Another structural correction for 1-oxo-1H-phenalene-2,3-dicarbonitriles: Synthesis of a potent BCL-2 inhibiting 7-phenoxy derivative

Article abstract of DOI:10.1002/ardp.202100151

Aromatic scaffolds are an important part of biologically active compounds and molecular probes used to study biochemical pathways and the involved targeted proteins of interest. 1-Oxo-1H-phenalene-2,3-dicarbonitrile-based compounds have been described as

Full text of DOI:10.1002/ardp.202100151

Received: 19 April 2021
Revised: 3 June 2021
Accepted: 7 June 2021
|
|
DOI: 10.1002/ardp.202100151
F U L L P A P E R
Another structural correction for 1‐oxo‐1H‐phenalene‐2,
3
7
‐dicarbonitriles: Synthesis of a potent BCL‐2 inhibiting
‐phenoxy derivative
1
1
2
3
Izidor Sosič | Martina Gobec | Christian Steinebach
| Martin Schlesinger
|
3
2
Gerd Bendas | Michael Gütschow
1
Faculty of Pharmacy, University of Ljubljana,
Ljubljana, Slovenia
Abstract
2
Department of Pharmaceutical & Medicinal
Aromatic scaffolds are an important part of biologically active compounds and
molecular probes used to study biochemical pathways and the involved targeted
proteins of interest. 1‐Oxo‐1H‐phenalene‐2,3‐dicarbonitrile‐based compounds have
been described as inhibitors of the BCL‐2 family of proteins, and this core structure
represents numerous possibilities for modifications that could lead to improved
inhibitory potencies. Many studies demonstrated intriguing characteristics of these
compounds in terms of reactivity and, interestingly, some contradictory literature
reports appeared about reaction outcomes to synthesize them. Here, we initially
provide a condensed overview of transformations performed on the phenalene
scaffold, followed by the resynthesis of a 6‐phenoxy‐substituted derivative. We
show that the initial determination of this particular structure was wrong and
provide two‐dimensional nuclear magnetic resonance (NMR) evidence to assign the
structure properly. When preparing new derivatives using the same synthetic route,
we observed 6‐ and 7‐substituted regioisomers. After confirming their structures by
NMR experiments, the ability of these compounds to inhibit BCL‐2 was evaluated.
The most potent 1‐oxo‐1H‐phenalene‐2,3‐dicarbonitrile derivatives inhibited BCL‐2
in the nanomolar range and showed double‐digit micromolar cytotoxicity against
four different cancer cell lines.
Chemistry, Pharmaceutical Institute,
University of Bonn, Bonn, Germany
3
Department of Pharmaceutical & Cell
Biological Chemistry, Pharmaceutical
Institute, University of Bonn, Bonn, Germany
Correspondence
Izidor Sosič, Faculty of Pharmacy, University
of Ljubljana, Aškerčeva cesta 7, SI‐1000
Ljubljana, Slovenia.
Email: izidor.sosic@ffa.uni-lj.si
Michael Gütschow, Department of
Pharmaceutical & Medicinal Chemistry,
Pharmaceutical Institute, University of Bonn,
An der Immenburg 4, D‐53121 Bonn,
Germany.
Email: guetschow@uni-bonn.de
Funding information
Slovenian Research Agency,
Grant/Award Number: J1‐2485 and core
funding P1‐0208
K E Y W O R D S
2D NMR spectroscopy, BCL‐2, phenalenones, ring expansion
1
|
INTRODUCTION
availability and manifold possibilities for systematic structural mod-
ifications, including those toward polycyclic aromatic compounds with
[
1–5]
Aromatic rings are frequently used structural components in drugs
and represent a common part of different medicinal chemistry ap-
proaches. They offer an easy and well‐characterized synthetic
electrophilic properties.
In 2005, a series of acenaphthylene‐based
molecules were synthesized, and the core structure of the most
exciting compound was reported as 8‐oxo‐8H‐acenaphtho[1,2‐b]
This is an open access article under the terms of the Creative Commons Attribution‐NonCommercial‐NoDerivs License, which permits use and distribution in any
medium, provided the original work is properly cited, the use is non‐commercial and no modifications or adaptations are made.
©
2021 The Authors. Archiv der Pharmazie published by Wiley‐VCH Verlag GmbH on behalf of Deutsche Pharmazeutische Gesellschaft
Arch. Pharm. 2021;e2100151.
wileyonlinelibrary.com/journal/ardp
|
1 of 13
https://doi.org/10.1002/ardp.202100151

2
of 13
SOSIČ ET AL.
|
H
such S
electivities.
with n‐butylamine (4.0 equiv, 3 h) resulted in the formation of 6‐, 9‐,
N
Ar products were performed to confirm different regios-
[
23,24]
It was revealed that the reaction of phenalene B
[
24]
and 3‐substituted derivatives (Figure 2a).
In contrast, a reaction of
B with three different primary amines (1.1 equiv, 12 h) mainly gave
6
‐substituted products with trace amounts of 9‐substituted deriva-
[23]
tives.
Secondary amines (1.1 equiv, 2 h) produced 6‐substituted
major products (30–49% yield) accompanied by 9‐ and 6,9‐
[
23]
disubstituted derivatives (Figure 2a).
FIGURE 1 The misassigned structure A and the corrected
compound B. The numbering of both core scaffolds is shown. The
colored numbers show the reactive carbon atoms, with red denoting
the most reactive one
H
The S
N
Ar reactions with thiol‐based reagents resulted in a
single 6‐substituted regioisomer (Figure 2b) with yields in a si-
milar range as for 6‐amino‐substituted compounds. Usually, for
H
aromatic thiols, 4 equivalents were needed, and the S
N
Ar pro-
[
16,18,19,28]
ceeded in MeCN at room temperature in 2–48 h.
To
[
6]
pyrrole‐9‐carbonitrile (compound A, Figure 1). This highly electron‐
react aliphatic thiols with phenalene B, MeOH was applied as a
[
13]
deficient scaffold was deemed as an attractive precursor for the
solvent. Chen et al.
synthesized different 6‐thiol‐substituted
[6,7]
preparation of useful fluorophores,
fluorescent markers for tumor
derivatives using 4.0 equivalents of the appropriate
[
8,9]
cells,
cysteine.
bioactive properties, such as cytotoxicity against tumor cells,
and fluorescence probes for imaging cysteine and homo-
S‐nucleophile at 0–5°C in an overnight reaction (Figure 2b). For
[
10]
Moreover, derivatives of A were shown to have different
the reaction with mercaptopropionic acid in MeOH, opposing
[10]
[7,11]
DNA
fibroblast growth factor re-
and inhibition of the B‐cell lymphoma 2 (BCL‐2)
outcomes were described. Zhang et al.
noted that for a suc-
[12]
intercalation and apoptosis induction,
cessful reaction only 2.0 equivalents of the nucleophile were
[23]
[13]
ceptor 1 inhibition,
required (2 h, 46% yield), whereas Li et al.
reported only 9%
[14–20]
family of proteins.
conversion in refluxing MeOH for 48 h despite using 5.0
equivalents of mercaptopropionic acid.
Despite numerous studies describing physicochemical and
biological properties of 8‐oxo‐8H‐acenaphtho[1,2‐b]pyrrole‐9‐
carbonitrile (A) derivatives, it was only in 2013/2014 when the ori-
Another contrasting observation in the chemistry of these
compounds was the outcome of reacting primary amines for
1–3 h in MeCN with phenalenones that possessed thiol‐ and
secondary amine‐based substituents at position 6. Some studies
reported substitutions (3.0 equiv amine) to occur at position 6
[
21,22]
ginal authors published two separate corrigendum papers,
which the core structure A was revised. They used two‐dimensional
2D) nuclear magnetic resonance (NMR) techniques to reassign the
molecule as 1‐oxo‐1H‐phenalene‐2,3‐dicarbonitrile (scaffold B,
in
(
[
7,23]
[18,20,28]
(Figure 2c),
whereas others
reported Michael addi-
[
23]
Figure 1). Moreover, the authors
and, independently, Lenk
provided mechanistic explanations for the formation of
‐oxo‐1H‐phenalene‐2,3‐dicarbonitrile (B) from starting ace-
naphthenequinone, as well as an unambiguous structure confirma-
tions (10 equiv amine) to give 3‐substituted derivatives in up to
35% yield (Figure 2d) without the nucleophilic substitution at
position 6.
[
24]
et al.
1
A different route was applied for the 6‐phenoxy‐substituted
phenalenones (Figure 2e). These were synthesized in high overall
yields (approx. 70%) by initial nucleophilic aromatic substitution
of 5‐bromoacenaphthylene‐1,2‐dione, followed by sequential
Knoevenagel condensation with malononitrile and base‐
[
24]
tion by X‐ray diffraction.
The class of phenalenones is accessible through various pre-
parative methods, some of which have been discovered re-
[
25–27]
cently.
The electronic characteristics of the aromatic system B
[
16–18]
allowed access to many derivatives thereof, which were primarily
mediated cyclization.
H
obtained via nucleophilic substitution of aromatic hydrogens (S
N
Ar )
We became aware of compounds with the 1‐oxo‐1H‐
phenalene‐2,3‐dicarbonitrile scaffold because of their ability to
potently inhibit members of the BCL‐2 family of proteins. Based
on our interest in BCL‐2 as an anticancer target and the in‐depth
overview of synthetic access toward such compounds, our initial
motivation was to resynthesize one of the previously described
[
24]
at positions C‐6 and C‐9 by N‐ and S‐nucleophiles.
In Figure 2, a
condensed overview of reactions performed on this planar and highly
electron‐deficient polycyclic system is represented, giving the reader
an impression of the phenalene scaffold's reactivity.
The majority of reactions on the phenalenone scaffold were
performed with primary or cyclic secondary amines at room tem-
perature, almost exclusively with MeCN as solvent. Usually, an amine
excess of 1.0–4.0 equivalents was used and the reaction times
spanned from 1 to 24 h, yielding compounds in up to 50% yield
[
18]
pan‐BCL‐2 inhibitors.
Considering the generally low yields
reported for 6‐amino‐ and 6‐thio‐substituted derivatives by
H
S
N
Ar reactions and the contrasting data on the products ob-
tained, we decided to initially prepare a BCL‐2 inhibitor with the
6‐phenoxy‐phenalene core. As the next step, it was intended to
employ the same synthetic route toward novel 6‐substituted
phenalene derivatives and evaluate their potential to inhibit
BCL‐2.
[
7]
(
Figure 2a). In the initial paper, primary amines (4.0 equiv, 1 h) gave
‐substituted products predominantly, whereas cyclic secondary
amines (4.0 equiv, 24 h) resulted in 6‐substituted, 9‐substituted, and
,9‐disubstituted products. Later on, detailed structural analyses of
6
6

