitrile (1.5 mmol) in 3 mL CH2Cl2 in podgy
silica gel containing with CH2Cl2 and then column chromatography
to yield a red power, which was used in the next step without
further purification.
refluxed for 2 h. After the reaction was complete as monitored by
TLC, the solvent was evaporated under reduced pressure and the
product was purified by column chromatography on silica gel using
dichloromethane/petroleum ether (1:1).
5.1.4.1. 3-(o-Tolyloxy)-8-oxo-8H-acenaphtho[1,2-b]pyrrole-9-carbo-
nitrile (2a). Yield: 0.14 g, 42%. Mp ¼ 210e211 ꢀC. 1H NMR
(400 MHz, CDCl3):
d
8.96 (d, J ¼ 8.0 Hz, 1H), 8.63 (d, J ¼ 8.0 Hz, 1H),
5.1.3.1. 2-[2-Oxo-2H-6-(o-tolyloxy)acenaphthylen-1-ylidene]-malo-
8.48 (d, J ¼ 8.0 Hz, 1H), 8.01 (t, J ¼ 8.0 Hz, 1H), 7.39 (d, J ¼ 7.6 Hz, H),
7.35 (d, J ¼ 8.0 Hz, H), 7.32 (d, J ¼ 7.6 Hz, 1H), 7.14 (d, J ¼ 8.0 Hz, 1H),
6.87 (d, J ¼ 8.0 Hz, H), 2.19 (s, 3H). TOF MS (EIþ): C22H12N2O2, calcd
336.0899, found 336.0890.
nonitrile (10a). Yield: 0.14 g, 42%. Mp ¼ 185e187 ꢀC. 1H NMR
(400 MHz, CDCl3):
d
8.61 (d, J ¼ 7.8 Hz, 1H), 8.00 (d, J ¼ 8.0 Hz, 1H),
7.88 (d, J ¼ 7.6 Hz, 1H), 7.39 (d, J ¼ 7.2 Hz, 2H), 7.33 (t, J ¼ 7.6 Hz, 2H),
7.30 (d, J ¼ 7.6 Hz, 1H), 7.14 (d, J ¼ 7.6 Hz, 1H), 2.21 (s, 3H). TOF MS
(EIþ): C22H12N2O2, calcd 336.0899, found 336.0887.
5.1.4.2. 3-(m-Tolyloxy)-8-oxo-8H-acenaphtho[1,2-b]pyrrole-9-carbo-
nitrile (2b). Yield: 0.18 g, 53%. Mp ¼ 222e223 ꢀC. 1H NMR
5.1.3.2. 2-[2-Oxo-2H-6-(m-tolyloxy)acenaphthylen-1-ylidene]-malono-
(400 MHz, CDCl3):
d
8.91 (d, J ¼ 8.4 Hz, 1H), 8.65 (d, J ¼ 8.4 Hz, 1H),
nitrile (10b). Yield: 0.16 g, 48%. Mp ¼ 176e177 ꢀC. 1H NMR (400 MHz,
8.46 (d, J ¼ 7.6 Hz, 1H), 7.87 (t, J ¼ 7.6 Hz, 1H), 7.41 (d, J ¼ 8.4 Hz, 2H),
7.19 (d, J ¼ 8.4 Hz, 1H), 7.03 (d, J ¼ 7.6 Hz, 2H), 2.43 (s, 3H). TOF MS
(EIþ): C22H12N2O2, calcd 336.0899, found 336.0894.
CDCl3):
d
8.59 (d, J ¼ 8.0 Hz, 1H), 8.54 (d, J ¼ 8.0 Hz, 1H), 8.03
(d, J ¼ 7.6 Hz, 1H), 7.86 (d, J ¼ 8.4 Hz, 2H), 7.39 (t, J ¼ 7.6 Hz, 1H), 7.16
(d, J ¼ 8.4 Hz, H), 7.02 (t, J ¼ 7.6 Hz, 1H), 7.00 (d, J ¼ 7.6 Hz, 1H), 2.43
(s, 3H). TOF MS (EIþ): C22H12N2O2, calcd 336.0899, found 336.0890.
