Targeting G-quadruplex DNA structures in the telomere and oncogene promoter regions by benzimidazole?carbazole ligands

Article abstract of DOI:10.1016/j.ejmech.2018.01.091

Recent studies support the idea that G-quadruplex structures in the promoter regions of oncogenes and telomere DNA can serve as potential therapeutic targets in the treatment of cancer. Accordingly, several different types of organic small molecules that stabilize G-quadruplex structures and inhibit telomerase activity have been discerned. Here, we describe the binding of benzimidazole-carbazole ligands to G-quadruplex structures formed in G-rich DNA sequences containing the promoter regions of human c-MYC, c-KIT1, c-KIT2, VEGF and BCL2 proto-oncogenes. The fluorescence spectroscopic data indicate that benzimidazole-carbazole ligands bind and stabilize the G-quadruplexes in the promoter region of oncogenes. The molecular docking studies provide insights into the mode and extent of binding of this class of ligands to the G-quadruplexes formed in oncogene promoters. The high stability of these G-quadruplex structures was validated by thermal denaturation and telomerase-catalyzed extension of the 3' end. Notably, benzimidazole-carbazole ligands suppress the expression of oncogenes in cancer cells in a dose-dependent manner. We anticipate that benzimidazole-carbazole ligands, by virtue of their ability to stabilize G-quadruplex structures in the promoter regions of oncogenes, might reduce the risk of cancer through the loss of function in the proteins encoded by these genes.

Full text of DOI:10.1016/j.ejmech.2018.01.091

Accepted Manuscript
Targeting G-quadruplex DNA structures in the telomere and oncogene promoter
regions by benzimidazole‒carbazole ligands
Mangesh H. Kaulage, Basudeb Maji, Sanjeev Pasadi, Asfa Ali, Santanu
Bhattacharya, Kalappa Muniyappa
PII:
S0223-5234(18)30113-2
10.1016/j.ejmech.2018.01.091
EJMECH 10168
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 17 November 2017
Revised Date: 17 January 2018
Accepted Date: 27 January 2018
Please cite this article as: M.H. Kaulage, B. Maji, S. Pasadi, A. Ali, S. Bhattacharya, K. Muniyappa,
Targeting G-quadruplex DNA structures in the telomere and oncogene promoter regions by
benzimidazole‒carbazole ligands, European Journal of Medicinal Chemistry (2018), doi: 10.1016/
j.ejmech.2018.01.091.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
Targeting G-quadruplex DNA structures in the telomere and oncogene promoter regions by
benzimidazole‒carbazole ligands
a,b,1
b,1
a
b
b∗
Mangesh H. Kaulage
, Basudeb Maji , Sanjeev Pasadi , Asfa Ali , Santanu Bhattacharya
a∗
and Kalappa Muniyappa
a
Department of Biochemistry, Indian Institute of Science, Bangalore 560012, India.
b
Department of Organic Chemistry, Indian Institute of Science, Bangalore 560012, India.
These authors contributed equally to this work.
1
∗
To whom correspondence should be addressed: Email: kmbc@iisc.ac.in or sb@iisc.ac.in
Graphical abstract
1

ACCEPTED MANUSCRIPT
Highlights
•
•
•
•
Benzimidazole-carbazole ligands stabilize promoter and telomere G4 DNA
Benzimidazole-carbazole ligands stabilize parallel topology of G4 DNA
Ligand-mediated G4 DNA stabilization inhibits extension of 3ꞌ end of telomere
Ligand-mediated G4 DNA stabilization at the promoters suppresses gene expression
Abstract
Recent studies support the idea that G-quadruplex structures in the promoter regions of
oncogenes and telomere DNA can serve as potential therapeutic targets in the treatment of
cancer. Accordingly, several different types of organic small molecules that stabilize G-
quadruplex structures and inhibit telomerase activity have been discerned. Here, we describe the
binding of benzimidazole-carbazole ligands to G-quadruplex structures formed in G-rich DNA
sequences containing the promoter regions of human c-MYC, c-KIT1, c-KIT2, VEGF and BCL2
proto-oncogenes. The fluorescence spectroscopic data indicate that benzimidazole-carbazole
ligands bind and preferentially stabilize the parallel topology of G-quadruplexes in the promoter
region of oncogenes. The molecular docking studies provide insights into the mode and extent of
binding of this class of ligands to the G-quadruplexes formed in oncogene promoters. The high
stability of these G-quadruplex structures was validated by thermal denaturation and telomerase-
catalyzed extension of the 3ꞌ end. Notably, benzimidazole-carbazole ligands suppress the
expression of oncogenes in cancer cells in a dose-dependent manner. We anticipate that
benzimidazole-carbazole ligands, by virtue of their ability to stabilize G-quadruplex structures in
2