SOSIČ ET AL.
3 of 13
|
(a)
(b)
(c)
(d)
(e)
FIGURE 2 Schematic overview of phenalene scaffold reactivity. Please note that not all of these structures were officially corrected.
For the sake of clarity, however, all reactions and positions of substitution are shown on the revised core structure B despite initially described
on the misassigned scaffold A. (a) Reaction with amines; (b) reaction with thiols; (c) nucleophilic substitution at position 6; (d) reactions at
position 3; (e) reactions with phenols

4
of 13
SOSIČ ET AL.
|
SCHEME 1 Resynthesis of the previously published 6‐(4‐isopropylphenoxy)‐1‐oxo‐1H‐phenalene‐2,3‐dicarbonitrile (3ʹ). Reagents and
conditions: (a) Br
2
, reflux; (b) 4‐isopropylphenol, K
2
3
CO , dimethylformamide, 75°C, 24 h; (c) CH
2
(CN)
2
, 10 mol% K
2
3
CO , wet CH
3
CN, reflux, 1 h
[
24]
1
2
2
|
RESULTS AND DISCUSSION
Chemistry
malononitrile and catalytic amounts of K
2
CO
3
.
Its H NMR
spectrum corresponded entirely to the previously described data,
.1
|
which presumed the formation of 6‐substituted phenalene (structure
[
16–18]
3ʹ in square brackets, Scheme 1).
However, when performing
1
13
1
1
To prepare the 6‐(4‐isopropylphenoxy)‐substituted phenalene 3ʹ
the complete assignment of H and C chemical shifts with H‐ H
[
16–18]
1
13
(
Scheme 1), literature procedures were used.
Bromination of
homonuclear correlation spectroscopy (COSY), H‐
C
hetero-
the starting acenaphthenequinone yielded the 5‐bromo derivative 1,
nuclear single‐quantum correlation (HSQC) spectroscopy, nuclear
1
13
which was subjected to nucleophilic aromatic substitution with
Overhauser effect spectroscopy (NOESY), and H‐ C heteronuclear
multiple‐bond correlation (HMBC) spectroscopy, we could demon-
strate that this product was the 7‐(4‐isopropylphenoxy)‐substituted
4
‐isopropylphenol to obtain the diaryl ether 2. The final product
was obtained in a one‐pot, two‐step sequence by reacting 2 with
1
13
FIGURE 3
H‐ C heteronuclear multiple‐bond correlation spectrum (CDCl
3
) of compound 3. Only the most significant cross‐peaks that
correlate protons and carbons, which are connected through three covalent bonds, are noted. Circled cross‐peaks in blue indicate correlations
of H‐3ʹ, H‐5ʹ (7.38 ppm) to CH(CH ) (33.84 ppm) and to C1ʹ (151.62 ppm), whereas the red circle indicates a key correlation of H‐9 (8.63 ppm)
3 2
to C1 (176.33 ppm)