5.1.4.3. 3-(p-Tolyloxy)-8-oxo-8H-acenaphtho[1,2-b]pyrrole-9-carbo-
nitrile (2c). Yield: 0.20 g, 60%. Mp ¼ 230e231 ꢀC. 1H NMR
(400 MHz, CDCl3):
d
8.92 (d, J ¼ 8.4 Hz, 1H), 8.63 (d, J ¼ 8.4 Hz, 1H),
5.1.3.3. 2-[2-Oxo-2H-6-(p-tolyloxy)acenaphthylen-1-ylidene]-malo-
nonitrile (10c). Yield: 0.18 g, 54%. Mp ¼ 227e228 ꢀC. 1H NMR
8.45 (d, J ¼ 7.6 Hz,1H), 7.86 (t, J ¼ 7.6 Hz, 1H), 7.32 (d, J ¼ 8.4 Hz, 2H),
7.09 (d, J ¼ 8.4 Hz, 2H), 7.01 (d, J ¼ 7.6 Hz, 1H), 2.44 (s, 3H). TOF MS
(EIþ): C22H12N2O2, calcd 336.0899, found 336.0892.
(400 MHz, CDCl3):
d
8.59 (d, J ¼ 8.0 Hz, 1H), 8.57 (d, J ¼ 8.0 Hz, 1H),
8.01 (d, J ¼ 8.0 Hz,1H), 7.87 (t, J ¼ 8.0 Hz,1H), 7.31 (d, J ¼ 8.0 Hz, 2H),
7.11 (d, J ¼ 8.0 Hz, 2H), 6.98 (d, J ¼ 8.0 Hz, 1H), 2.43 (s, 3H). TOF MS
(EIþ): C22H12N2O2, calcd 336.0899, found 336.0888.
5.1.4.4. 3-(2-Methoxyphenoxy)-8-oxo-8H-acenaphtho[1,2-b]pyrrole-
9-carbonitrile (3a). Yield: 0.23 g, 65%. Mp ¼ 250e251 ꢀC. 1H NMR
(400 MHz, CDCl3):
d
8.97 (d, J ¼ 8.0 Hz, 1H), 8.63 (d, J ¼ 8.4 Hz, 1H),
5.1.3.4. 2-[2-Oxo-2H-6-(2-methoxyphenoxy)acenaphthylen-1-ylidene]-
malononitrile (10d). Yield: 0.15 g, 43%. Mp ¼ 194e195 ꢀC. 1H
8.46 (d, J ¼ 8.0 Hz, 1H), 7.87 (t, J ¼ 8.0 Hz, 1H), 7.37 (t, J ¼ 8.0 Hz, 1H),
7.23 (s, H), 7.15 (d, J ¼ 7.2 Hz, H), 7.11 (d, J ¼ 8.0 Hz, H), 7.09 (d,
J ¼ 7.2 Hz, 1H), 3.75 (s, 3H). TOF MS (EIþ): C22H12N2O3, calcd
352.0848, found 352.0840.
NMR (400 MHz, CDCl3):
d
8.62 (d, J ¼ 8.0 Hz, 1H), 8.59
(d, J ¼ 8.0 Hz, H), 8.00 (d, J ¼ 8.0 Hz, 1H), 7.88 (d, J ¼ 8.0 Hz, 1H), 7.35 (t,
J ¼ 8.0 Hz, 1H), 7.23 (t, J ¼ 8.0 Hz, 1H), 7.13 (t, J ¼ 8.4 Hz, 2H), 6.84 (d,
J ¼ 8.4 Hz, 1H), 3.76 (s, 1H). TOF MS (EIþ): C22H12N2O3, calcd 352.0848,
found 352.0839.
