Fragment-Based Design, Synthesis, and Biological Evaluation of 1-Substituted-indole-2-carboxylic Acids as Selective Mcl-1 Inhibitors

Article abstract of DOI:10.1002/ardp.201600251

Based on a known selective Mcl-1 inhibitor, 6-chloro-3-(3-(4-chloro-3,5-dimethylphenoxy)propyl)-1H-indole-2-carboxylic acid, we applied a fragment-based approach to obtain new molecules that extended into the p1 pocket of the BH3 groove and then exhibited binding selectivity for the Mcl-1 over the Bcl-2 protein. After we deconstructed the 1H-indole-2-carboxylic acid from the parental molecule, a benzenesulfonyl was substituted at the 1-position to adopt a geometry preferred for accessing the p1 pocket according to the binding mode of the parental molecule identified by X-ray crystallography. A linear relationship between the free energy of ligand binding (ΔG) and the count of non-hydrogen heavy atoms (HAC) was maintained during the molecular growing to occupy the p1 pocket. Finally, we not only obtained compound 12 with a 7.5-fold selectivity to Mcl-1 (K_i = 0.48 μM by fluorescence polarization) over Bcl-2 (K_i = 3.6 μM), but also provided evidence that additional occupation of the p1 pocket is more favorable for Mcl-1 than for Bcl-2 binding, and contributes more to Mcl-1 inhibition than occupation of the p2 pocket. Compound 12 exhibited a selective killing ability on Mcl-1-dependent cancer cells.

Full text of DOI:10.1002/ardp.201600251

RCHH HARM
A P
Arch. Pharm. Chem. Life Sci. 2016, 349, 1–12
Archiv der Pharmazie
Full Paper
Fragment-Based Design, Synthesis, and Biological Evaluation
of 1-Substituted-indole-2-carboxylic Acids as Selective Mcl-1
Inhibitors
Ziqian Wang1ꢀ, Wenjie Xu1ꢀ, Ting Song2, Zongwei Guo2, Lu Liu1, Yudan Fan2, Anhui Wang1,
and Zhichao Zhang1
1
State Key Laboratory of Fine Chemicals, School of Chemistry, Dalian University of Technology, Dalian,
People’s Republic of China
School of Life Science and Technology, Dalian University of Technology, Dalian, People’s Republic of
2
China
Based on a known selective Mcl-1 inhibitor, 6-chloro-3-(3-(4-chloro-3,5-dimethylphenoxy)propyl)-1H-
indole-2-carboxylic acid, we applied a fragment-based approach to obtain new molecules that
extended into the p1 pocket of the BH3 groove and then exhibited binding selectivity for the Mcl-1
over the Bcl-2 protein. After we deconstructed the 1H-indole-2-carboxylic acid from the parental
molecule, a benzenesulfonyl was substituted at the 1-position to adopt a geometry preferred for
accessing the p1 pocket according to the binding mode of the parental molecule identified by X-ray
crystallography. A linear relationship between the free energy of ligand binding (DG) and the count
of non-hydrogen heavy atoms (HAC) was maintained during the molecular growing to occupy the p1
pocket. Finally, we not only obtained compound 12 with a 7.5-fold selectivity to Mcl-1 (Ki ¼ 0.48 mM
by fluorescence polarization) over Bcl-2 (Ki ¼ 3.6 mM), but also provided evidence that additional
occupation of the p1 pocket is more favorable for Mcl-1 than for Bcl-2 binding, and contributes more
to Mcl-1 inhibition than occupation of the p2 pocket. Compound 12 exhibited a selective killing
ability on Mcl-1-dependent cancer cells.
Keywords: Fragment-based drug design / Ligand efficiency / Mcl-1 inhibitor / p1 pocket / Selective
Received: August 30, 2016; Revised: November 8, 2016; Accepted: November 9, 2016
DOI 10.1002/ardp.201600251
Additional supporting information may be found in the online version of this article at the publisher’s web-site.
:
pro-apoptotic proteins possesses four conserved hydrophobic
residues that insert into four hydrophobic sub-pockets, p1–p4,
Introduction
Mediated by the BH3 domain shared by both anti-apoptotic
(Bcl-2, Bcl-xL, Mcl-1, etc.) and pro-apoptotic (multi-domain
proteins Bax and Bak, and BH3-only proteins) members, the
Bcl-2 family proteins form heterodimers to play a key role in
regulation of apoptotic pathways [1, 2]. The BH3 domain of
in the BH3-binding groove of anti-apoptotic proteins, while
an aspartate makes a salt bridge with a conserved arginine
residue on anti-apoptotic members [3, 4].
The anti-apoptotic Bcl-2 proteins are overexpressed in a
large number of tumors preventing cells from undergoing
apoptosis [5]. They are thus attractive yet challenging targets
for developing new anticancer therapies [1, 6, 7]. The BH3
binding groove of Bcl-2 and Bcl-xL are the primary focus for
the design of Bcl-2 inhibitors [8]. The specific Bcl-2/Bcl-xL
Correspondence: Dr. Zhichao Zhang, State Key Laboratory of Fine
Chemicals, School of Chemistry, Dalian University of Technology,
Dalian 116012, People’s Republic of China.
E-mail: zczhang@dlut.edu.cn
Fax: þ86-411-84986032
ꢀThese authors contributed equally to this work.
ß 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com
1