SOSIČ ET AL.
5 of 13
|
phenalene 3, the regioisomer of the presumed compound 3ʹ
substitution pattern, which was further supported by correlation
peaks of both H‐9 (8.44 ppm) and H‐6 (8.69 ppm) to C7 (127.29 ppm)
(Figure 4). For phenalenone 5, a key three‐bond correlation between
H‐9 (8.28 ppm) and C1 (185.30 ppm) suggested the 6‐position of the
Br atom (Figure 5).
(
Scheme 1). Our attempts to additionally obtain compound 3ʹ failed.
1
1
We assigned the structure by first performing a H‐ H COSY
experiment (Figure S1) to distinguish between ortho‐coupled aro-
matic protons (H‐8 and H‐9, doublets, J = 8.6 Hz) and protons at
1
13
positions 4, 5, and 6. Next, we used the H‐ C HSQC experiment
When we attempted to prepare the 6‐benzyl(methyl)amino‐
substituted derivative (Scheme 3), the same phenomenon occurred
as we again observed two distinct products. Several rounds of
chromatographic purifications eventually led to their successful iso-
lation and characterization. 2D NMR data showed that these were
indeed the 7‐ and 6‐substituted regioisomers 7 and 8, respectively
(Scheme 3).
(
Figure S2) to determine which hydrogen atoms are connected to
1
1
which carbons. In the H‐ H NOESY data (Figure S3), the through‐
space correlation between the signal at 7.37 and 3.00 ppm indicated
the spatial proximity of the aromatic H‐3ʹ and H‐5ʹ and the CH
1
13
proton of the isopropyl group at 4ʹ. Based on H‐ C HSQC and
1
1
H‐ H NOESY, carbons of the phenoxy ring C2ʹ, C6ʹ and C3ʹ, C5ʹ
1
1
could be appropriately assigned. Other cross‐peaks in the H‐ H
After assigning the most evident protons and carbons of com-
1
1
1
13
NOESY spectrum showing the proximity of aromatic protons cor-
pound 7 by analyzing H‐ H COSY (Figure S8), H‐ C HSQC
1
1
1
1
roborated data from the H‐ H COSY experiment.
(Figure S9), and H‐ H NOESY (Figures S10a and S10b) spectra, the
The crucial correlations were, however, demonstrated in the
remaining carbons and the substitution pattern were determined
1
13
1
13
H‐ C HMBC spectrum. Namely, a three‐bond correlation of H‐9
with the H‐ C HMBC experiment (Figure 6). Again, three‐bond
couplings between H‐9 (8.41 ppm) and C1 (173.96 ppm), H‐9
(8.41 ppm) and C7 (159.65 ppm), as well as H‐6 (8.69 ppm) and C7
(159.65 ppm) indicated the 7‐substituted phenalene ring (Figure 6).
(
8.63 ppm) to C1 (176.33 ppm) evidenced the 4‐isopropylphenoxy
moiety at position 7 of the phenalene ring (Figure 3, red circle). The
long‐range coupling correlations of H‐9 (8.62 ppm) to C7
1
(
163.83 ppm) as well as H‐6 (8.90 ppm) to C7 (163.82 ppm) ad-
The phenyl carbons were unambiguously assigned via three‐bond H,
13
ditionally confirmed the suggested substitution pattern (Figure 3). As
C coupling correlation (blue circles in Figure 6) of aliphatic CH
2
a result of simultaneous correlations of H‐4 (8.45 ppm) to C3
protons (4.98 ppm) to C2ʹ, C6ʹ (127.13 ppm) and H‐3ʹ, H‐5ʹ
(7.40 ppm) to C1ʹ (136.13 ppm). The signals of other quaternary
phenalene carbons were resolved by analyzing simultaneous cou-
plings of H‐4 (8.35 ppm) to C3 (128.11 ppm) and C6 (135.35 ppm),
1
(
131.06 ppm), C3a (128.64 ppm), and C6 (132.19 ppm), other car-
bons assignations could be assigned with this experiment (Figure 3).
Correlation peaks of H‐3ʹ and H‐5ʹ (7.38 ppm) to CH(CH
33.84 ppm) and to C1ʹ (151.62 ppm) (Figure 3, blue circles) enabled
3 2
)
1
(
H‐9 (8.41 ppm) to C3a (129.09; ppm), and H‐8 (7.39 ppm) to C9a
further proof of the correct assignment of carbons and protons of
the phenoxy ring.
(119.36 ppm) and C6a (122.05 ppm) (Figure 6; the latter two cor-
relations are not shown with arrows on the structure).
We next subjected 5‐bromo derivative 1 to the same reaction
conditions that would enable access to a bromo‐substituted phena-
lene (Scheme 2). After consumption of the starting material, two
distinct products were visible by thin‐layer chromatography (TLC) of
the crude reaction mixture. Despite significant difficulties, we sepa-
rated and purified both products by several consecutive runs of
chromatographic purification. Interestingly, 2D NMR analyses
showed that regioisomers, that is, 7‐bromo‐ (4; Figures 4, S4, and S5)
and 6‐bromo‐substituted (5; Figures 5, S6, and S7) phenalenones
were formed (Scheme 2).
The position of the N‐benzylmethylamine moiety in compound 8
was deciphered similarly. After determining which phenalene pro-
1
13
tons are ortho‐coupled, we searched for the key H‐ C HMBC
correlation of a single aromatic proton with C1. From Figure 7, it can
be observed that the proton at position 9 (8.66 ppm) is the only one
that exhibited a notable three‐bond correlation with a C1 peak at
1
1
176.90 ppm. Other 2D spectra of compound 8, that is, H‐ H COSY
1
13
and H‐ C HSQC, are shown in Figures S11 and S12, respectively.
It has to be emphasized that we encountered numerous diffi-
culties with the purification of all final compounds, a phenomenon
[
24]
Evidently, the ring expansion of 1 can occur in two ways de-
pending on which carbonyl group reacts with malononitrile in the
initial Knoevenagel condensation. However, following a postulated
mechanism, an intermediate tetracyclic 2,3‐dihydroxycyclopropane‐
previously also observed by others.
This was probably due to the
intrinsic reactivity of the phenalene scaffold and due to the struc-
tural similarity of regioisomeric products, which made purifications
1
,1‐dicarbonitrile is formed whose deprotonation determines the
[
24]
position of the carbonyl group in the final phenalenone.
the regiopreference of the reaction is hardly predictable.
Hence,
To determine the structures of both regioisomers, the same
step‐by‐step combination of 2D NMR experiments as for compound
3
was employed, whereby key correlations to indirectly suggest the
1
13
position of the Br atom were obtained in the H‐ C HMBC spectra.
The most evident three‐bond correlations for 7‐bromo‐substituted
phenalene 4 are depicted in Figure 4. In particular, a correlation
of H‐9 (8.44 ppm) to C1 (177.20 ppm) indicated the 7‐bromo
SCHEME 2 Synthesis of 7‐ and 6‐bromo‐substituted phenalenes
4
and 5. Reagents and conditions: (a) CH
2 2 2 3
(CN) , 10 mol% K CO ,
MeCN, reflux, 1 h

6
of 13
SOSIČ ET AL.
|
1
13
FIGURE 4
H‐ C HMBC spectrum (DMSO‐d
6
) of compound 4. Only the most significant cross‐peaks that correlate protons and carbons,
which are connected through three covalent bonds, are noted. Circled cross‐peaks in red indicate correlations of H‐9 (8.44 ppm) to C1
177.20 ppm) and C7 (127.29 ppm), as well as of H‐6 (8.69 ppm) to C7 (127.29 ppm). Additional three‐bond correlation peaks noted (blue
arrows next to the structure) are H‐4 (8.46 ppm) to C3 (131.35 ppm), H‐5 (8.06 ppm) to C3a (122.82 ppm), and C6a (130.63 ppm), H‐6
(
1
(
8.69 ppm) to C4 (135.11 ppm), H‐8 (8.37 ppm) to C9a (127.03 ppm) and C6a (130.64 ppm), as well as H‐9 (8.44 ppm) to C3a (133.78 ppm).
DMSO, dimethyl sulfoxide; HMBC, heteronuclear multiple‐bond correlation
of both compounds challenging. For instance, to obtain compounds
pure enough for structural characterization and biochemical assays
pro‐apoptotic proteins, such as BIM, can bind. Our interest in BCL‐2
inhibition is also based on inhibitor of apoptosis (IAP)‐addressing
[
31]
(
see below), at least two rounds of silica gel chromatographic pur-
PROTACs with potential antitumor effects.
ifications were necessary. Of note, we tried to synthesize
The synthesized compounds 3, 4, 5, 7, and 8 were evaluated for
their inhibition of BCL‐2 using an enzyme‐linked immunosorbent
assay (ELISA) (Table 1), in which the ability of compounds to com-
petitively displace a BIM‐derived peptide from BCL‐2 was measured.
Briefly (see Section 4 for details), the biotinylated BIM peptide was
attached to a streptavidin‐coated well plate and mixtures of different
inhibitor concentrations and His‐tagged BCL‐2 protein were added,
followed by applying the anti‐His secondary antibody conjugated to
horseradish peroxidase and a colorimetric readout upon addition of
o‐phenylenediamine. The data were calculated as the residual ac-
tivities (RAs) of BCL‐2 in the presence of 50 µM of each compound.
Compounds that showed more than 50% inhibition of the BCL‐2
were subjected to dose‐dependent inhibitory activity measurements.
The binding curves for phenalenones 3, 4, 7, and 8 are shown in
Figure S13.
H
6
1
4
‐substituted derivatives 3ʹ and 8 also by a direct S
N
Ar reaction of
‐oxo‐1H‐phenalene‐2,3‐dicarbonitrile (B, Figure 1) with either
‐isopropylphenol or N‐benzylmethylamine. However, a set of two
different reaction conditions, both based on previous literature
procedures, did not yield the desired products (Scheme S1).
2.2
|
Biochemistry
The commitment of cells to apoptosis, an ancient cell suicide pro-
gram, is controlled by the BCL‐2 family of proteins, which contains
both pro‐apoptotic and pro‐survival members. Cytotoxic stimuli can
activate pro‐apoptotic BH3‐only proteins and some of them initiate
apoptosis signaling by binding to pro‐survival BCL‐2 members. BCL‐2
itself, for example, decreases the propensity of cells for apoptosis,
confers cancer cells' resistance to therapeutic agents, and represents
The 7‐phenoxy‐substituted compound 3 exhibited slightly dif-
ferent BCL‐2 inhibition (IC50 = 110 ± 30 nM) in comparison to pre-
[
29,30]
[18]
an important target for anticancer therapy.
BCL‐2 forms a
viously published data,
where the IC50 was determined to be
well‐defined hydrophobic surface binding groove, into which
739 nM. All four new phenalenones proved to be less potent BCL‐2