5.1.4.5. 3-(4-Methoxyphenoxy)-8-oxo-8H-acenaphtho[1,2-b]pyrrole-
9-carbonitrile (3b). Yield: 0.24 g, 68%. Mp ¼ 253e254 ꢀC. 1H NMR
(400 MHz, CDCl3):
d
8.94 (d, J ¼ 8.0 Hz, 1H), 8.63 (d, J ¼ 8.4 Hz, 1H),
5.1.3.5. 2-[2-Oxo-2H-6-(3-methoxyphenoxy)acenaphthylen-1-ylidene]-
8.46 (d, J ¼ 8.0 Hz,1H), 7.88 (t, J ¼ 8.0 Hz,1H), 7.26 (d, J ¼ 7.2 Hz, 2H),
7.13 (d, J ¼ 7.2 Hz, 2H), 7.04 (d, J ¼ 8.8 Hz, 1H), 3.83 (s, 3H). TOF MS
(EIþ): C22H12N2O3, calcd 352.0848, found 352.0841.
malononitrile (10e). Yield: 0.14 g, 40%. Mp ¼ 178e179 ꢀC. 1H NMR
(400 MHz, CDCl3):
d
8.59 (d, J ¼ 8.4 Hz, 1H), 8.54 (d, J ¼ 8.4 Hz, 2H),
8.04 (d, J ¼ 8.0 Hz, 1H), 7.86 (d, J ¼ 8.0 Hz, 1H), 7.40 (t, J ¼ 8.0 Hz, 1H),
7.06 (d, J ¼ 8.0 Hz, 1H), 6.87 (d, J ¼ 8.4 Hz, H), 6.80 (d, J ¼ 8.4 Hz, 2H),
3.84 (s, 1H). TOF MS (EIþ): C22H12N2O3, calcd 352.0848, found
352.0841.
5.1.4.6. 3-(4-Methoxyphenoxy)-8-oxo-8H-acenaphtho[1,2-b]pyrrole-
9-carbonitrile (3c). Yield: 0.30 g, 85%. Mp ¼ 257e258 ꢀC. 1H NMR
(400 MHz, CDCl3):
d
8.92 (d, J ¼ 8.0 Hz, 1H), 8.62 (d, J ¼ 8.8 Hz, 1H),
8.44 (d, J ¼ 8.0 Hz, 1H), 7.86 (t, J ¼ 8.0 Hz, 1H), 7.15 (d, J ¼ 7.2 Hz, 2H),
7.03 (d, J ¼ 7.2 Hz, 2H), 7.00 (d, J ¼ 8.8 Hz, 1H), 3.88 (s, 3H). TOF MS
(EIþ): C22H12N2O3, calcd 352.0848, found 352.0839.
5.1.3.6. 2-[2-Oxo-2H-6-(4-methoxyphenoxy)acenaphthylen-1-ylidene]-
malononitrile (10f). Yield: 0.19 g, 54%. Mp ¼ 233e234 ꢀC. 1H NMR
(400 MHz,
CDCl3):
d
8.44
(d,
J ¼ 8.4 Hz, 1H),
8.35
(d, J ¼ 9.2 Hz, 2H), 8.18 (d, J ¼ 8.0 Hz, 1H), 8.11 (d, J ¼ 8.0 Hz, 1H), 7.89
(t, J ¼ 8.0 Hz, 1H), 7.65 (t, J ¼ 9.2 Hz, 1H), 7.30 (t, J ¼ 9.2 Hz, 2H), 7.25 (d,
J ¼ 8.4 Hz, 1H). TOF MS (EIþ): C22H12N2O3, calcd 352.0848, found
352.0839.