RCHH HARM
A P
Arch. Pharm. Chem. Life Sci. 2016, 349, 1–12
Z. Wang et al.
Archiv der Pharmazie
inhibitor Navitoclax is currently in phase I/II clinical trials
[9, 10]. However, the cancer killing ability of Navitoclax based
on disruption of the heterodimers between anti-apoptotic
and pro-apoptotic Bcl-2 members is limited to Bcl-2-depen-
dent tumors and it is encountering resistance in Mcl-1
expressing cancer cells [11, 12].
during the molecular growing to progressively occupy p1
pocket. Finally, we not only obtained compound 12 with a
7.5-fold selectivity to Mcl-1 over Bcl-2, but also provided an
evidence that occupation of the p1 pocket is more favorable
for Mcl-1 binding than Bcl-2, and contributes more to Mcl-1
inhibition than occupation of the p2 pocket.
Mcl-1 protein possesses an overall similar but divergent
BH3-binding groove compared with Bcl-2 and Bcl-xL, which
exhibits unique anti-apoptotic functions. Mcl-1 has been
documented as the other arm of the antiapoptotic Bcl-2
family to help cancer evade apoptosis [13]. So far, nine series
of selective Mcl-1 inhibitors that were predicted and/or
identified to bind in the BH3-binding groove of Mcl-1 have
been investigated by Wang, Fesik, Zaneta Nikolovska-
Coleska, our group, and others since 2012 [14, 15]. However,
structural insights into features that contribute to Mcl-1
binding selectivity and affinity are still elusive. Most of these
known selective Mcl-1 inhibitors focused on the optimization
of p2 occupation because it is the deepest and largest sub-
pocket among four sub-pockets in BH3-binding groove of
Mcl-1 and it showed some plastic property that distinguishes
Mcl-1 from the corresponding one on Bcl-2/Bcl-xL. By
contrast, p4 was not a favorable pocket for Mcl-1 binding
selectivity [14]. Of note, neither the reported selective Mcl-1
inhibitors nor Bcl-2/Mcl-1 dual inhibitors explore the p1
pocket although this pocket could be involved in substrate
binding energy improvement.
Most of the known selective Mcl-1 inhibitors or their
parental leads were obtained from HTS (high-throughput
screening) or found by accident. The others were developed
by fragment-based drug design (FBDD), in which a minimal
binding fragment was discovered through NMR-based
screening or deconstruction from a known inhibitor [14].
With the aid of structure-based drug design, the final drug-
like candidates were obtained through fragment growing,
linking, or scaffold merging to markedly increase the affinity
of the starting fragments [16]. FBDD has been rapidly
developed over the past decades and showed significant
advantages over HTS [17, 18]. ABT-737, the parental molecule
of Navitoclax, was also discovered by this strategy [19]. It is
necessary to obtain the structural binding information of the
starting fragment for the further optimization by exploiting
the structure–activity relationship analysis (SAR) observed
during molecular evolution [20, 21]. Additionally, the
conservation of binding mode during molecular growing
could be reflected by a linear relationship between molecular
weight and binding affinity. The prediction map would
provide critical and quantitative assessments to guide the
future drug development [22–24].
Results and discussion
Rationale
According to the mutation studies of Bcl-2 and Mcl-1 in
complex with BH3 peptide, respectively, together with our
previous finding in SAR studies of Bcl-2/Mcl-1 dual inhibitors,
mimicking D67 of Bim BH3 peptide to capture a conserved
R146 or R263 is a widely recognized molecular basis to confer
a major binding energy to a molecule binding either Bcl-2 or
Mcl-1 [25]. Additionally, eight of nine series reported
selective Mcl-1 inhibitors could form a hydrogen bound
with R263 and/or N260 on Mcl-1 protein [14, 15]. We then
decided to maintain the interactions with R263 to contribute
binding energy when we designed new selective Mcl-1
inhibitors.
With respect to the four hydrophobic pocket p1, p2, p3, and
p4, we focused on the p1 pocket because we found significant
features that might be exploited to contribute Mcl-1
selectivity. First, the p1 of Mcl-1 is not as open as that in
Bcl-2/Bcl-xL because the a-helix 3 is well formed in the Mcl-1
complex but poorly formed in the Bcl-2 complex (Fig. 1).
Second, the p1 and p2 pockets are less contiguous in Mcl-1
than that in Bcl-2 when complexed with Bim because M231, a
Here, we chose a fragment from a known molecule that has
been identified by X-ray studies to bind Mcl-1 selectively and
its relatively high ligand efficiency (LE) value could guarantee
the binding mode conservation. With the aid of structure-
based drug design, a series of inhibitors that engaged into the
p1 pocket was obtained through fragment growing. A linear
relationship between DG and HAC was well maintained
Figure 1. Overlay of Mcl-1 from the Mcl-1BLR:hBim BH3
complex (PDB ID: 2PQK, green) and Bcl-2 from Bcl-2:Bim BH3
complex (PDB ID: 4B4S, light blue). M231 on Mcl-1 that is located
between p1 and p2 and the corresponding Y59 of Bcl-2 were
highlighted in orange. Positively charged residues on Mcl-1 that
construct the p1 pocket was marked with dark blue.
ß 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com
2