SOSIČ ET AL.
7 of 13
|
1
13
FIGURE 5 A crucial segment of the H‐ C HMBC spectrum (DMSO‐d
6
) of 5. A three‐bond correlation between H‐9 (8.28 ppm) and C1
(
185.30 ppm) is shown. DMSO, dimethyl sulfoxide; HMBC, heteronuclear multiple‐bond correlation
5‐diphenyltetrazolium bromide (MTT) assay. Both BCL‐2 inhibitors
exhibited double‐digit micromolar cytotoxicity against all cancer cell
lines (Table 1). The discrepancy between in vitro BCL‐2 inhibition
and cytotoxicity can be attributed to many factors, such as poor
cellular permeability of compounds or the fact that the selected cell
lines are less dependent on BCL‐2, as indicated by CRISPR‐based
[
32]
knockout screens from the DepMap database (see Figure S14).
SCHEME 3 Synthesis of 7‐ and 6‐benzyl(methyl)amino‐
substituted derivatives 7 and 8. Reagents and conditions: (a)
N‐benzylmethylamine, K
b) CH (CN) , 10 mol% K
2
CO
3
, dimethylformamide, 75°C, 24 h;
3
, MeCN, reflux, 1 h
3
|
CONCLUSION
(
2
2
2
CO
In this study, we demonstrated the importance of using a combina-
tion of 2D NMR techniques to determine substituent patterns on the
phenalene ring properly. The usefulness of this approach was clearly
presented by correcting the structure of a previously misassigned 4‐
isopropylphenol‐substituted phenalene 3ʹ. Despite showing this on a
limited number of examples, our findings suggest that the one‐pot
Knoevenagel condensation and base‐mediated ring expansion per-
formed on 5‐substituted‐acenaphthylene‐1,2‐diones can generate
two possible regioisomers (Schemes 2 and 3). If the acenaphthene-
inhibitors in comparison to 3. Clearly, the replacement of the
‐aryloxy group by the 7‐benzylamino was disadvantageous (3 vs. 7).
7
Owing to the low number of new derivatives it was impossible to
draw reliable conclusions about structure–activity relationships
(
SAR). Given that 7‐substituted compounds 3, 4, and 7 inhibited BCL‐
to a different extent, a significantly greater number of compounds
would be needed to gain a better understanding of SAR for
‐substituted compounds. Nevertheless, we assessed the cytotoxi-
2
7
quinone is first subjected to cyclization to yield the phenalene ring B,
H
city of the two most potent compounds 3 and 4 against four cancer
cells (U87, human primary glioblastoma cell line; MV3, melanoma cell
line; A2780, human ovarian cancer cell line; MDA‐MB‐231, human
breast cancer cell line) by using 3‐(4,5‐dimethylthiazole‐2‐yl)‐2,
followed by the S
N
Ar reactions, the final products obtained are
mostly 6‐substituted phenalenones (Figures 2a and 2b).
We are aware that the previously studied 6‐arylthio derivatives
performed even better as BCL‐2 inhibitors than related phenoxy

8
of 13
SOSIČ ET AL.
|
1
13
FIGURE 6
H‐ C HMBC spectrum (DMSO‐d
6
) of compound 7. Only the most significant cross‐peaks that correlate protons and carbons,
which are connected through three covalent bonds, are noted. Circled cross‐peaks in blue indicate correlations of aliphatic CH
(
2
protons
4.98 ppm) to C2ʹ, C6ʹ (127.13 ppm) and H‐3ʹ, H‐5ʹ (7.40 ppm) to C1ʹ (136.13 ppm). The red‐circled cross‐peaks indicate a key correlation of H‐9
8.41 ppm) to C1 (173.96 ppm) and a three‐bond coupling between aliphatic CH protons (4.98 ppm) and C7 (159.65 ppm). DMSO, dimethyl
(
2
sulfoxide; HMBC, heteronuclear multiple‐bond correlation
[
16,18]
compounds.
However, the low‐yielding reactions to-
and the components were visualized under UV light and/or through
staining with the relevant reagent. Preparative normal‐phase flash
column chromatography was performed on Merck Silica Gel 60
(particle size, 0.040–0.063 mm; Merck). For reversed‐phase column
chromatography Isolera Biotage One Flash Chromatography system
(Biotage) and SNAP Biotage KP‐C18‐HS column (30 g) was used with
a gradient of 0.1% trifluoroacetic acid (TFA) in deionized water and
MeCN as eluent (gradient 0% MeCN for 2 column volumes; 10–90%
MeCN in 15 column volumes; 90% MeCN for 5 column volumes).
Melting points were determined on a Reichelt hot‐stage appara-
ward phenalene derivatives, their tedious purifications, and poor
[
33]
aqueous solubility
precluded our further work with these com-
pounds despite encouraging results in terms of BCL‐2 inhibition and
despite knowing that several possibilities to explore the SAR still
exist. Nevertheless, we believe that the findings presented in this
report further clarify the structural issues of this compound class and
thus provide additional guidelines for proper structure determina-
tion of substituted phenalene derivatives.
1
13
tus and are uncorrected. H and C NMR spectra were recorded at
4
4
4
|
EXPERIMENTAL
Chemistry
295 K on a Bruker Avance 500 MHz spectrometer (Bruker) or Bruker
1
Avance III 400 MHz spectrometer operating at frequencies for
H
13
.1
.1.1
|
NMR at 500 or 400 MHz, and for C NMR at 125 or 101 MHz. The
chemical shifts (δ) are reported in parts per million (ppm) and are
referenced to the deuterated solvent used. The coupling constants (J)
are given in Hz, and the splitting patterns are designated as follows:
s, singlet; d, doublet; app d, apparent doublet; dd, double doublet; t,
|
General
Reagents and solvents were obtained from commercial sources
Acros Organics, Sigma‐Aldrich, TCI Europe, Alfa Aesar, Fluorochem)
13
(
triplet; m, multiplet. All C NMR spectra were proton decoupled.
and were used as received. For reactions involving air‐ or moisture‐
sensitive reagents, solvents were distilled before use, and these re-
actions were carried out under an argon atmosphere. Reactions were
monitored using analytical TLC plates (Merck 60 F254, 0.20 mm),
Resonance assignments were made on the basis of one‐ and two‐
1
13
1
1
dimensional NMR techniques, which include H, C, H‐ H COSY,
1
13
1
13
1
1
H‐ C HSQC, H‐ C HMBC, and H‐ H NOESY experiments.
Standard parameter sets (COSYGPSW, NOESYPHSW, HSQCETGP,

SOSIČ ET AL.
9 of 13
|
1
13
FIGURE 7 A crucial segment of H‐ C HMBC spectrum (DMSO‐d ) of compound 8. A three‐bond correlation between H‐9 (8.28 ppm) and
6
C1 (185.30 ppm) is shown. DMSO, dimethyl sulfoxide; HMBC, heteronuclear multiple‐bond correlation
TABLE 1 Inhibitory potencies of
Cytotoxicity
compounds 3–5, 7, and 8 toward BCL‐2
MDA‐
MB‐231
a
determined using an ELISA assay and
U87
MV3
A2780
IC50 (µM) BCL‐2 or RA (%)
at 50 µM
cytotoxicity of compounds 3 and 4 against
Compound
IC50 (µM) IC50 (µM) IC50 (µM) IC50 (µM)
b
four cancer cell lines
3
4
5
7
8
0.11 ± 0.03
0.61 ± 0.14
69 ± 2%
39 ± 5
23 ± 4
n.d.
36 ± 4
31 ± 3
n.d.
39 ± 10
19 ± 4
n.d.
22 ± 7
22 ± 4
n.d.
2.14 ± 0.33
2.71 ± 0.20
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
Abbreviations: ELISA, enzyme‐linked immunosorbent assay; n.d., not determined; RA, residual activity.
a
Obtained from three independent experiments. For compound 5, only RA of BCL‐2 at 50 µM
compound is given. Venetoclax was used as a positive control in the BCL‐2 ELISA assay, and the IC50
was determined to be 0.017 ± 0.005 µM.
b
The cytotoxicity data were obtained from three independent experiments.
and HMBCGP) from the Bruker pulse library were used. High‐
resolution mass measurements were performed on a Thermo Sci-
entific Q Exactive Plus mass spectrometer (Thermo Fisher Scientific)
at the Faculty of Pharmacy, University of Ljubljana. The purity of the
compounds was determined by HPLC‐UV obtained on a liquid
chromatography–mass spectrometry (LC‐MS) instrument (Applied
Biosystems API 2000 LC/MS/MS, HPLC Agilent 1100; Agilent) or on
a Thermo Scientific Dionex UltiMate 3000 UHPLC modular system
(Thermo Fisher Scientific), equipped with a photodiode array de-
®
tector set to 254 nm, whereby an Acquity UPLC BEH Phenyl Col-
umn (2.1 × 100 mm; 1.7 μm) was used, thermostated at 40°C, with a
flow rate set to 0.3 ml/min. The purities of the test compounds used
for the biological evaluations were confirmed to be ≥95% purity by
LC, unless stated otherwise.