ACCEPTED MANUSCRIPT
the promoter regions of oncogenes, might reduce the risk of cancer through the loss of function
in the proteins encoded by these genes.
Keywords: Oncogenes, G-quadruplex DNA, benzimidazole‒carbazole conjugates, thiazole
orange, fluorescence intercalator displacement, polymerase stop assay, transcriptional regulation.
1
. Introduction
The human genome contains highly repetitive sequences: among them, the most
polymorphic are the G-rich sequences, which readily fold into a variety of G-quadruplex
structures [1]. A typical G-quadruplex comprises of a planar cyclic array of four guanine
residues held together by Hoogsteen hydrogen bonds [1]. By stacking on top of each other, these
tetrads produce a right-handed quadruple helical structure. The G-rich sequences give rise to
diverse G-quadruplex structures by folding inter- or intra-molecularly and can form hybrid-type
mixed parallel/antiparallel G-strands [1-3]. The intramolecular G-quadruplex structures can
adopt diverse topologies; depending on the orientation of strands, size/sequence of loops,
glycosidic torsion angles, and type of metal ions [1-3]. The computational studies and high-
throughput sequencing have revealed that approximately 716310 distinct G-quadruplex forming
sequences exist in the human genome [4, 5].
A significant body of work has also shown that G-quadruplex forming sequences are not
randomly distributed throughout the genome, but tend to cluster in specific regions. The non-
random distribution of G-quadruplex forming sequences and their evolutionary conservation in
genomes posit that these sequences may be associated with a wide range of fundamental
biological processes [4, 6, 7]. The most common form of DNA in cells is the B-form, in which
3

ACCEPTED MANUSCRIPT
G/C base pairing precludes the formation of G-quadruplex structures. However, the guanines
within a G/C-rich region can fold into G-quadruplex structures as a consequence of transient
strand separation during the processes of transcription, replication and recombination [8, 9]. The
research activity in the area of G-quadruplex structures has hugely expanded in the past decade
due to their existence at biologically relevant loci under physiological conditions and their
potential druggability in human cells [10-12].
The telomeres protect chromosomes from degradation and genomic instability; they also
regulate telomerase activity [13, 14]. Telomerase is a reverse transcriptase that plays a key role
in telomere length homeostasis by adding TTAGGG repeats to the ends of chromosomes [15,
1
6]. For optimal telomerase activity, unfolded telomeric G-rich single strands are a necessity; on
the other hand, folding of this overhang into a G-quadruplex structure inhibits telomere
elongation by telomerase. Given the fact that telomerase is overexpressed in a majority of human
cancer cells, it has emerged as a potential target in the development of anti-cancer drugs [17].
Among several different approaches to telomerase inhibition, the stabilization of G-quadruplexes
by small molecules has received much attention [18]. As human telomeric single-stranded DNA
(ssDNA) folds into a variety of intramolecular G-quadruplex structures [19-23], the structural
features that are considered during the development of ligands include the G-tetrad surface,
discrete grooves resulting from combinations of syn- and anti- deoxyguanosine conformations, a
central channel with a negative charge and motifs within the flexible loop regions [24]. Owing to
their potential roles in disease pathology, several hundreds of small molecules that interact with
and stabilize telomeric G-quadruplex structures have been reported in literature [25].
Previous analyses using bioinformatics tools have revealed that G-quadruplex DNA motifs
are embedded within the promoters of human oncogenes as well in several genes that encode
4