RCHH HARM
A P
Arch. Pharm. Chem. Life Sci. 2016, 349, 1–12
1-Substituted-indole-2-carboxylic Acids as Selective Mcl-1 Inhibitors
Archiv der Pharmazie
residue located between p1 and p2, is more solvent exposed
on Mcl-1 than the corresponding Y59 on Bcl-2. Third, the
surface of Mcl-1 is more electropositive than that of other Bcl-
2 proteins. In particular, as showed in Fig. 1, a number of
positively charged residues are present on a-helix 3, the a-
helix 3 to a-helix 4 loop, and a-helix 4 in Mcl-1, which
construct the p1 pocket [26]. These differences in p1 pocket
may contribute to a specific binding properties of Mcl-1.
Additionally, p1 is much narrower than p2, which may provide
more potency for additional substituents to expand the
occupation. The known nine series of selective Mcl-1
inhibitors and all the pan-Bcl-2 inhibitors do not thus far
explore the p1 pocket.
Here, we choose 1-methyl-indole-2-carboxylic acid (com-
pound 1, compound 13 in original literature) as the starting
fragment [27]. 1 exhibited a Ki value of 160 mM to bind Mcl-1
and its binding position in BH3-binding groove of Mcl-1 has
been identified by X-ray studies of its analogs, showing that
the carboxylic acids of 1 interact with R263 of Mcl-1 and the
indole core could stably locate at the p2 pocket during the
following molecular merge. Further, suitable substituents at
1-position of 1H-indole-2-carboxylic acid was designed to
access the p1 pocket (Supporting Information Fig. S1).
of sodium 4-hydroxybenzenesulfonate with different substi-
tuted benzyl bromides [31]. Reaction of sodium salt with
SOCl2 in DMF afforded sulfonyl chloride which was coupled
with methyl indole-2-carboxylate in DMF to give the
corresponding methyl esters of 7–12. Alkaline hydrolysis of
these esters yielded carboxylic acids 7–12 [32].
Structure-based design, synthesis, and active
evaluation
Based on the binding mode of compound 1 and our aims, a
benzenesulfonyl was substituted at 1-position because a
sulfonamide moiety could adopt an appropriate geometry for
engaging into the p1 pocket as it may be oriented in parallel
with the groove. Chemical accessiblility was also taken into
account. 1-(Phenylsulfony)-1H-indole-2-carboxylic acid (2) was
then obtained.
The binding affinity (Ki value) of 2 was evaluated using
fluorescence polarization assays (FPA), which measured its
ability to competitively displace a Bid-derived peptide from
Mcl-1/Bcl-2. R-(–)-Gossypol was used as a positive control. It
was found that 2 binds to Mcl-1 with a threefold affinity
improvement than compound 1 (Ki ¼ 50 mM vs. 160 mM)
toward Mcl-1. A Ki value of 80 mM for 2 binding to Bcl-2
protein was also found (Table 1).
Chemistry
Compound 2 was prepared by benzenesulfonyl chloride and
methyl indole-2-carboxylate according to literature proce-
dures [28]. Then, analogs 3–6 were synthesized in which the
4-position of phenylsulfonyl moiety was replaced with –Ph,
–O–Ph, –t-Amyl, and –O–i-Bu, respectively. These analogs
were prepared by a four-step synthesis, as shown in Scheme 1.
4-substituted benzenesulfonic acids were prepared by com-
mercially available substituted benzenes and ClSO3H [29].
They were then converted to the corresponding sulfonyl
chlorides by SOCl2 [30]. The amidation reaction with methyl
indole-2-carboxylate and subsequent ester hydrolysis with
LiOH yielded final compounds 3–6.
Next, a heteronuclear single quantum coherence (HSQC)
NMR spectroscopy experiment using 15N labeled Mcl-1 protein
validated our design strategy. Figure 2a shows a plot of the
chemical shift perturbations against the overall Mcl-1 protein
residues. Approximately 65% of the residues that were
perturbed above the threshold value (0.06 ppm) were located
in the p2 pocket. Compared with the residues affected by
compound 1 in the previous report, the residues of R263, L267,
F270, A271, A227, and F228 exhibited a similar chemical shift
upon the addition of compound 2. Moreover, M231, L232
L246, V249, and V253 are newly emergent residues with
significant chemical shift (DCS > 0.07 ppm), and these residues
are closer to the p1 pocket than the above-mentioned
residues. The NMR-derived models (Fig. 2b) demonstrate that
the indole core keeps the original binding mode of the
Compounds 7–12 were synthesized by a different four-step
synthesis, as shown in Scheme 2. The O-substituted sodium
4-hydroxybenzenesulfonates were prepared by benzylation
Scheme 1. Synthesis of compounds 3–6.
Reagents and conditions: (a) ClSO3H, dry
CHCl3, ice-water bath, nitrogen atmosphere
protection, 1.5 h; (b) SOCl2, two drops DMF,
80°C, 2 h; (c) NaH, dry DMF, room tempera-
ture, 1 h; (d) LiOH, MeOH/THF/water, 70°C,
5 h; HCl, pH ¼ 2.
ß 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com
3

RCHH HARM
A P
Arch. Pharm. Chem. Life Sci. 2016, 349, 1–12
Z. Wang et al.
Archiv der Pharmazie
Scheme 2. Synthesis of compounds 7–12. Reagents and conditions: (a) substituted benzyl bromide, 1 N NaOH, isopropanol, 75°C,
20 h; (b) SOCl2, dry DMF, 5 min; ice water, 5 min; (c) NaH, dry DMF, room temperature, 1 h; (d) LiOH, MeOH/THF/water, 70°C, 5 h; HCl,
pH ¼ 2.
starting fragment, and benzenesulfonyl orients toward the p1
pocket. We proposed that elaboration at para position of
benzenesulfonyl with hydrophobic substitutions could probe
the p1 pocket and then achieve further gains in affinity and
selectivity.
even found for compound 10 with larger hindrance of
3,5-dimethyl-substitution, most likely due to the wider p1
pocket of Bcl-2 than that of Mcl-1 as showed in Fig. 1. To
improve the selectivity, we then substituted difluoro and
chlorine with small size to yield compounds 11 and 12.
Additionally, we suspected that these polar moieties might
interact with the more positive Mcl-1 surface better than
Bcl-2. As expected, compound 12 showed a 7.5-fold selectivity
to bind Mcl-1 (Ki ¼ 0.48 mM) over Bcl-2 (Ki ¼ 3.6 mM). To test if
the binding site of compound 12 maintained the original
fragment in the p2 and extended into p1 as we suspected,
1H-15N HSQC NMR spectroscopy studies were performed. The
chemical shift changes in the presence of 12 were mapped and
plotted against Mcl-1 residues (Fig. 3a). Chemical shift
perturbations of F228, L232, L246, V249, V253, L267, F270,
and R263 which located at the p2 pockets were also observed
as caused by compound 2. Moreover, L232, L235, D236, V243,
K244, S245 located on a-helix 3 and 4 that construct p1 are
newly emergent residues with significant chemical shift
perturbations (DCS > 0.065 ppm) caused by compound 12,
indicating that 12 did extend deeply into the p1 pocket. I237,
K238, N239, and D242 that located on an unstructured loop
connecting a-helix 3 and a-helix 4 and surrounded p1 also
showed significant chemical shift perturbations (DCS > 0.06
ppm). NMR-derived structure of 12 bound to Mcl-1 is
shown in Fig. 3b. We then overlay the compound 12 and
a-helix in Bim. As shown in Fig. 3c, 12 mimics D67 to maintain
the interaction with R263 as its starting fragment compound 1
Compound 3, with a further elongation of benzenesul-
fonyl, showed a similar fivefold affinity improvement toward
both Mcl-1 and Bcl-2 compared with 2 (Ki ¼ 9.8 and 19.3 mM,
respectively). A twofold selectivity between Bcl-2 and Mcl-1
was observed for compound 3. Encouraged by the binding
improvement of 3, and to further improve the selectivity
between Mcl-1 and Bcl-2, we tried to increase the Mcl-1
affinity by fitting the steric hindrance of the specific M231 in
the p1 of Mcl-1. Compounds 4, 5, 6, and 7 with oxygen atom as
a linker or alkyl substituents of various sizes at the para
position of the benzenesulfonyl were prepared to obtain
proper length and flexibility for substituent to adapt to p1.
From this set of compounds, a trend emerged showing much
improved Mcl-1 affinity than Bcl-2 that compound 7 exhibited
a 6.5-fold selectivity for binding to Mcl-1 (Ki ¼ 1.3 mM) over
Bcl-2 (Ki ¼ 8.3 mM).
Consequently, analogs with different substituents at o-, m-,
p-position of benzyloxyl were prepared to further probe the
p1 pocket, yielding compounds 8–12 (Table 2). According to
FPA, although compounds 8 and 9 with 2- or 3-methyl
substituted showed improvement in affinity, they increased
equally against Bcl-2 and Mcl-1 compared with compound 7.
An obvious loss of selectivity between the two proteins was
ß 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com
4