10 of 13
SOSIČ ET AL.
|
The InChI codes of the investigated compounds, together with
(400 MHz, DMSO‐d
(hept, J = 7.0 Hz, 1H, CH(CH
7.21–7.26 (m, 2H, Ar–H), 7.39–7.45 (m, 2H, Ar–H), 7.93 (dd, J = 8.4,
6
) δ 1.25 (d, J = 7.0 Hz, 6H, CH(CH
3 2
) ), 2.98
some biological activity data, are provided as Supporting
Information.
3 2
)
), 6.97 (d, J = 7.9 Hz, 1H, Ar–H),
7
.1 Hz, 1H, Ar–H), 8.04 (d, J = 7.9 Hz, 1H, Ar–H), 8.12 (dd d, J = 7.0,
1
3
0
.6 Hz, 1H, Ar–H), 8.54 (dd, J = 8.4, 0.6 Hz, 1H, Ar–H); C NMR
) δ 24.1, 33.7, 111.3, 120.7, 122.7, 123.1, 124.4,
27.8, 127.8, 128.1, 128.3, 128.4, 146.7, 148.1, 152.2, 160.1, 186.2,
189.0; LC‐MS (ESI) (90% H O to 100% MeCN in 10 min, then 100%
4.1.2
|
Synthesis of compounds 1–3
(125 MHz, CDCl
3
1
5
‐Bromoacenaphthylene‐1,2‐dione (1)
2
[
34,35]
Compound 1 was synthesized as described previously.
Ace-
R
MeCN to 20 min, DAD 220–400 nm), t = 12.67 min, 94% purity, m/z
+
naphthenequinone (6.02 g, 33.0 mmol) was weighed into a 250 ml
[M+H] calcd. for C21
H
17
O
3
: 317.11, found: 316.9; HRMS (ESI) m/z
+
round‐bottom flask, followed by the addition of bromine (9.4 ml,
[M+H] calcd. for C21
H
17
O
3
: 317.1172, found: 317.1162.
1
82 mmol, 5.5 equiv). The reaction mixture was stirred at 75°C for
2
h. During the reaction, the flask was fitted with a reflux condenser,
7‐(4‐Isopropylphenoxy)‐1‐oxo‐1H‐phenalene‐2,3‐dicarbonitrile (3)
a T‐shaped adapter allowing a positive flow of argon, and an outlet
This compound was synthesized based on a previously described
[
24]
that led to a trap of 1 M NaOH (200 ml). After the reaction was
one‐pot sequence.
acenaphthylene‐1,2‐dione (2, 104 mg, 0.33 mmol) in MeCN (15 ml)
at room temperature, malononitrile (22 mg, 0.33 mmol) and K CO
3
To a solution of 5‐(4‐isopropylphenoxy)
2
complete, it was cooled to room temperature and diluted with H O
(
100 ml). Then, the mixture was cooled to 0°C and vigorously stirred
2
while a 40% aqueous solution of NaHSO
3
(100 ml) was added. The
O (200 ml). Pure
(4.6 mg, 0.033 mmol, 0.1 equiv) were added consecutively. The re-
action mixture was then heated at reflux for 1 h (note that the color
of the mixture went from yellow to dark green during heating). After
that time, the solvent was removed under reduced pressure and the
pure product was obtained by two consecutive column chromato-
graphy steps using first EtOAc/petroleum ether (1:3, v/v) and second
EtOAc/petroleum ether (1:4, v/v) as eluent systems. Yield (63 mg,
orange solid was filtered and washed with H
2
compound was obtained by column chromatography using EtOAc/
petroleum ether (1:4, v/v) as an eluent system. Yield (6.38 g, 74%);
3
4
f
light orange solid; mp: 216–218°C (lit. mp: 230−232°C); R = 0.36
1
(
EtOAc/petroleum ether, 1:3); H NMR (500 MHz, dimethyl sulfoxide
[
6
DMSO]‐d ) δ 7.96 (d, J = 7.5 Hz, 1H, Ar–H), 8.04 (dd, J = 8.3, 7.2 Hz,
1
H, Ar–H), 8.15 (app d, J = 7.2 Hz, 1H, Ar–H), 8.21 (d, J = 7.5 Hz, 1H,
52%); orange wax‐like solid; mp: 184–187°C; R
f
= 0.47 (EtOAc/pet-
) δ 1.31 (d, J = 6.9 Hz,
), 7.04 (d,
1
3
1
Ar–H), 8.39 (app d, J = 8.3 Hz, 1H, Ar–H); C NMR (125 MHz,
DMSO‐d ) δ 123.0, 123.0, 127.4, 129.8, 130.4, 130.6, 131.0, 131.5,
32.9, 145.2, 187.5, 187.7; LC‐MS (ESI) (90% H O to 100% MeCN in
roleum ether, 1:3); H NMR (500 MHz, CDCl
3
6
6H, CH(CH
)
3 2
), 3.00 (hept, J = 6.9 Hz, 1H, CH(CH
3
)
2
1
1
2
2
J = 8.6 Hz, 1H, Ar–H, H‐8), 7.12–7.16 (m, 2H, Ar–H, H‐2ʹ, H‐6ʹ),
7.35–7.39 (m, 2H, Ar–H, H‐3ʹ, H‐5ʹ), 7.86 (dd, J = 8.4, 7.4 Hz, 1H,
Ar–H, H‐5), 8.45 (dd, J = 7.4, 1.0 Hz, 1H, Ar–H, H‐4), 8.63 (d,
0 min, then 100% MeCN to 20 min, diode array detector [DAD]
+
20–400 nm), t
R
= 10.40 min, 98% purity, m/z [M+H] calcd. for
C
12
H
6
BrO
2
: 260.95, found: 260.9; high‐resolution mass spectro-
J = 8.6 Hz, 1H, Ar–H, H‐9), 8.91 (dd, J = 8.4, 1.0 Hz, 1H, Ar–H, H‐6);
+
13
metry (HRMS) (electrospray ionization [ESI]) m/z [M+H] calcd. for
3 3 2 3 2
C NMR (125 MHz, CDCl ) δ 24.2 (CH(CH ) ), 33.9 (CH(CH ) ),
C
12
H
6
BrO
2
: 260.9546, found: 260.9549.
112.5 (C8), 113.0 (CN), 113.2 (CN), 120.7 (C2), 121.0 (C2ʹ, C6ʹ),
1
1
22.4 (C9a), 122.7 (C3a), 124.0 (C6a), 127.0 (C5), 128.7 (C3a ),
5
‐(4‐Isopropylphenoxy)acenaphthylene‐1,2‐dione (2)
This compound was synthesized as described previously.
to a suspension of 5‐bromoacenaphthylene‐1,2‐dione (1, 261 mg,
.0 mmol) and 4‐isopropylphenol (136 mg, 1.0 mmol) in di-
methylformamide (DMF) (15 ml), CO (152 mg, 1.1 mmol,
.1 equiv) was added and the mixture stirred at 70°C for 5 h. The
128.8 (C3ʹ, C5ʹ), 131.1 (C3), 132.2 (C6), 134.7 (C4), 136.7 (C9), 147.6
(C4ʹ), 151.6 (C1ʹ), 163.8 (C7), 176.3 (C1); LC‐MS (ESI) (90% H O to
[18]
Briefly,
2
100% MeCN in 10 min, then 100% MeCN to 20 min, DAD
+
1
220–400 nm), t
R
= 12.70 min, 99% purity, m/z [M+H] calcd. for
+
K
2
3
C
24
H
17
N
2
O
2
: 365.12, found: 364.7; HRMS (ESI) m/z [M+H] calcd.
1
for C24
H
17 2 2
N O : 365.1285, found: 365.1279.
reaction mixture was then cooled to room temperature and poured
into a saturated aqueous solution of NaCl (50 ml), followed by ex-
traction with EtOAc (3 × 80 ml). The combined organic phases were
extracted with saturated aqueous solution of NaCl (5 × 50 ml), dried
4.1.3
|
Synthesis of compounds 4 and 5
[
24]
2
with Na SO
4
, filtered, and evaporated under reduced pressure. The
This procedure was performed as described previously.
To a so-
product was purified by column chromatography using EtOAc/pet-
roleum ether (1:3, v/v) as an eluent system. Yield (269 mg, 85%);
lution of 5‐bromoacenaphthylene‐1,2‐dione (1, 522 mg, 2.0 mmol) in
MeCN (20 ml) at room temperature, malononitrile (132 mg,
beige solid; mp: 143–146°C; R
f
= 0.49 (EtOAc/petroleum ether, 1:3);
), 2.99
), 6.98 (d, J = 7.9 Hz, 1H, Ar–H),
2 3
2.0 mmol) and K CO (28 mg, 0.2 mmol, 0.1 equiv) were added con-
1
H NMR (500 MHz, CDCl
3
) δ 1.31 (d, J = 7.0 Hz, 6H, CH(CH
3
)
2
secutively. The reaction mixture was then heated at reflux for 1 h
(note that the color of the mixture went from yellow to dark blue
during heating). TLC showed the consumption of the starting mate-
rial with two main products observed. The solvent was then removed
under reduced pressure and the crude material loaded on silica for
(
sep, J = 7.0 Hz, 1H, CH(CH
3
)
2
7
.12–7.16 (m, 2H, Ar–H), 7.33–7.37 (m, 2H, Ar–H), 7.84 (dd, J = 8.4,
7
.0 Hz, 1H, Ar–H), 8.00 (d, J = 7.9 Hz, 1H, Ar–H), 8.11 (app d,
1
J = 7.0 Hz, 1H, Ar–H), 8.59 (app d, J = 8.4 Hz, 1H, Ar–H); H NMR