ACCEPTED MANUSCRIPT
transcription factors, but are under-represented in the promoters of housekeeping genes and
tumor suppressor genes [24, 26]. A number of studies have provided persuasive evidence
supporting functional roles of the promoter G-quadruplexes of various human genes such as INS
[27], c-MYC [28], VEGF [29], HIF-1α [30], BCL-2 [31], MtCK [32], KRAS [33], c-KIT [34],
RET [35], PDGF-A [36] and ACC1 [37] and also their potential druggability [1, 18, 38]. Several
classes of small molecules that bind and stabilize telomeric G-quadruplex structures have been
reported [25, 39, 40]. Most of the reported ligands have a planar aromatic surface and interact
with the external surface of the G-quartet by π-stacking interactions. Along with this selectivity
and affinity of a ligand can be enhanced by the electrostatic as well as H-bonding interactions of
the neutral/cationic side chain with the grooves/loops of the quadruplex structure [39, 41].
In previous research works, a variety of experimental techniques have revealed that
benzimidazole‒carbazole ligands selectively bind and stabilize telomeric G-quadruplex
structures, inhibit telomerase activity and proliferation of human cancer cells [42, 43].
Additionally, these ligands were found to induce topological conversion from non-parallel to
parallel forms in human telomeric G-quadruplex structures [42, 43]. In this study, we
demonstrate the ability of benzimidazole-carbazole ligands to bind, stabilize and target G-
quadruplex structures formed at telomeres and in the promoter regions of human c-MYC, c-KIT1,
c-KIT2, VEGF and BCL2 proto-oncogenes. The data presented here suggest that benzimidazole-
carbazole ligands can inhibit telomerase-catalyzed extension of the 3ꞌ end [42]. The biological
evidence that these ligands have important physiological consequences are presented: they
suppress the expression of oncogenes in human cells thus indicating cellular relevance to
oncogene promoter G-quadruplexes and potential to reduce the risk of cancer.
5

ACCEPTED MANUSCRIPT
2
. Results and Discussion
2
.1. Ligands used in this study
The benzimidazole scaffolds are a valuable class of heterocyclic compounds that have
many applications in the pharmaceutical industry. The benzimidazole containing compounds
interact with various cellular targets (DNA being the primary target) and elicit a variety of
cellular responses. Owing to their planarity, benzimidazole derivatives have the potential to
interact with DNA via the intercalation binding mode [44-46]. Another class of compounds,
which contain the bis-benzimidazole group (for example: Hoechst 33342 and Hoechst 33258)
interact through the minor groove of duplex DNA [47, 48]. The benzimidazole-transition metal
2
+
ion conjugates also interact with DNA. For example, benzimidazole-Cu complexes intercalate
into duplex DNA [49]. The binding activity of benzimidazole moiety (containing groove
binders) is also studied with respect to their interaction with G-quadruplex DNA [50].
The compounds containing a carbazole moiety express a wide range of pharmacological
activities including anticancer, antiviral, antifungal, anticoagulant, antiparasitic, analgesic,
anticonvulsant, antiulcer and antihypertensive activities [51-55]. The chemistry and biology of
carbazoles have attracted great interest in the context of their use in the synthesis of DNA
binders due to their desirable electronic structures and large π-conjugated surface. In addition,
various functional groups can be attached to a structurally rigid carbazolyl ring. The ligands
containing carbazole moiety bind to the minor groove of A/T-rich sequences and also interact by
intercalation as they contain a flat chromophore [56]. The research stimulated by the attractive
properties of carbazoles has led to the synthesis of carbazole derivatives that recognize and
stabilize a variety of G-quadruplex DNA structures [42, 43, 49, 56-60].
6