RCHH HARM
A P
Arch. Pharm. Chem. Life Sci. 2016, 349, 1–12
1-Substituted-indole-2-carboxylic Acids as Selective Mcl-1 Inhibitors
Archiv der Pharmazie
Table 1. Fragment 1 and its derivatives: Name, structure, and binding affinity determined by fluorescence polarization
assays (FPA).
Ki (mM)
Compounds
Structure
Mcl-1
Bcl-2
R-(ꢁ)-Gossypol
0.19 ꢂ 0.01
0.32 ꢂ 0.01
1
160
ND
2
3
50 ꢂ 5.70
9.8 ꢂ 0.73
80 ꢂ 7.38
19.3 ꢂ 1.21
4
5
6
6.3 ꢂ 0.43
4.8 ꢂ 0.40
5.9 ꢂ 0.44
17.2 ꢂ 0.97
11.6 ꢂ 0.88
14.5 ꢂ 0.62
7
1.3 ꢂ 0.25
8.3 ꢂ 0.45
does, and it mimics L62 and I58 hotspots simultaneously. It
could explore the p1 as deep as the I58 of Bim could reach.
To further characterize the binding affinities of the
compounds to Mcl-1 and Bcl-2, we applied an enzyme-linked
immunosorbent assay (ELISA) for compounds 7, 10, and 12. As
shown in Table 3, compounds 7 and 10 exhibited 6.2-fold and
10.8-fold selectivity for binding to Mcl-1 over Bcl-2, while
compound 8 exhibited only 4.3-fold selectivity. These data
were consistent with the FP results.
between DG and HAC was found for compounds 2–12,
indicating the conservation of the binding mode upon
elaboration of the starting fragment. The parental 1H-
indole-2-carboxylic acid core constantly remained interaction
with R263 in Mcl-1 and the 1-position substituent access
an additional p1 pocket. Moreover, a different slope was
found for the growing path toward different targets. Each
additional heavy atom contribute approximately 0.27 kcal
molꢁ1 energy to Mcl-1, while a minor one was found for Bcl-2
which is 0.17 kcal molꢁ1. It illustrated that the occupation
of p1 pocket is more favorable for binding Mcl-1 than Bcl-2.
We then plotted the molecular weight (MW) against the pKi
(ꢁlogKi) to further quantify this differences (Fig. 4b, Support-
ing Information Table S1). The trend line with a value for
the slope of 61, implied by the compounds’ data, along with
the path of ideal optimization, showed that an increase of 1
pKi unit can be expected for every 61 mass units added to the
To our knowledge, compounds 2–12 are the first series of
I58 mimicking inhibitors that developed by fragment-based
strategy to extend a starting fragment into the p1 pocket.
Thus, this is a promising series to study the contribution of
p1 pocket to Mcl-1 binding. We then plotted the DG of
these inhibitors against the HAC to monitor the efficiency
during fragment growing. As shown in Fig. 4a (Supporting
Information Table S1),
a
linear (R2 ¼ 0.9) relationship
ß 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com
5

RCHH HARM
A P
Arch. Pharm. Chem. Life Sci. 2016, 349, 1–12
Z. Wang et al.
Archiv der Pharmazie
Figure 2. (a) Chemical shift perturbations of Mcl-1 residues bound to compound 2. (b) Mcl-1 residues shown to be perturbed in HSQC
NMR in the presence of 2 (green).
compound. For Bcl-2 protein, the value for the slope increased
to about 87.
occupation is an effective way to increase Mcl-1 binding
affinity of designed compounds.
Furthermore, we compared our optimization process to
other seven known Mcl-1 selective inhibitors, all of which are
optimized for p2 occupation (Supporting Information
Table S2). The present compounds in this study showed more
binding efficiency than other five series (Fig. 5a). We analyzed
the MW/pKi curve of molecules previously (Fig. 5b). The
average value of the slope of the trend line is around 85. That
is, for those molecules optimized toward p2 occupation, an
increase of 1 pKi unit can be expected for every 85 mass units
added to the compound. Therefore, optimization for p1
Compound 12 selectively induces apoptosis in
Mcl-1-dependent cancer cells
To confirm the specificity of 7, 10, 12 for Mcl-1 protein in
cellular models, we analyzed its activity in four cell lines (Fig. 6
and Supporting Information Fig. S2). NCI-H23 cells depend on
Mcl-1 for survival. In contrast, HL-60 cells are Bcl-2-dependent
cells and H22 and MCF-7 cells are dependent on both Mcl-1
and Bcl-2. Figure 6a showed Western blot analysis of the
ß 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com
6