SOSIČ ET AL.
11 of 13
|
purification by column chromatography using CH
2
Cl
2
as an eluent.
The combined organic phases were extracted with saturated aqu-
eous solution of NaCl (5 × 50 ml), dried with Na SO , filtered, and
The first‐eluting compound (5) and the second‐eluting compound (4)
were further purified by automated reversed‐phase flash chroma-
tography using 0.1% TFA in deionized water and MeCN as an eluent
system. After the chromatographic purification, fractions containing
each of the products were combined separately and organic volatiles
were evaporated in vacuo. The remaining aqueous solution was
2
4
evaporated under reduced pressure. The product was purified by
column chromatography using EtOAc/petroleum ether (1:3, v/v) as
an eluent system. Yield (104 mg, 26%); dark red solid; mp:
1
162–163°C;
R
f
= 0.30 (EtOAc/petroleum ether, 1:3);
) δ 3.13 (s, 3H, NCH ), 4.76 (s, 2H, PhCH
H NMR
(500 MHz, CDCl
3
3
2
), 7.15 (d,
made alkaline with a saturated aqueous solution of NaHCO
extracted with CH Cl (2 × 30 ml). The combined organic phases
were dried with Na SO , filtered, and volatile components evapo-
rated under reduced pressure to afford pure products.
3
and
J = 8.0 Hz, 1H, Ar–H), 7.34–7.38 (m, 1H, Ar–H), 7.39–7.45 (m, 4H,
Ar–H), 7.60 (dd, J = 8.4, 7.0 Hz, 1H, Ar–H), 7.96 (d, J = 7.0 Hz,
1H, Ar–H), 8.05 (d, J = 8.0 Hz, 1H, Ar–H), 8.28 (app d, J = 8.4 Hz, 1H,
2
2
2
4
13
Ar–H); C NMR (125 MHz, CDCl
23.6, 124.3, 126.4, 127.2, 127.8, 128.7, 129.0, 130.0, 136.7, 148.8,
154.9, 185.6, 190.4; LC‐MS (ESI) (90% H O to 100% MeCN in
3
) δ 40.4, 60.1, 113.9, 121.7, 121.8,
1
7
‐Bromo‐1‐oxo‐1H‐phenalene‐2,3‐dicarbonitrile (4)
2
Yield (129 mg, 21%); orange solid; mp: 238–239°C;
R
f
= 0.64
R
10 min, then 100% MeCN to 20 min, DAD 220–400 nm), t = 11.58
1
+
(
CH
2
Cl
2
); H NMR (500 MHz, DMSO‐d
6
) δ 8.06 (dd, J = 8.4, 7.5 Hz,
min, 97% purity, m/z [M+H] calcd. for C20
H
16NO
2
: 302.11, found:
: 302.1176,
+
1
1
H, Ar–H, H‐5), 8.37 (d, J = 8.0 Hz, 1H, Ar–H, H‐8), 8.44 (d, J = 8.0 Hz,
301.8; HRMS (ESI) m/z [M+H] calcd. for C20
H
16NO
2
H, Ar–H, H‐9), 8.46 (app d, J = 7.5 Hz, 1H, Ar–H, H‐4), 8.69 (app d,
found: 302.1173.
1
3
J = 8.4 Hz, 1H, Ar–H, H‐6); C NMR (125 MHz, DMSO‐d
6
) δ 113.4
(
CN), 113.7 (CN), 120.1 (C2), 122.8 (C3a), 127.0 (C9a), 127.3 (C7),
1
29.6 (C5), 130.6 (C6a), 131.4 (C3), 132.9 (C9), 133.3 (C8), 133.8
O to
4.1.5 Synthesis of compounds 7 and 8
|
(
C3a1), 135.1 (C4), 135.7 (C6), 177.2 (C1); LC‐MS (ESI) (90% H
2
[
24]
1
2
00% MeCN in 10 min, then 100% MeCN to 20 min, DAD
This procedure was performed as described previously.
To a
+
20–400 nm), t
BrN
for C21
R
= 10.97 min, 96% purity, m/z [M+H] calcd. for
solution of 5‐[benzyl(methyl)amino]acenaphthylene‐1,2‐dione (6,
+
C
15
H
6
2
O: 308.96, found: 308.9; HRMS (ESI) m/z [M+H] calcd.
75 mg, 0.25 mmol) in MeCN (15 ml) at room temperature, mal-
H
17
O
3
: 308.9658, found: 308.9646.
ononitrile (17 mg, 0.25 mmol) and
2 3
K CO (3.5 mg, 0.025 mmol,
0
.1 equiv) were added consecutively. The reaction mixture was then
6
‐Bromo‐1‐oxo‐1H‐phenalene‐2,3‐dicarbonitrile (5)
heated at reflux for 1 h (note that the color of the mixture went from
yellow to dark blue during heating). After that time, the solvent was
removed under reduced pressure and the crude product (showing
two major spots on TLC) loaded on silica for purification by column
chromatography using EtOAc/petroleum ether (1:1, v/v) as an eluent
system. Under these conditions, only the first‐eluting compound (7)
was purified successfully. For the second‐eluting compound (8), three
additional column chromatography purifications using EtOAc/
petroleum ether (1:1, v/v) as an eluent system were needed.
Yield (186 mg, 30%); orange solid; mp: 241–244°C;
f
R = 0.67
1
(
CH
2
Cl
2
); H NMR (400 MHz, DMSO‐d
6
) δ 8.11 (dd, J = 8.3, 7.1 Hz,
1
H, Ar–H, H‐8), 8.29 (dd, J = 7.1, 0.6 Hz, 1H, Ar–H, H‐9), 8.32 (d,
J = 7.8 Hz, 1H, Ar–H, H‐4), 8.35 (d, J = 7.8 Hz, 1H, Ar–H, H‐5), 8.43
1
3
(
dd, J = 8.3 Hz, 0.6 Hz, 1H, Ar–H, H‐7); C NMR (101 MHz, DMSO‐
1
d
6
) δ 111.7 (CN), 113.2 (CN), 124.3 (C9), 124.4 (C4), 126.8 (3a ),
1
28.7 (C3a), 129.4 (C6a), 130.2 (C9a), 130.9 (C2), 130.9 (C7), 131.1
(
C8), 132.8 (C5), 142.2 (C6), 155.5 (C3), 185.30 (C1); HPLC (95%
H
4
2
O [with 0.1% TFA] to 95% MeCN in 10 min, then 95% MeCN for
min), t
R
= 7.21 min, 96% purity, detection at 254 nm; HRMS (ESI)
7‐[Benzyl(methyl)amino]‐1‐oxo‐1H‐phenalene‐2,3‐dicarbonitrile (7)
+
m/z [M+H] calcd. for C15
H
6
BrN
2
O: 308.9658, found: 308.9650.
This compound eluted first from the column. Yield (36 mg, 41%); dark
blue wax‐like solid; mp: 244–248°C; R
f
= 0.36 (EtOAc/petroleum
) δ 3.35 (s, 3H, NCH ), 4.99
), 7.32–7.35 (m, 3H, Ar–H, H‐2ʹ, H‐6ʹ, H‐4ʹ), 7.38 (d,
1
ether, 1:1); H NMR (500 MHz, DMSO‐d
6
3
4.1.4
|
Synthesis of compound 6
(s, 2H, PhCH
2
J = 9.0 Hz, 1H, Ar–H, H‐8), 7.39–7.42 (m, 2H, Ar–H, H‐3ʹ, H‐5ʹ), 7.75
(app t, J = 7.9 Hz, 1H, Ar–H, H‐5), 8.34 (d, J = 7.3 Hz, 1H, Ar–H, H‐4),
8.41 (d, J = 9.0 Hz, 1H, Ar–H, H‐9), 8.69 (d, J = 8.2 Hz, 1H, Ar–H, H‐6);
5
‐[Benzyl(methyl)amino]acenaphthylene‐1,2‐dione (6)
‐Bromoacenaphthylene‐1,2‐dione (1, 350 mg, 1.34 mmol) was sus-
pended in DMF (15 ml), followed by the addition of N‐
benzylmethylamine (163 mg, 1.34 mmol) and CO (204 mg,
.47 mmol, 1.1 equiv). The reaction mixture was stirred at 80°C for
2 h. Because TLC showed the presence of the starting compound 1,
5
13
13
C NMR (125 MHz, DMSO‐d
6
6
) δ C NMR (1125 MHz, DMSO‐d ) δ
K
2
3
43.1 (CH ), 60.0 (CH ), 113.9 (CN), 114.7 (CN), 115.6 (C8), 118.8
3
2
1
1
(C2), 119.4 (C9a), 121.9 (C3a), 122.1 (C6a), 124.5 (C5), 127.1 (C2ʹ,
1
C6ʹ), 127.6 (C4ʹ), 128.1 (C3), 128.8 (C3ʹ, C5ʹ), 129.1 (C3a ), 134.0
additional amounts of N‐benzylmethylamine (81 mg, 0.67 mmol,
.5 equiv) and K CO (93 mg, 0.67 mmol, 0.5 equiv) were added and
(C4), 134.9 (C9), 135.4 (C6), 136.1 (C1ʹ), 159.7 (C7), 174.0 (C1); LC‐
0
2
3
MS (ESI) (90% H
2
O to 100% MeCN in 10 min, then 100% MeCN to
+
the mixture stirred at 80°C for additional 4 h. Then, it was cooled to
room temperature and poured into a saturated aqueous solution
of NaCl (100 ml), followed by extraction with EtOAc (4 × 80 ml).
20 min, DAD 220–400 nm), t
calcd. for C23 O: 350.12, found: 349.8; HRMS (ESI) m/z [M+H]
calcd. for C23 O: 350.1288, found: 350.1281.
R
= 11.45 min, 98% purity, m/z [M+H]
+
H
16
N
3
H
16
N
3