ACCEPTED MANUSCRIPT
In previous research work, benzimidazole-carbazole conjugates were shown to
selectively stabilize telomeric G-quadruplex DNA over duplex DNA [42]. These ligands possess
carbazole as a central pharmacophore with varying number of flanking benzimidazole moieties
and they preferentially bind to telomeric G-quadruplex DNA over duplex DNA. In this study, the
ability of benzimidazole-carbazole conjugates to bind the G-quadruplex structures formed by G-
rich DNA sequences (Table 1) in the promoter region of human c-MYC, c-KIT1, c-KIT2, VEGF
and BCL2 proto-oncogenes was investigated. Five benzimidazole-carbazole conjugate ligands
were used (Fig. 1), of which the synthesis of ligand 2 and 4 have been reported previously [42];
newly synthesized ligands 1, 3 and 5 were included to widen the pharmacological scope. The
strategy employed in the synthesis of ligand 1 was based on the observation that carbazole-
benzimidazole pharmacophore efficiently binds to and stabilizes telomeric G-quadruplex DNA
[42]. While protonatable hydroxyethyl piperazine moieties in ligand 1 can electrostatically
interact with negatively charged phosphodiester backbone, the hydroxyl moieties can function as
hydrogen bond donor. Further, ligand 2 has three protonatable piperazine or morpholine moieties
which can impart electrostatic interactions and ligand 3 contains an extra hard positive charge
that might facilitate ligand-DNA interaction. Both ligand 2 and 3 are devoid of the hydroxyethyl
group and can disable the H-bond donor property. In the case of ligands 4 and 5, the
hydroxyethyl group along with three protonatable moieties, either N-methyl-piperazine or
dimethylamine, was added. Thus, keeping the carbazole-bis-benzimidaole central
pharmacophore common, the peripheral functional groups were varied to study their effect on G-
quadruplex binding efficacy and changes in their specificity towards different topological
structures.
7

ACCEPTED MANUSCRIPT
Figure. 1. Structure of benzimidazole‒carbazole conjugates used in this study.
Table 1. Telomeric and oncogene promoter oligonucleotide sequences used in this study.
Promoter
C-Myc
VEGF
5'<-----Sequence----->3'
TGGGGAGGGTGGGGAGGGTGGGGAAGG
CGGGGCGGGCCGGGGGCGGGGT
TTAGGGTTAGGGTTAGGGTTAGGG
AGGGAGGGCGCTGGGAGGAGGG
CGGGCGGGCGCGAGGGAGGGG
Telomere
c-Kit
c-Kit-2
BCL2
AGGGGCGGGCGCGGGAGGAAGGGGGCGGGAGCGGGGCTG
2
.2. Fluorescence emission spectral titrations
To investigate the ability of different benzimidazole‒carbazole conjugates (ligands 1-5) to
bind G-quadruplexes formed by the telomere and oncogene promoter sequences (c-MYC, c-
KIT1, c-KIT2, VEGF and BCL2), fluorescence titration experiments were carried out. In these
8

ACCEPTED MANUSCRIPT
experiments, a fixed concentration of a specified ligand was titrated against increasing
concentrations of the indicated G-quadruplex. All the ligands exhibited very weak emissions
after excitation at 330 nm, which could be due to the collisional quenching of the singlet-excited
state by polar water molecules. The ligand 1 alone displayed an extremely weak fluorescence
emission (Fig. 2A). The addition of increasing concentrations of telomere G-quadruplex resulted
in a substantially enhanced emission (maximum at 462 nm). In parallel, the ability of ligand 1 to
bind oncogene promoter (c-MYC, BCL2, c-KIT1, c-KIT2, and VEGF) G-quadruplexes was
measured. The fluorescence intensity increased significantly as the ligand was titrated with each
of the G-quadruplex DNA substrates derived from the indicated oncogene promoter G-rich
sequences (Fig. 2B-2F).
Similarly, fluorescence titration experiments were performed with ligands 2-5 and the
results corroborated the effect of ligand 1. The influence of ligands 2-5 on the fluorescence
emission spectra was evident from the differences in emission profiles upon addition of
increasing concentrations of the oncogene promoter G-quadruplex DNA, which was followed by
saturation for each ligand (supplementary Fig. S1-S4). The enhancement in fluorescence
intensity suggests that the ligands encounter an entirely hydrophobic environment upon binding,
which in turn results in the retention of their intrinsic fluorescence emissions. Collectively, these
fluorescence titration measurements support the idea that benzimidazole-carbazole conjugates
interact with both telomere and oncogene promoter (c-MYC, c-KIT1, c-KIT2, VEGF, and BCL2)
G-quadruplexes.
9