RCHH HARM
A P
Arch. Pharm. Chem. Life Sci. 2016, 349, 1–12
1-Substituted-indole-2-carboxylic Acids as Selective Mcl-1 Inhibitors
Archiv der Pharmazie
Table 2. Compounds 8–12: Name, structure, and binding affinity by FPA (Ki, mM).
Ki (mM)
Compounds
Structure
Mcl-1
Bcl-2
R-(ꢁ)-Gossypol
0.19 ꢂ 0.01
0.32 ꢂ 0.01
8
0.73 ꢂ 0.12
5.2 ꢂ 0.17
9
0.75 ꢂ 0.08
0.65 ꢂ 0.04
5.3 ꢂ 0.23
2.8 ꢂ 0.03
10
11
12
0.64 ꢂ 0.03
0.48 ꢂ 0.01
4.7 ꢂ 0.03
3.6 ꢂ 0.02
expression levels of Bcl-2 and Mcl-1 in these cell lines. These
cell lines were treated with different concentrations of 12,
and then apoptosis was determined by Annexin V flow
cytometry. After 48 h treatment, 12 induced apoptosis in NCI-
H345 cells (IC50 ¼ 2.2 mM), which was 30-fold greater than that
in HL-60 cells (IC50 ¼ 63.3 mM), consistent with its binding
selectivity for Mcl-1 over Bcl-2. The activity of 12 in H22
(IC50 ¼ 9.2 mM) cells and MCF-7 cells (IC50 ¼ 19.2 mM) was not as
sensitive as that in NCI-H23 cells (Fig. 6b). Furthermore,
compound 10, with less selectivity for Mcl-1 over Bcl-2,
induced potent apoptosis in both NCI-H345 (IC50 ¼ 8.7 mM)
and HL-60 cells (IC50 ¼ 23.1 mM), while the less potent
compound 7 failed to sensitively induce apoptosis in both
of the cell lines (Supporting Information Fig. S2).
and MW/pKi of these molecules against the two proteins, we
found that the p1 occupation exhibits a higher efficiency for
binding Mcl-1 than occupation of p2, and it is more favorable
for Mcl-1 inhibition than that for Bcl-2 inhibition. Our
evidence illustrated that the p1 could be a key binding site
for selective Mcl-1 inhibitor design.
Experimental
Chemistry
General
All commercial reagents were purchased and used without
further purification or distillation unless otherwise stated.
1H NMR spectra was recorded on
a Bruker AV-500
spectrometer using DMSO as solvent. Chemical shifts were
reported in d values (ppm) with tetramethylsilane as the
internal reference, and J values were reported in hertz (Hz).
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
200–300 mesh.
Conclusion
Following a fragment-based strategy and structure-based
design, we not only obtained compound 12 that exhibits
7.5-fold binding selectivity toward Mcl-1 (Ki ¼ 0.48 mM by FP)
over Bcl-2 (Ki ¼ 3.6 mM), but also obtained a series of small
molecule tools to explore the p1 pocket of BH3-binding
groove on both Mcl-1 and Bcl-2. Through analysis of DG/HAC
ß 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com
7

RCHH HARM
A P
Arch. Pharm. Chem. Life Sci. 2016, 349, 1–12
Z. Wang et al.
Archiv der Pharmazie
Figure 3. (a) Plot of chemical shift changes of Mcl-1 amide upon addition of 12 (Mcl-1:12 ¼ 1:2) as a function of Mcl-1 residue
numbers. (b) Mcl-1 residues shown to be perturbed in HSQC NMR in the presence of 12 (green). Significant shift (DCS > 0.06 ppm)
including R263, M231 in p2, and the residues L235, V243, K238, and N239 in the p1 pocket are highlighted in dark gray. (c) Side view
of 12 bound to Mcl-1 overlaid with the bound Bim peptide.
Table 3. Competitive binding of 7, 10, and 12 to Mcl-1 and Bcl-2 protein determined by ELISA assay (IC50, mM).
IC50 (mM)
Compounds
Mcl-1
Bcl-2
R-(ꢁ)-Gossypol
0.35 ꢂ 0.14
3.0 ꢂ 0.3
0.72 ꢂ 0.07
18.6 ꢂ 1.1
5.2 ꢂ 0.5
7
10
12
1.2 ꢂ 0.32
0.75 ꢂ 0.18
9.6 ꢂ 0.8
ß 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com
8

RCHH HARM
A P
Arch. Pharm. Chem. Life Sci. 2016, 349, 1–12
1-Substituted-indole-2-carboxylic Acids as Selective Mcl-1 Inhibitors
Archiv der Pharmazie
Figure 4. (a) DG (in kcal molꢁ1) of compounds 2–12 plotted as a function of HAC in the compounds toward Mcl-1 and Bcl-2,
respectively. (b) Linear relationship between MW and pKi of compounds 2–12 toward Mcl-1 and Bcl-2, respectively.
The InChI codes of the investigated compounds together with
some biological activity data are provided as Supporting
Information.
Synthesis of (1,1’-biphenyl)-4-sulfonic acid. Biphenyl (3.08 g,
20 mmol) was dissolved in chloroform (150 mL). Chlorosul-
fonic acid (1.31 mL, 20 mmol) in 25 mL chloroform was added
dropwise over 30 min under an ice-water bath and a nitrogen
atmosphere protection. With the addition of chlorosulfonic
acid, a white solid precipitated gradually. The reaction
mixture was stirred for further 1 h, then the precipitate was
collected by filtration and washed with cold chloroform.
Yield: 2.8 g, 60%.
Synthesis of (1,1’-biphenyl)-4-sulfonyl chloride. (1,1’-Biphe-
nyl)-4-sulfonic acid (2.8 g, 12 mmol) was dissolved in dichloro
sulfoxide (25 mL) and treated with a catalytic amount of N,N-
dimethylformamide (two drops). The mixture was heated at
80°C for 2 h. After the completion of the reaction, the mixture
was concentrated in vacuo. The residual thionyl chloride was
1-(Phenylsulfonyl)-1H-indole-2-carboxylic acid (2)
Yield: white solid, 460 mg, 85%. 1H NMR (500 MHz, DMSO) d:
11.74 (s, 1H), 8.02 (d, J ¼ 12.0 Hz, 2H), 7.67 (m, J ¼ 7.5 Hz, 4H),
7.49–7.41 (m, 1H), 7.36–7.30 (m, 2H), 7.23 (t, J ¼ 7.5 Hz, 1H),
7.06 (dd, J ¼ 13.9, 6.0 Hz, 1H). TOF MS (EIþ): C15H11NO4S, found
301.04.
General procedure for the preparation of compounds 3, 4,
5, and 6
The general procedure is illustrated immediately below with
compound 3 as a specific example.
Figure 5. (a) DG (in kcal molꢁ1) for series of A–G plotted as a function of HAC in the compounds toward Mcl-1 protein. (b) Linear
relationship between MW and pKi of series of A–G toward Mcl-1 protein.
ß 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com
9