12 of 13
SOSIČ ET AL.
|
6
‐[Benzyl(methyl)amino]‐1‐oxo‐1H‐phenalene‐2,3‐dicarbonitrile (8)
Instruments Inc.) at 490 nm and IC50 values were determined after
fitting curves using GraphPad Prism (GraphPad Software). Three
independent experiments were performed with each inhibitor to
calculate average IC50 values and standard deviation.
This compound eluted second from the column. Yield (7 mg, 9%);
dark violet wax‐like solid; mp: 248–251°C; R
f
= 0.22 (EtOAc/petro-
) δ 3.51 (s, 3H, NCH ),
), 7.31 (d, J = 9.2 Hz, 1H, Ar–H, H‐5), 7.34–7.38 (m,
H, Ar–H, H‐2ʹ, H‐6ʹ, H‐4ʹ), 7.40–7.44 (m, 2H, Ar–H, H‐3ʹ, H‐5ʹ), 7.89
1
leum ether, 1:1); H NMR (500 MHz, DMSO‐d
6
3
5
.13 (s, 2H, PhCH
2
3
(
t, J = 8.1 Hz, 1H, Ar–H, H‐8), 8.06 (d, J = 9.2 Hz, 1H, Ar–H, H‐4), 8.67
dd, J = 7.5, 0.9 Hz, 1H, Ar–H, H‐9), 8.81 (dd, J = 8.3, 0.9 Hz, 1H, Ar–H,
4.2.2 Cell culture
|
(
13
H‐7); C NMR (125 MHz, DMSO‐d
6
) δ 44.0 (CH
3
), 59.8 (CH
2
), 106.5
A2780 (kind gift from Prof. Ulrich Jaehde, Pharmaceutical Institute,
University of Bonn) and MV‐3 (kind gift from Epo GmbH Berlin) cell
lines were cultivated in RPMI‐1640 medium (PAN Biotech) con-
taining 10% fetal calf serum (FCS; Sigma‐Aldrich), 1% penicillin/
streptomycin (PAN Biotech), and 1.5% L‐glutamine (all from PAN
Biotech). U87 cells (kind gift from Prof. Christa E. Müller, Pharma-
ceutical Institute, University of Bonn) were cultured in Dulbecco's
modified Eagle's medium supplemented with 10% FCS, 1% penicillin/
streptomycin, and 1.5% L‐glutamine. The same medium with the
addition of 1 mM pyruvate was used for MDA‐MB‐231 cells (kind gift
from Epo GmbH Berlin). All cells were grown in a humidified cham-
(
C2*), 113.1 (C3a), 114.0 (C5), 114.4 (CN*), 115.9 (CN*), 123.4 (C6a),
1
26.3 (C3), 126.6 (C8), 127.1 (C2ʹ, C6ʹ), 127.8 (C4ʹ), 129.0 (C3ʹ, C5ʹ),
1
1
29.2 (C3a ), 129.6 (C9a), 132.8 (C9), 135.5 (C1ʹ), 135.6 (C7), 136.6
(
2
C4), 160.5 (C6), 176.9 (C1); LC‐MS (ESI) (90% H O to 100% MeCN
R
in 10 min, then 100% MeCN to 20 min, DAD 220–400 nm), t =
+
1
0.81 min, 97% purity, m/z [M+H] calcd. for C23
H
16
N
3
O: 350.12,
+
found: 349.8; HRMS (ESI) m/z [M+H] calcd. for C23
H
16
N
3
O:
3
50.1288, found: 350.1279. *The assignations for these signals were
obtained from predictions using ChemDraw Ultra 18.0.
2
ber at 37°C and with 5% CO .
4.2
|
Biochemical assays
4.2.1
|
BCL‐2 competitive inhibition assay
4.2.3 Metabolic activity assay
|
A 0.06 µg/ml solution (1 mg/ml in 20% EtOH/H
2
O) of biotinylated
Cytotoxicity of BCL‐2 inhibitors was determined with the MTT assay.
4
BIM peptide (residues 81–106: (Biotin)‐β‐Ala‐β‐Asp‐Met‐Arg‐Pro‐
Glu‐Ile‐Trp‐Ile‐Ala‐Gln‐Glu‐Leu‐Arg‐Arg‐Ile‐Gly‐Asp‐Glu‐Phe‐Asn‐
Cells (1 × 10 cells [A2780, U87, MV‐3, or MDA‐MB‐231] per well)
were seeded on a 96‐well plate and left to adhere. Afterward,
compounds were added at different concentrations ranging from 1 to
50 µM. After 24‐h incubation, the MTT reagent was added and left
for 1 h before adding DMSO. Next, absorbance was measured at
690 nm on the Tecan plate reader. IC50 values were calculated from
three independent experiments in GraphPad prism.
2
Ala‐Tyr‐Tyr‐Ala‐Arg‐Arg‐NH [catalog no. 3526; Tocris Bioscience,
Bio‐Techne Ltd.]) was prepared in SuperBlock Blocking Buffer
(
Thermo Fisher Scientific). The solution (100 µl/well) was applied to a
®
streptavidin‐coated 96‐well plate (Pierce Streptavidin Coated High
Binding Capacity Plates, catalog no. 15500; Thermo Fisher Scientific)
and incubated at room temperature with shaking for 1.5 h. Different
concentrations of inhibitors (prepared from a 20 mM stock DMSO
solutions) in phosphate‐buffered saline (PBS) were incubated with
ACKNOWLEDGMENTS
We thank Maja Frelih for HRMS measurements. This study was funded
by the Slovenian Research Agency (research core funding No. P1‐0208,
Grant J1‐2485 to Izidor Sosič). Izidor Sosič and Martina Gobec ac-
knowledge short‐term fellowships from FEBS and EMBO, respectively.
2
0 nM purified His‐tagged BCL‐2 protein (catalog no. 10195‐H08E;
Sino Biological) for 1 h at room temperature. The BIM–
biotin–streptavidin‐coated plates were washed three times with
0
.05% Tween‐20 in PBS. Afterward, 100 ‐µl aliquots of the
inhibitor–protein solutions were transferred to the BIM–
biotin–streptavidin‐coated plate and incubated at room temperature
for 2 h. The plate was washed three times with 0.05% Tween‐20
in PBS, followed by applying 100 µl anti‐His‐tag mouse
mAb–horseradish peroxidase conjugate (catalog no. BZ‐652504;
BioLegend), 1:1000 dilution in SuperBlock Blocking Buffer to the
well, and incubating the plate for 1 h at room temperature. The plate
was then washed five times with 0.05% Tween‐20 in PBS, and 100 µl
of o‐phenylenediamine (prepared by dissolving one pre‐packaged
CONFLICTS OF INTERESTS
The authors declare that there are no conflicts of interests.
ORCID
Christian Steinebach
Michael Gütschow
https://orcid.org/0000-0001-5638-1955
http://orcid.org/0000-0002-9376-7897
REFERENCES
[1] R. D. Taylor, M. MacCoss, A. D. Lawson, J. Med. Chem. 2014, 57, 5845.
tablet [catalog no. P5412; Aldrich] in 10 ml of 0.1 M Na
.05 M citric acid buffer, pH 5, with added 6 µl of 30% H
use) was applied to the wells. The enzymatic activity was stopped
after 5 min with the addition of 2 mM H SO . Absorbance was
measured on BioTek Synergy HT microplate reader (BioTek
2
HPO
4
/
[2] M. Aldeghi, S. Malhotra, D. L. Selwood, A. W. Chan, Chem. Biol. Drug
Des. 2014, 83, 450.
0
2 2
O before
[3] M. D. Polêto, V. H. Rusu, B. I. Grisci, M. Dorn, R. D. Lins, H. Verli,
Front. Pharmacol. 2018, 9, 395.
2
4
[4] A. S. Tikhomirov, A. A. Shtil, A. E. Shchekotikhin, Recent Pat. Anti‐
Cancer Drug Discov. 2018, 13, 159.