ACCEPTED MANUSCRIPT
(
A)
(B)
(C)
6
1.6x10
c-MYC
BCL2
5
5
5
5
6
6
6
5
5
8.0x10
6.0x10
4.0x10
2.0x10
Telomere
2.0x10
1.6x10
1.2x10
6
5
5
1
.2x10
8.0x10
4.0x10
8
.0x10
.0x10
4
0
.0
0.0
400
0.0
400
4
00
440
480
520
560
440
480
520
560
440
480
520
560
Wavelength (nm)
Wavelength (nm)
Wavelength (nm)
(
D)
(E)
(F)
c-KIT1
VEGF
c-KIT2
6
6
5
5
5
6
6
5
1.6x10
6
4
2
.0x10
.0x10
.0x10
2.4x10
1.2x10
8.0x10
4.0x10
5
5
1.6x10
8.0x10
0
.0
0
.0
0.0
400
4
00
440
480
520
560
440
480
520
560
4
00
440
480
520
560
Wavelength (nm)
Wavelength (nm)
Wavelength (nm)
Figure. 2. Interaction of ligand 1 with different G-quadruplexes. The reaction mixtures
contained a fixed amount (0.6 µM) of the indicated ligand in 10 mM Tris-HCl buffer (pH 7.4)
with 100 mM KCl, 0.1 mM EDTA and increasing concentration of the specified G-quadruplex
(as indicated in each panel). The fluorescence spectral tracings were obtained at each time point
after the addition of an aliquot of G-quadruplex (0.15 µM). Panels A, B, C, D and E corresponds
to telomeric, c-MYC, BCL2, c-KIT1, c-KIT2 and VEGF G-quadruplex substrates, respectively.
The vertical arrow represents relative increase in fluorescence intensity following addition of the
specified G-quadruplex.
2
.3. Effect of benzimidazole-carbazole conjugates on the conformation of G-quadruplexes
The foregoing studies indicated that benzimidazole-carbazole conjugates interact with
telomere and oncogene promoter G-quadruplexes. Given the fact that G-quadruplexes can adopt
various folding topologies [1-2,6-7], we sought to assess the effect of benzimidazole-carbazole
1
0

ACCEPTED MANUSCRIPT
conjugates on the topology of oncogene promoter G-quadruplexes. Far-UV CD spectroscopy has
been extensively used to monitor the different conformations of G-quadruplex structures in
solution [61]. Thus, CD spectroscopy was employed in the current study to investigate the effect
of different benzimidazole-carbazole ligands on the conformation of G-quadruplexes.
The reaction mixtures (in 10 mM Tris-HCl buffer, pH 7.4, containing 5 µM G-quadruplex
DNA, 100 mM KCl and 0.1 mM EDTA; with or without different ligands) were heated at 95 °C
for 5 min followed by slow cooling to room temperature. Then the spectra were recorded in the
absence or presence of the indicated ligands. The c-MYC, c-KIT1, c-KIT2 and VEGF promoter
G-rich sequences consistently folded into parallel-stranded while the telomere and BCL2
sequences folded into hybrid-type mixed parallel/antiparallel G-strands. The addition of ligand 1,
+
4
or 5 to the G-quadruplex samples led to a structural inversion from the K stabilized hybrid-
type mixed parallel/antiparallel G-quadruplex to a stable telomere parallel-stranded G-
quadruplex. The positive peaks at 288 nm and 267 nm merged to produce a positive peak at 264
nm and a negative peak at 241 nm, indicating the formation of parallel-stranded G4 DNA (Fig.
3
A).
The telomeric G-rich sequence folds into a hybrid structure under physiological conditions;
at the same time, it folds into a parallel-stranded conformation under molecular crowding
conditions [62]. The ligands 2 and 3 failed to induce a structural transition from hybrid-type
mixed parallel/antiparallel G-quadruplex to a parallel-stranded structure in the telomere G-
quadruplex. A similar effect was seen in the case of the BCL2 promoter G-quadruplex structure
(
Fig. 3C). It is interesting to note that the ligands 1, 2, 3, 4 or 5 induced changes in the ellipticity
of human c-MYC, c-KIT1, c-KIT2 and VEGF promoter G-quadruplexes; however, the parallel-
stranded topology was retained (Fig. 3B, 3D-3F). These results suggest that these ligands
1
1