5.1.4.7. 3-(4-sec-Butylphenoxy)-8-oxo-8H-acenaphtho[1,2-b]pyrrole-
9-carbonitrile (6). Yield: 0.31 g, 82%. Mp ¼ 210e211 ꢀC. 1H NMR
(400 MHz, CDCl3):
d
8.92 (d, J ¼ 8.0 Hz, 1H), 8.65 (d, J ¼ 8.8 Hz, 1H),
8.46 (d, J ¼ 8.0 Hz,1H), 7.87 (t, J ¼ 8.0 Hz,1H), 7.33 (d, J ¼ 8.4 Hz, 2H),
7.14 (d, J ¼ 8.4 Hz, 2H), 7.04 (d, J ¼ 8.0 Hz, 1H), 2.70 (m, 1H), 1.65 (m,
J ¼ 8.0 Hz, 2H), 1.30 (d, J ¼ 8.0 Hz, 3H), 0.88 (t, J ¼ 8.0 Hz, 3H). 13C
5.1.3.7. 2-[2-Oxo-2H-6-(4-sec-butylphenoxy)acenaphthylen-1-ylidene]-
malononitrile (10g). Yield: 0.20 g, 53%. Mp ¼ 176e177 ꢀC. 1H NMR
NMR (400 MHz, CDCl3):
d 176.17, 163.64, 151.70, 147.53, 136.46,
(400 MHz, CDCl3):
d
8.58 (d, J ¼ 8.0 Hz,1H), 8.56 (d, J ¼ 8.0 Hz,1H), 8.02
134.54, 132.05, 130.91,128.63,128.54, 126.80, 123.89, 122.55, 122.25,
120.84, 120.55, 113.04, 112.87, 112.31, 41.53, 30.77, 21.32, 11.54. TOF
MS (EIþ): C25H18N2O2, calcd 378.1368, found 378.1376.
(d, J ¼ 8.0 Hz, 1H), 7.85 (t, J ¼ 8.0 Hz, 1H), 7.32 (d, J ¼ 8.4 Hz, 2H), 7.14 (d,
J ¼ 8.4 Hz, 2H), 7.00 (d, J ¼ 8.0 Hz, 1H), 2.38 (m, 1H), 1.64 (m, J ¼ 8.0 Hz,
2H), 1.29 (d, J ¼ 8.0 Hz, 3H), 0.88 (t, J ¼ 8.0 Hz, 3H). TOF MS (EIþ):
C25H18N2O2, calcd 378.1368, found 378.1376.
5.2. Biological assay
5.1.4. General procedure for the preparation of 2aec, 3ae3c and 6
To a solution was added the corresponding compounds 10aeg
(1 mmol) and K2CO3 (165 mg, 1.2 mmol) in CH3CN (5 mL) was
5.2.1. Reagents, plasmid and antibodies
A 21-residue Bid BH3 peptide (residues 79e99) [19,20] bearing
a 6-carboxyfluorescein succinimidyl esterfluorescence tag (FAM-Bid)

Z. Zhang et al. / European Journal of Medicinal Chemistry 46 (2011) 3909e3916
3915
was synthesized at HD Biosciences (ShangHai, China). Recombinant
human Bcl-2 protein was purchased from Santa Cruz Biotechnology
(sc-4096, Santa Cruz) [21]. For expression of human Mcl-1 as our
report described [9]. Antibodies were Bcl-2 antibody (sc-783, Santa
Cruz) and rabbit antibody against Mcl-1 (BS1220, Bioworld).
protein bank in the RCSB [23] The 3D structures of the inhibitors
were generated using Chembio3D Ultra 11.0 followed by energy
minimization. AutoDock 4.0 program equipped with ADT was used
to perform the automated molecular docking [24]. Grid maps
covering residues that were perturbed more than the threshold
value of 0.1 ppm in the BH3 binding groove of the proteins were
defined for all inhibitors in the AutoDock calculations using a grid
5.2.2. Cell lines
ꢀ
The human leukemia cell line K562, human breast adenocarci-
noma cell line MCF-7, and human myeloid leukemia cell line HL-60
were purchased from China Center for Type Culture Collection
(Wuhan, China). Cells were cultured in 75 cm2 Falcon flasks (Becton
Dickinson, Franklin Lakes) in phenol-red-free RPMI medium 1640
(Gibco/BRL) supplemented with 10% fetal bovine serum, HEPES
(10 mM), sodium bicarbonate (24 mM), sodium pyruvate (1 mM),
spacing of 0.375 A. The GA-LS algorithm was adopted using default
settings. For each docking job, 100 hybrid GA-LS runs were carried
out. A total of 100 possible binding conformations were generated
ꢀ
and grouped into clusters based on a 1.0 A cluster tolerance. The
docking models were analyzed and represented using ADT.