SOSIČ ET AL.
13 of 13
|
[
[
[
5] U. Neumann, M. Gütschow, Bioorg. Chem. 1995, 23, 72.
6] Y. Xiao, F. Liu, X. Qian, J. Cui, Chem. Commun. 2005, 239.
7] F. Liu, Y. Xiao, X. Qian, Z. Zhang, J. Cui, D. Cui, R. Zhang, Tetrahedron
[25] P. De Bonfils, E. Verron, C. Sandoval‐Altamirano, P. Jaque, X. Moreau,
G. Gunther, P. Nun, V. Coeffard, J. Org. Chem. 2020, 85, 10603.
[26] T. Fukuyama, T. Sugimori, S. Maetani, I. Ryu, Org. Biomol. Chem.
2018, 16, 7583.
2
005, 61, 11264.
[8] Y. Liu, Y. Xu, X. Qian, Y. Xiao, J. Liu, L. Shen, J. Li, Y. Zhang, Bioorg.
Med. Chem. Lett. 2006, 16, 1562.
[27] F. Ospina, A. Ramirez, M. Cano, W. Hidalgo, B. Schneider,
F. Otálvaro, J. Org. Chem. 2017, 82, 3873.
[9] L. Xiong, M. Yu, M. Cheng, M. Zhang, X. Zhang, C. Xu, F. Li, Mol.
BioSyst. 2009, 5, 241.
[28] T. Song, Z. Wang, F. Ji, Y. Feng, Y. Fan, G. Chai, X. Li, Z. Li, Z. Zhang,
Angew. Chem. Int. Ed. 2016, 55, 14250.
[
10] M. Zhang, M. Yu, F. Li, M. Zhu, M. Li, Y. Gao, L. Li, Z. Liu,
[29] R. Delbridge, S. Grabow, A. Strasser, D. L. Vaux, Nat. Rev. Cancer
2016, 16, 99.
J. Zhang, D. Zhang, T. Yi, C. Huang, J. Am. Chem. Soc. 2007, 129,
1
0322.
11] F. Liu, X. Qian, J. Cui, Y. Xiao, R. Zhang, G. Li, Bioorg. Med. Chem.
006, 14, 4639.
[30] A. Anantram, M. Degani, Future Med. Chem. 2019, 11, 2287.
[31] C. Steinebach, Y. L. D. Ng, I. Sosič, C. S. Lee, S. Chen, S. Lindner,
L. P. Vu, A. Bricelj, R. Haschemi, M. Monschke, E. Steinwarz,
K. G. Wagner, G. Bendas, J. Luo, M. Gütschow, J. Krönke, Chem. Sci.
2020, 11, 3474.
[
[
[
[
[
[
[
[
[
[
[
[
[
[
2
12] Z. Zhang, Y. Yang, D. Zhang, Y. Wang, X. Qian, F. Liu, Bioorg. Med.
Chem. 2006, 14, 6962.
13] Z. Chen, X. Wang, W. Zhu, X. Cao, L. Tong, H. Li, H. Xie, Y. Xu, S. Tan,
D. Kuang, J. Ding, X. Qian, J. Med. Chem. 2011, 54, 3732.
14] Z. Zhang, L. Jin, X. Qian, M. Wei, Y. Qang, J. Qang, Y. Yang, Q. Xu,
Y. Xu, F. Liu, ChemBioChem 2007, 8, 113.
[32] R. Meyers, J. Bryan, J. McFarland, B. A. Weir, A. E. Sizemore, H. Xu,
N. V. Dharia, P. G. Montgomery, G. S. Cowley, S. Pantel, A. Goodale,
Y. Lee, L. D. Ali, G. Jiang, R. Lubonja, W. F. Harrington,
M. Strickland, T. Wu, D. C. Hawes, V. A. Zhivich, M. R. Wyatt,
Z. Kalani, J. J. Chang, M. Okamoto, K. Stegmaier, T. R. Golub,
J. S. Boehm, F. Vazquez, D. E. Root, W. C. Hahn, A. Tsherniak, Nat.
Genet. 2017, 49, 1779.
15] Z. Zhang, T. Song, T. Zhang, J. Gao, G. Wu, L. An, G. Du, G., Int.
J. Cancer 2011, 128, 1724.
16] Z. Zhang, G. Wu, F. Xie, T. Song, X. Chang, J. Med. Chem. 2011, 54,
[33] Z. Zhang, T. Song, X. Li, Z. Wu, Y. Feng, F. Xie, G. Wu, C. Liu, J. Qin,
1
101.
17] Z. Zhang, H. Yang, G. Wu, Z. Li, T. Song, X. Li, Eur. J. Med. Chem.
011, 46, 3909.
H. Chen, Eur. J. Med. Chem. 2013, 59, 141.
[
34] C. G. Grant, S. O. Kang, B. P. Hay, J. Org. Chem. 2013, 78, 7735.
35] C. Favero, M. B. Closs, G. B. Galland, R. Stieler, E. Rosseto,
K. Bernardo‐Gusmão, J. Catal. 2019, 377, 63.
2
[
18] T. Song, X. Li, X. Chang, X. Liang, Y. Zhao, G. Wu, S. Xie, P. Su, Z. Wu,
Y. Feng, Z. Zhang, Bioorg. Med. Chem. 2013, 21, 11.
19] T. Song, X. Yu, Y. Liu, X. Li, G. Chai, Z. Zhang, ChemBioChem 2015,
1
6, 757.
SUPPORTING INFORMATION
20] J. Zhu, Z. Wang, Z. Guo, X. Zhang, T. Song, Y. Guo, T. Ji, Z. Zhang, Z.,
Arch. Pharm. 2020, 353, e2000005.
Additional Supporting Information may be found online in the
supporting information tab for this article.
21] T. Song, Q. Chen, X. Li, G. Chai, Z. Zhang, J. Med. Chem. 2013, 56,
9
366.
22] T. Song, X. Li, Y. Yang, G. Wu, S. Xie, P. Su, Y. Feng, Z. Zhang, Z.,
Bioorg. Med. Chem. 2014, 22, 663.
How to cite this article: I. Sosič, M. Gobec, C. Steinebach, M.
Schlesinger, G. Bendas, M. Gütschow, Arch. Pharm. 2021,
e2100151. https://doi.org/10.1002/ardp.202100025
23] H. Li, F. Liu, P. Pellechia, M. D. Smith, X. Qian, G. Wang, Q. Wang, Q.
Tetrahedron 2014, 70, 5872.
24] R. Lenk, A. Tessier, P. Lefranc, V. Silvestre, A. Planchat, V. Blot,
D. Dubreuil, J. Lebreton, J. Org. Chem. 2014, 79, 9754.