RCHH HARM
A P
Arch. Pharm. Chem. Life Sci. 2016, 349, 1–12
Z. Wang et al.
Archiv der Pharmazie
Figure 6. (a) The levels of Mcl-1 and Bcl-2 protein in NCI-H23, H22, MCF-7, and HL-60 cells were examined by Western blot. (b) Cells
were treated with graded concentration of 12 for 48 h, and the percentage of apoptotic cells was determined by Annexin V flow
cytometry.
removed by toluene (15 mL ꢃ 3) under the condition of
vacuum. Yield: 2.4 g, 80%.
1-((4-Phenoxyphenyl)sulfonyl)-1H-indole-2-carboxylic acid
(4)
Synthesis of methyl 1-([1,1’-biphenyl]-4-ylsulfonyl)-1H-in-
dole-2-carboxylate. The NaH (60%, 632 mg, 15.8 mmol) was
added in three portions to a solution of methyl 1H-indole-2-
carboxylate (1.1 g, 6.3 mmol) in dry N,N-dimethylformamide
(10 mL) under stirring at room temperature for 1 h, then (1,1’-
biphenyl)-4-sulfonyl chloride (2.4 g, 9.5 mmol) was added to
this mixture. After 1 h, the solution was poured into 20 mL
cooled water. The resulting precipitate was filtered and
purified by silica gel chromatography (ethyl acetate/petro-
leum ether ¼ 1:6) to generate 1.6 g of the product as the white
powder solid. Yield: 65%.
Synthesis of 1-((1,1’-biphenyl)-4-ylsulfonyl)-1H-indole-2-
carboxylic acid (3). To a solution of methyl 1-(phenyl-
sulfonyl)-1H-indole-2-carboxylate (1.6 g, 4.1 mmol) in tetra-
hydrofuran/methanol/water (5:5:1, 22 mL) was added lithium
hydroxide monohydrate (688 mg, 28.7 mmol). The reaction
mixture was heated at 70°C for 5 h. After cooling to room
temperature, it was concentrated, followed by addition of
water (10 mL). The pH was adjusted to 2 with 1 N HCl and
extracted with ethyl acetate (15 mL ꢃ 3). The combined
extraction was dried over anhydrous sodium sulfate, filtered,
and concentrated. The crude product was purified by re-
crystallization in a mixture of dichloromethane and hexanes
to afford compound 3 as white solid.
Yield: white solid, 2.04 g, 26%. 1H NMR (500MHz, DMSO) d:
11.73 (s, 1H), 8.08 (d, J ¼ 9.0 Hz, 2H), 8.00 (d, J ¼ 8.5 Hz, 1H), 7.68
(d, J ¼ 8.0 Hz, 1H), 7.45 (q, J ¼ 9.0 Hz, 3H), 7.21–7.37 (m, 3H),
7.09–7.15 (m, 4H). TOF MS (EIþ): C21H15NO5S, found 393.07.
1-((4-(tert-Pentyl)phenyl)sulfonyl)-1H-indole-2-carboxylic
acid (5)
Yield: white solid, 2.37 g, 32%. 1H NMR (500 MHz, DMSO) d:
11.73 (s, 1H), 8.00 (t, J ¼ 8.5 Hz, 3H), 7.67 (d, J ¼ 7.5 Hz, 3H),
7.41–7.52 (m, 1H), 7.24–7.35 (m, 2H), 1.59 (q, J ¼ 7.5 Hz, 2H),
1.22 (s, 6H), 0.54 (t, J ¼ 7.5 Hz, 3H). TOF MS (EIþ): C20H21NO4S,
found 371.12.
1-((4-Isobutoxyphenyl)sulfonyl)-1H-indole-2-carboxylic
acid (6)
Yield: white solid, 1.72 g, 24%. 1H NMR (500 MHz, DMSO) d:
11.94 (s, 1H), 7.93 (d, J ¼ 9.0 Hz, 2H), 7.82 (d, J ¼ 8.0 Hz, 2H),
7.42 (s, 1H), 7.34–7.40 (m, 1H), 7.12 (d, J ¼ 7.8 Hz, 2H), 6.86–
6.93 (m, 1H), 4.69 (m, 1H), 1.29 (s, 6H). TOF MS (EIþ):
C
18H17NO5S, found 359.08.
General procedure for the preparation of compounds 7–12
The general procedure is illustrated immediately below with
compound 7 as a specific example.
Synthesis of sodium 4-(benzyloxy)benzenesulfonate. To a
suspension of sodium 4-hydroxybenzenesulfonate (1.96 g,
10 mmol), benzyl bromide (1.9 mL, 16 mmol), and NaOH
(0.44 g, 11 mmol in 10 mL water) in isopropanol (25 mL). The
mixture was stirred at 75°C for 20 h. After cooled to
room temperature, the white solids that crystallized out of
solution were filtered, washed with cold isopropanol. Yield:
1.9 g, 68%.
1-((1,1’-Biphenyl)-4-ylsulfonyl)-1H-indole-2-carboxylic acid
(3)
Yield: white solid, 2.11 g, 28%. 1H NMR (500 MHz, DMSO) d
11.74 (s, 1H), 8.09 (dd, J ¼ 24.7, 8.5 Hz, 2H), 7.92 (d, J ¼ 8.6 Hz,
1H), 7.73–7.59 (m, 3H), 7.53–7.41 (m, 4H), 7.38–7.31 (m, 1H),
7.23 (t, J ¼ 7.5 Hz, 1H), 7.06 (dd, J ¼ 13.4, 5.5 Hz, 2H). TOF MS
(EIþ): C21H15NO4S, found 377.07.
ß 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com
10