2-mercaptoethanol (0.05 mM), penicillin (60 mg/mL) and strepto-
Acknowledgments
mycin (100 mg/mL). Cell culture was kept in a humidified incubator
with 5% CO2 at 37 ꢀC.
This work was supported by the Fundamental Research Funds
for the Central Universities and partly supported by the National
Natural Science Foundation of China (Grant 30772622).
5.2.3. Fluorescence polarization-based binding assay
For the competitive binding assay for Bcl-2 protein, FAM-Bid
peptide (10 nM) and Bcl-2 protein (40 nM) were preincubated in
the assay buffer (100 mM potassium phosphate, pH 7.5; 100 mg/mL
References
bovine gamma globulin; 0.02% sodium azide). The detergent Triton
X-100 was freshly prepared daily as a 1% (v/v) stock in 50 mM KPi.
For each experiment, a control containing Bcl-2 and FAM-Bid
peptide (equivalent to 0% inhibition), and another control con-
taining only FAM-Bid, were included on each assay plate. Next,
serial dilutions of compounds with 0.01% Triton X-100 were added.
After 30 min incubation, the polarization values were measured
using the Spectra Max M5 Detection System in a black 96-well
plate. Saturation experiments determined that FAM-Bid binds to
the Bcl-2 protein with a Kd value of 8 nM. For Mcl-1, assays were
performed in the same manner as that for Bcl-2 with the following
exceptions: 50 nM Mcl-1 and 10 nM FAM-Bid peptide were used in
the assay buffer (25 mM Tris, pH 8.0; 150 mM NaCl). FAM-Bid
peptide binds to the Mcl-1 protein with a Kd value of 1.9 nM. The
Ki value for each inhibitor was calculated using the equation we
have developed for FP-based assays [9].
[1] D. Hanahan, R.A. Weinberg, The hallmarks of cancer, Cell 100 (2000) 57e70.
[2] R.W. Johnstone, A.A. Ruefli, S.W. Lowe, Apoptosis: a link between cancer
genetics and chemotherapy, Cell 108 (2002) 153e164.
[3] S.G. Das, J.M. Doshi, D. Tian, S.N. Addo, B. Srinivasan, D.L. Hermanson, C. Xing,
Structureeactivity relationship and molecular mechanisms of ethyl 2-amino-
4-(2-ethoxy-2-oxoethyl)-6-phenyl-4H-chromene-3-carboxylate (sHA 14-1)
and its analogues, J. Med. Chem. 19 (2009) 5937e5949.
[4] M.R. Warr, G.C. Shore, Unique biology of Mcl-1: therapeutic opportunities in
cancer, Curr. Mol. Med. 8 (2008) 138e147.
[5] A.S. Azmi, R.M. Mohammad, Non-peptidic small molecule inhibitors against
Bcl-2 for cancer therapy, J. Cell. Physiol. 1 (2009) 13e21.
[6] M. Konopleva, R. Contractor, T. Tsao, I. Samudio, P.P. Ruvolo, S. Kitada, X. Deng,
D. Zhai, Y.X. Shi, T. Sneed, M. Verhaegen, M. Soengas, V.R. Ruvolo, T. McQueen,
W.D. Schober, J.C. Watt, T. Jiffar, X. Ling, F.C. Marini, D. Harris, M. Dietrich,
Z. Estrov, J. McCubrey, W.S. May, J.C. Reed, M. Andreeff, Mechanisms of
apoptosis sensitivity and resistance to the BH3 mimetic ABT-737 in acute
myeloid leukemia, Cancer Cell 10 (2006) 375e388.