ACCEPTED MANUSCRIPT
preferentially bind to the parallel-stranded G-quadruplexes formed by G-rich promoter sequences
derived from human c-MYC, c-KIT1, c-KIT2 and VEGF oncogenes. Based on these observations,
we propose that the benzimidazole-carbazole ligands induce and stabilize parallel-stranded G-
quadruplex topologies. Moreover, the piperazine residues at the end of the benzimidazole
scaffold (ligand 1, 4, and 5) play an important role in the structural alteration of hybrid G-
quadruplex structures into parallel ones. Interestingly, replacement of the piperazine residues
with morpholine (ligand 2 and 3) disable the molecules to bring about any such transformational
changes.
(
A)
(B)
(C)
1
5
0
5
0
5
1
5
Telomere G4 alone
Telomere G4 + 1
Telomere G4 + 2
Telomere G4 + 3
Telomere G4 + 4
Telomere G4 + 5
c-MYC G4 alone
c-MYC G4 + 1
c-MYC G4 + 2
c-MYC G4 + 3
c-MYC G4 + 4
c-MYC G4 + 5
BCL2 G4 alone
BCL2 G4 + 1
BCL2 G4 + 2
BCL2 G4 + 3
BCL3 G4 + 4
BCL2 G4 + 5
1
0
5
0
5
1
10
5
0
-
-5
-
250
300
350
400
250
300
Wavelength (nm)
350
400
250
300
350
400
Wavelength (nm)
Wavelength (nm)
(
D)
(E)
(F)
1
5
0
5
0
5
15
10
5
c-KIT1 G4 alone
15
VEGF G4 alone
c-KIT2 G4 alone
c-KIT1 G4 + 1
c-KIT1 G4 + 2
c-KIT1 G4 + 3
c-KIT1 G4 + 4
c-KIT1 G4 + 5
VEGF G4 + 1
VEGF G4 + 2
VEGF G4 + 3
VEGF G4 + 4
VEGF G4 + 5
c-KIT2 G4 + 1
c-KIT2 G4 + 2
c-KIT2 G4 + 3
c-KIT2 G4 + 4
c-KIT2 G4 + 5
1
1
0
5
0
5
0
-
-5
-
250
300
350
400
250
300
350
400
250
300
350
400
Wavelength (nm)
Wavelength (nm)
Wavelength (nm)
Figure. 3. The effect of benzimidazole-carbazole ligands on the conformation of different G-
quadruplex substrates. The CD spectra were obtained from reaction mixtures containing 5 µM of
the indicated G-quadruplex in 10 mM Tris-HCl buffer (pH 7.4) with 100 mM KCl, 0.1 mM
EDTA in the presence or absence of different ligands. Inset: black, red, blue, dark cyan, pink and
olive CD spectral tracings correspond to G-quadruplex in the absence or presence of ligands 1, 2,
1
2

ACCEPTED MANUSCRIPT
3
, 4 and 5, respectively. Panels A, B, C, D, E, and F correspond to telomeric, c-MYC, BCL2, c-
KIT1, c-KIT2 and VEGF G4 DNA substrates respectively.
2
.4. Fluorescent intercalator displacement assay
The fluorescent intercalator displacement (FID) assay has been previously used to
determine the specificity and affinity of putative G4 DNA ligands [1-2,6-7]. Thiazole orange
TO), a fluorescent DNA-binding ligand, binds to G-quadruplexes with high affinity; it does so
(
in a manner consistent with binding at a single site, leading to a 500 to 3000-fold increase in
fluorescence emission [65]. The enhancement in fluorescence emission appears to depend on the
topology of G-quadruplex that adopted by a given G-rich sequence. The affinity of a given
ligand for G-quadruplex is expressed in terms of DC values, which represents the concentration
5
0
of the ligand required to reduce the fluorescence intensity by 50%. A low DC value suggests
5
0
high binding affinity and vice versa.
We have previously shown that benzimidazole-carbazole conjugates display lower DC50
values for telomeric G-quadruplexes compared to duplex DNA, suggesting higher affinity of the
ligands towards G-quadruplex structures [42]. These findings were further validated by
determining the concentrations of ligands 1-5 required to reduce the fluorescence intensity of
oncogene promoter (c-MYC, c-KIT1, c-KIT2, VEGF, and BCL2) G-quadruplexes (Fig. 4;
supplementary Fig. S5-S7 and Table 2). The ligand 1 showed higher affinity towards c-MYC, c-
KIT1, c-KIT2, and VEGF with DC50 values of 1.63 µM, 1.62 µM, 1.36 µM and 1.42 µM
respectively (Table 2). However, a higher concentration of ligand 1 was required to displace 50%
of TO from telomeric and BCL2 G-quadruplexes. The DC50 values for ligand 1 were 2.79 µM
and 2.86 µM in the case of telomeric and BCL2 G-quadruplexes, respectively. To assert the
1
3