RCHH HARM
A P
Arch. Pharm. Chem. Life Sci. 2016, 349, 1–12
1-Substituted-indole-2-carboxylic Acids as Selective Mcl-1 Inhibitors
Archiv der Pharmazie
Synthesis of 4-(benzyloxy)benzenesulfonyl chloride. To a
stirred suspension of sodium 4-(benzyloxy)benzenesulfonate
(1.9 g, 6.8 mmol) in dry N,N-dimethylformamide (10 mL),
dichloro sulfoxide (638 mL, 8.8 mmol) was added dropwise
at room temperature. After the reaction mixture was stirred
for 5 min, poured onto ice, stirred further for 5 min and the
precipitated white solid was filtered. The product was pure
enough for next step without further purification. Yield:
1.7 g, 88%.
7.17–7.28 (m, 3H), 7.03–7.10 (m, 1H), 5.18 (s, 2H). TOF MS (EIþ):
C
22H16ClNO5S, found 441.04.
1H-15N HSQC
1H-15N HSQC spectra were recorded for 0.1 mM 15N-labeled
Mcl-1. Inhibitor solution was added at a 2:1 ratio. The
1H-15N HSQC spectra were recorded on a Bruker Advance
DRX600 MHz spectrometer, processed by NMRPipe and
visualized by NmrViewJ 8.0. Chemical shift changes were
2
2 1/2
The amidation reaction and subsequent ester hydrolysis
used same methods in compound 3.
calculated as [(D1H ppm) þ (0.2 ꢃ D15N ppm)
]
as a
function of the residue number.
Molecular docking
1-((4-(Benzyloxy)phenyl)sulfonyl)-1H-indole-2-carboxylic
acid (7)
Molecular modeling was performedwith AutoDock 4.0. The3D
structures of the human Mcl-1 (hMcl-1; PDB ID: 2PQK) was
obtained from the Protein Data Bank. The 3D structures of the
inhibitors were generated using Chembio3D Ultra 11.0
followed by energy minimization. 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 spacing of
Yield: white solid, 1.63 g, 40%. 1H NMR (500MHz, DMSO) d:
11.94(s, 1H), 8.08(d, J ¼ 9.0 Hz, 2H), 7.98(d, J ¼ 8.5 Hz,1H), 7.65–
7.78 (m, 5H), 7.58 (s, 1H), 7.45 (d, J ¼ 8.0 Hz, 2H), 7.27–7.38 (m,
3H), 5.28 (s, 2H). TOF MS (EIþ): C22H17NO5S, found 407.08.
1-((4-((2-Methylbenzyl)oxy)phenyl)sulfonyl)-1H-indole-2-
carboxylic acid (8)
Yield: white solid, 1.47 g, 35%. 1H NMR (500 MHz, DMSO) d:
11.91 (s, 1H), 8.04 (d, J ¼ 8.0 Hz, 1H), 7.96 (d, J ¼ 9.0 Hz, 2H),
7.68 (d, J ¼ 8.0 Hz, 2H), 7.42–7.52 (m, 2H), 7.31–7.39 (m, 2H),
7.15–7.29 (m, 4H), 5.16 (s, 2H), 2.28 (s, 3H). TOF MS (EIþ):
0.375A. 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
̊
C
23H19NO5S, found 421.10.
tolerance. The docking models were analyzed and represented
using PyMol molecular graphics system.
1-((4-((3-Methylbenzyl)oxy)phenyl)sulfonyl)-1H-indole-2-
carboxylic acid (9)
Pharmacology
Yield: white solid, 1.26 g, 30%. 1H NMR (500 MHz, DMSO) d:
11.92 (s, 1H), 8.06 (d, J ¼ 8.0 Hz, 1H), 7.98 (d, J ¼ 9.0 Hz, 2H),
7.43 (dt, J ¼ 14.8, 6.5 Hz 4H), 7.19–7.28 (m, 4H), 7.08–7.12 (m,
2H), 5.16 (s, 2H), 2.30 (s, 3H). TOF MS (EIþ): C23H19NO5S, found
421.10.
Fluorescence polarization-based binding assay (FPA)
For the competitive binding assay for Mcl-1 protein, FAM-
Bid peptide (10 nM) and Mcl-1 protein (55 nM) were
preincubated in the assay buffer (25 mM Tris, pH 8.0;
150 mM NaCl). Next, serial dilutions of compounds were
added. After a 1-min incubation, the polarization values
were measured using the Spectra Max M5 Detection System
in a black 96-well plate. Free peptide controls (fluorescent
peptide only) and bound peptide controls (fluorescent
peptide in the presence of Mcl-1) were included on each
assay plate. The percentage of inhibition was derived from the
equation: % inhibition ¼ 100ꢃ (1ꢁ(mP ꢁ mPf)/(mPb ꢁ mPf)) in
which mPf is the free peptide control and mPb is the
bound peptide control. The computer program for calculating
Ki values for FPA-based assays is available free of charge at:
http://sw16.im.med.umich.edu/software/calc Ki/.
1-((4-((3,5-Dimethylbenzyl)oxy)phenyl)sulfonyl)-1H-
indole-2-carboxylic acid (10)
Yield: white solid, 1.61 g, 37%. 1H NMR (400 MHz, DMSO) d:
11.92 (s, 1H), 8.03 (d, J ¼ 8.4 Hz, 1H), 7.94 (d, J ¼ 9.0 Hz, 2H),
7.68 (d, J ¼ 7.8 Hz, 1H), 7.49 (t, J ¼ 7.8 Hz, 1H), 7.35 (dd, J ¼ 16.0,
8.4 Hz, 2H), 7.19 (d, J ¼ 9.0 Hz, 2H), 6.99 (d, J ¼ 7.4 Hz, 3H), 5.07
(s, 2H), 2.25 (s, 6H). TOF MS (EIþ): C24H21NO5S, found 435.11.
1-((4-((3,4-Difluorobenzyl)oxy)phenyl)sulfonyl)-1H-
indole-2-carboxylic acid (11)
Yield: white solid, 1.77 g, 40%. 1H NMR (400 MHz, DMSO) d:
13.58 (s, 1H), 8.00 (d, J ¼ 9.0 Hz, 3H), 7.66 (d, J ¼ 7.7 Hz, 1H),
7.50–7.58 (m, 1H), 7.40–7.49 (m, 2H), 7.31 (dd, J ¼ 14.1, 6.9 Hz,
3H), 7.21 (d, J ¼ 9.0 Hz, 2H), 5.15 (s, 2H). TOF MS (EIþ):
Western blot
After cells were lysed in NP-40 lysis buffer and centrifuged, the
supernatant was subjected to Western blot by using specific
antibody. Electrophoresis and transfer to membrane were
performed routinely. Protein signals were detected by the ECL
method (Amersham Pharmacia Biotech).
C
22H15F2NO5S, found 443.06.
1-((4-((3-chlorobenzyl)oxy)phenyl)sulfonyl)-1H-indole-2-
carboxylic acid (12)
Yield: white solid, 1.76 g, 40%. 1H NMR (500 MHz, DMSO) d:
11.71 (s, 1H), 7.99 (d, J ¼ 7.5 Hz, 2H), 7.65 (t, J ¼ 8.9 Hz, 1H), 7.51
(s, 1H), 7.43 (dt, J ¼ 14.8, 6.2 Hz 4H), 7.31 (t, J ¼ 7.5 Hz, 1H),
Apoptosis assay
Phosphatidylserine (PS) exposure was quantified by surface
Annexin V-FITC staining. Cells were seeded in each well of 6-
ß 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com
11