[7] M.H. Kang, Z.S. Wan, Y.H. Kang, R. Sposto, C.P. Reynolds, Mechanism of
synergy of N-(4-hydroxyphenyl) retinamide and ABT-737 in acute lympho-
blastic leukemia cell lines: Mcl-1 inactivation, J. Natl. Cancer. Inst 8 (2008)
580e595.
[8] P.H. Bernardo, T. Sivaraman, K.-F. Wan, J. Xu, J. Krishnamoorthy, C.M. Song,
L. Tian, J.S.F. Chin, D.S.W. Lim, H.Y.K. Mok, V.C. Yu, J.C. Tong, C.L.L. Chai,
Structural insights into the design of small molecule inhibitors that selectively
antagonize Mcl-1, J. Med. Chem. 53 (2010) 2314e2318.
[9] Z. Zhang, T. Song, T. Zhang, G. Wu, J. Gao, L. An, G. Du, A novel BH3 mimetic S1
potently induces Bax/Bak-dependent apoptosis by targeting both Bcl-2 and
Mcl-1, Int, J. Cancer 7 (2011) 1724e1735.
[10] Z. Zhang, G. Wu, F. Xie, T. Song, X. Chang, 3-Thiomorpholin-8-oxo-8H-ace-
naphtho[1,2-b]pyrrole-9-carbonitrile (S1) based molecules as potent, dual
inhibitors of B-cell lymphoma 2 (Bcl-2) and myeloid cell leukemia sequence 1
(Mcl-1): structure-based design and structureeactivity relationship Studies,
J. Med. Chem. 4 (2011) 1101e1105.
5.2.4. Western blot
Cells were lysed in digitonin lysis buffer. (62.5 mMTriseHCl, pH6.8,
2% w/v SDS,10% glycerol, 50 mM DTT,1 mM PMSF) [9]. Electrophoresis
and transfer to membrane were performed routinely. The membranes
were blocked and probed with indicated antibody, followed by
peroxidase-conjugated IgG. The expected proteins were detected by
the ECL method (Amersham Pharmacia Biotech, Pittsburgh, PA).
5.2.5. MTT assay
Standard procedures were used [22]. Tumor cells were plated in
[11] H.G. Rule, S.B. Thompson, Acenaphthenone and acenaphthenequinone,
J. Chem. Soc. (1937) 1761e1763.
[12] A.J. Fatiadi, For an extensive review on the application of malononitrile in the
Knoevenagel reaction, Synthesis 3 (1978) 165e204.
96-well (1 ꢂ104 cells per well) plates, and were cultured in the
presence of various concentrations of compounds (0.01e100 mM)
from DMSO stock (1 mM or 10 mM) for 48 h. The experiments were
performed in triplicates. Control cells were treated with DMSO equal
to the highest percentage of solvent used in the experimental condi-
tions. At the end of the exposure time, the medium was removed and
3-[4,5-dimethylthiazol-2-yl]-2,5-diphentetrazoliumbromide (MTT)
wasaddedtoeachwell. Theplateswere incubatedat 37 ꢀC for 4 h. The
solution was removed and DMSO was added to each well. The IC50
value was obtained from plots of % viability against the dose of the
compound added.
[13] H. Junek, B. Hornischer, H. Sterk, Synthesen mit Nitrilen, 20. Mitt: Untersu-
chungen über den Farbstoff aus Acen-phthenchiaon und Malonitril von,
Monatsh. Chem. 99 (1968) 2121e2125.
[14] Y. Xiao, F. Liu, X. Qian, J. Cui, A new class of long-wavelength fluorophores:
strong red fluorescence, convenient synthesis and easy derivation, Chem.
Commun. (2005) 239e240.
[15] B. Ku, C. Liang, J.U. Jung, B.-H. Oh, Evidence that inhibition of BAX activation by
BCL-2 involves its tight and preferential interaction with the BH3 domain of
BAX, Cell Res. 21 (2011) 627e641.