RCHH HARM
A P
Arch. Pharm. Chem. Life Sci. 2016, 349, 1–12
Z. Wang et al.
Archiv der Pharmazie
well tissue culture plates and treated with compounds for 48 h
after 24 h. Cells were washed twice with PBS and incubated
with a 1:40 solution of FITC-conjugated Annexin V for 10 min
at room temperature. Stained cells were analyzed by flow
cytometry.
[15] K. Doi, R. Li, S. Sung, H. Wu, Y. Liu, W. Manieri,
G. Krishnegowda, A. Awwad, A. Dewey, X. Liu, S. Amin,
C. Cheng, Y. Qin, E. Schonbrunn, S. Sebti, H. Wang,
J. Biol. Chem. 2012, 287, 10224–10235.
[16] M. N. Schulz, R. E. Hubbard, Curr. Opin. Pharmacol. 2009,
9, 615–621.
This work was supported by the National Natural Science
Foundation of China (21372036, 81570129, 81430083,
21502015, and 21402022).
[17] M. Orita, K. Ohno, T. Niimi, Drug Discov. Today 2009, 14,
321–328.
[18] H. Chen, X. Zhou, A. Wang, Y. Zheng, Y. Gao, J. Zhou,
Drug Discov. Today 2015, 20, 105–113.
The authors have declared no conflict of interest.
[19] T. Oltersdorf, S. W. Elmore, A. R. Shoemaker,
R. C. Armstrong, D. J. Augeri, B. A. Belli, M. Bruncko,
T. L. Deckwerth, J. Dinges, K. J. Tomaselli, B. Wang,
M. D. Wendt, H. Zhang, S. W. Fesik, S. H. Rosenberg,
Nature 2005, 435, 677–681.
References
[1] R. M. Brady, A. Vom, M. J. Roy, B. J. Smith, R. M. Moss,
E. Hatzis, D. C. S. Huang, J. P. Parisot, I. P. Street,
G. Lessen, J. Med. Chem. 2014, 57, 1323–1343.
[2] S. Cory, D. C. S. Huang, J. M. Adams, Oncogene 2003, 22,
8590–8607.
[20] S. B. Shuker, P. J. Hajdilk, R. P. Meadows, S. W. Fesik,
Science 1996, 274, 1531–1534.
[21] D. A. Erlanson, R. S. McDowell, T. O’Brien, J. Med. Chem.
2004, 47, 3463–3482.
[22] S. Schultes, C. Graaf, E. E. J. Haaksma, I. J. P. Esch,
R. Leurs, O. Kramer, Drug Discov. Today 2010, 7,
157–162.
[3] M. Vogler, D. Dinsdale, M. J. S. Dyer, G. M. Cohen,
Cell Death Differ. 2009, 16, 360–367.
[4] E. F. Lee, P. E. Czabotar, B. J. Smith, K. Zobel, M. Colman,
W. D. Fairlie, Cell Death Differ. 2007, 14, 1711–1713.
[5] B. E. Sleebs, W. J. A. Kersten, S. Kulasegaram,
G. Nikolakopoulos, E. Hatzis, R. M. Moss, J. P. Parisot,
H. Yang, L. Chen, G. Lessene, J. Med. Chem. 2013, 56,
5514–5540.
[6] P. H. Bernardo, T. Sivaraman, K. Wan, J. Xu,
J. Krishnamoorthy, J. S. F. Chin, D. S. W. Lim, C. L.
L. Chai, J. Med. Chem. 2010, 53, 2314–2318.
[7] G. Lessene, P. E. Czabotar, P. M. Colman, Nat. Rev. Drug
Discov. 2008, 7, 989–1000.
[23] R. A. E. Carr, M. Congreve, C. W. Murray, D. C. Rees, Drug
Discov. Today 2005, 10, 987–992.
[24] P. J. Hajduk, J. Med. Chem. 2006, 49, 6972–6976.
[25] E. F. Lee, P. E. Czabotar, M. F. van Delft, E. M. Michalak,
M. J. Boyle, S. N. Willis, H. Puthalakath, P. Bouillet,
P. M. Colman, D. C. S. Huang, W. D. Fairlie, J. Cell Biol.
2008, 180, 341–355.
[26] 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, PNAS 2007, 104, 6217–6222.
[27] A. Friberg, D. Vigil, B. Zhao, R. N. Daniels, J. P. Burke,
P. M. Garcia-Barrantes, D. Camper, B. A. Chauder, T. Lee,
E. T. Olejniczak, S. W. Fesik, J. Med. Chem. 2013, 56,
15–30.
[8] M. R. Arkin, J. A. Wells, Nat. Rev. Drug Discov. 2004, 3,
301–317.
[9] C. Tse, A. R. Shoemaker, J. Adickes, M. G. Anderson,
J. Chen, S. Jin, E. F. Johnson, K. C. Marsh, M. J. Mitten,
S. W. Elmore, Cancer Res. 2008, 68, 3421–3428.
[10] W. H. Wilson, O. A. O’Connor, M. S. Czuczman,
A. S. LaCasce, J. F. Gerecitano, J. P. Leonard, A. Tulpule,
K.Dunleavy,H.Xiong, R.A.Humerickhouse, LancetOncol.
2010, 11, 1149–1159.
[11] M. F. van Delft, A. H. Wei, K. D. Mason, C. J. Vandenberg,
L. Chen, P. E. Czabotar, S. N. Willis, C. L. Scott, C. L. Day,
S. Cory, D. C. S. Huang, Cancer Cell 2006, 10, 389–399.
[12] D. Yecies, N. E. Carlson, J. Deng, A. Letai, Blood 2010,
115, 3304–3313.
[28] J. Basset, M. Romero, T. Serra, M. D. Pujol, Tetrahedron
2012, 68, 356–362.
[29] R. G. Schult, J. Org. Chem. 1961, 26, 5195–5196.
[30] P. M. O’Brien, D. F. Ortwine, A. G. Pavlovsky,
J. A. Picard, D. R. Sliskovic, B. D. Roth, R. D. Dyer,
L. L. Johnson, C. F. Man, H. Hallak, J. Med. Chem. 2000,
43, 156–166.
[31] E. Nuti, S. Santamaria, F. Casalini, K. Yamamoto,
L. Marinelli, V. L. Pietra, E. Novellino, E. Orlandini,
S. Nencetti, A. M. Marini, S. Salerno, S. Taliani,
F. D. Settimo, H. Nagase, A. Rossello, Eur. J. Med. Chem.
2013, 62, 379–394.
[13] M. R. Warr, G. C. Shore, Curr. Mol. Med. 2008, 8, 138–147.
[14] L. Chen, M. E. Lanning, S. Fletcher, Austin J. Anal. Pharm.
Chem. 2014, 1, 1015–1020.
€
[32] M. I. Attia, M. Timmermann, P. Hogger, C. Herdeis,
Eur. J. Org. Chem. 2007, 22, 3669–3675.
ß 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com
12