[16] (a) M. Bruncko, T.K. Oost, B.A. Belli, H. Ding, M.K. Joseph, A. Kunzer,
D. Martineau, W.J. McClellan, M. Mitten, Studies leading to potent, dual
inhibitors of Bcl-2 and Bcl-xL, J. Med. Chem. 4 (2007) 641e662;
(b) A.M. Petros, E.T. Olejniczak, S.W. Fesik, Structural biology of the Bcl-2
family of proteins, Biochim. Biophys. Acta 1644 (2004) 83e94.
[17] P.E. Czabotar, E.F. Lee, M.F. van Delft, C.L. Day, B.J. Smith, D.C.S. Huang,
W.D. Fairlie, M.G. Hinds, P.M. Colman, Structural insights into the degradation
of Mcl-1 induced by BH3 domains, Proc. Natl. Acad. Sci. U.S.A. 15 (2007)
6217e6222.
5.3. Molecular docking
The 3D structures of the human Mcl-1 (hMcl-1; PDB ID: 2NLA)
and human Bcl-2 (Bcl-2; PDB ID: 1GJH) were obtained from the

3916
Z. Zhang et al. / European Journal of Medicinal Chemistry 46 (2011) 3909e3916
[18] Q. Liu, T. Moldoveanu, T. Sprules, E. Matta-Camacho, N. Mansur-Azzam,
K. Gehring, Apoptotic regulation by MCL-1 through hetero-dimerization,
J. Biol. Chem. 285 (25) (2010) 19615e19624.
[19] G. Wang, Z. Nikolovska-Coleska, C.-Y. Yang, R. Wang, G. Tang, J. Guo,
S. Shangary, S. Qiu, W. Gao, D. Yang, J. Meagher, J. Stuckey, K. Krajewski,
S. Jiang, P.P. Roller, H.O. Abaan, Y. Tomita, S. Wang, Structure-based design of
potent small-molecule inhibitors of anti-apoptotic Bcl-2 proteins, J. Med.
Chem. 49 (2006) 6139e6142.
[20] Z. Nikolovska-Coleska, R. Wang, X. Fang, H. Pan, Y. Tomita, P. Li, P.P. Roller,
K. Krajewski, N.G. Saito, J.A. Stuckey, S. Wang, Development and optimization
of a binding assay for the XIAP BIR3 domain using fluorescence polarization,
Anal. Biochem. 332 (2004) 261e273.
[21] G. Hacker, A. Weber, BH3-only proteins trigger cytochrome c release, but
how? Arch. Biochem. Biophys 462 (2007) 150e155.
[22] S.P. Wu, L.Z. Sun, J.K. Willson, L. Humphrey, R. Kerbel, M.G. Brattain, Repres-
sion of autocrine transforming growth factor beta 1 and beta 2 in quiescent
CBS colon carcinoma cells leads to progression of tumorigenic properties, Cell
Growth Differ. 4 (1993) 115e123.
[23] (a) H.M. Berman, J. Westbrook, Z. Feng, G. Gilliland, T.N. Bhat, H. Weissig,
I.N. Shindyalov, P.E. Bourne, The Protein Data Bank, Nucleic Acids Res. 28
(2000) 235e242.(b) H.M. Berman, T. Battistuz, T.N. Bhat, W.F. Bluhm,
P.E. Bourne, K. Burkhardt, Z. Feng, G.L. Gilliland, L. Iype, S. Jain, P. Fagan,
J. Marvin, D. Padilla, V. Ravichandran, B. Schneider, N. Thanki, H. Weissig,
J.D. Westbrook, C. Zardeckia, The Protein Data Bank, Acta Crystallogr., Sect. D:
Biol. Crystallogr. 58 (2002) 899e907.
[24] G.M. Morris, D.S. Goodsell, R.S. Halliday, R. Huey, W.E. Hart, R.K. Belew, A.J. Olson,
Automated docking using a Lamarckian genetic algorithm and an empirical
binding free energy function, J. Comput. Chem. 19 (1998) 1639e1662.