ACCEPTED MANUSCRIPT
generality of the principle, DC50 values were determined for ligands 2, 4 and 5. The ligands 2, 4
and 5 displayed lower DC values for c-MYC, c-KIT1, c-KIT2 and VEGF over telomeric and
5
0
BCL2 G-quadruplexes (Table 2). Notably, ligands 2, 4 and 5 displayed higher efficiency
compared to ligand 1, indicating higher affinity for G4 DNA. This higher affinity may be due to
the presence of the extra protonated side chain coupled to the carbazole N-centre. However,
higher DC50 values for ligand 3 indicates that the existence of an additional protonatable
morpholine moiety along the carbazole N-side chain energetically disfavors the G-quadruplex
binding activity. Importantly, all the ligands (1‒5) displayed increased binding affinity towards
parallel G-quadruplex structures (c-MYC, c-KIT1, c-KIT2, and VEGF) over hybrid-type
structures (telomere and BCL2). The standard G-quadruplex DNA binding molecule TMPyP4
showed higher affinity towards all the G-quadruplex DNA than the carbazole-benzimidazole
ligands (supplementary Fig. S8 and Table 2). It is important to note that, unlike carbazole-
benzimidazole ligands, TMPyP4 does not show any preference among different G-quadruplex
DNA topologies. Moreover, all the carbazole-benzimidazole ligands showed much higher
selectivity toward ds-DNA over G-quadruplex DNA than TMPyP4 (Table 2).
1
4

ACCEPTED MANUSCRIPT
(
(
(
A)
(B)
(C)
6
6
6
6
8
.0x10
.0x10
.0x10
6
6
6
6
6
6
c-MYC
BCL2
6
.0x10
.0x10
.0x10
6.0x10
Telomere
6
4
4
2
4.0x10
2.0x10
2.0x10
0
.0
0.0
0
.0
5
20 540 560 580 600 620 640 660
520 540 560 580 600 620 640 660
Wavelngth (nm)
5
20 540 560 580 600 620 640 660
Wavelngth (nm)
Wavelngth (nm)
D)
(E)
(F)
6
6
6
5
c-KIT2
6
6
3x10
2x10
1x10
2.25x10
1.50x10
7.50x10
3.00x10
c-KIT1
VEGF
6
6
1
.50x10
0
.00
0
0.00
5
20 540 560 580 600 620 640 660
5
20 540 560 580 600 620 640 660
520 540 560 580 600 620 640 660
Wavelngth (nm)
Wavelngth (nm)
Wavelngth (nm)
G)
100
80
60
40
20
0
Telomere G4
c-MYC G4
BCL2 G4
c-KIT1 G4
c-KIT2 G4
VEGF G4
0
2
4
6
8
10 12 14
Ligand
(
mM)
Fig. 4. Thiazole orange (TO) displacement by ligand 1 from telomeric, c-MYC, BCL2, c-KIT1, c-
KIT2 or VEGF G-quadruplex-TO adduct. The fluorescence spectra of TO/G-quadruplex was
obtained in a buffer containing 10 mM Tris-HCl (pH 7.4), 100 mM KCl, and 0.1 mM EDTA in
the presence of increasing concentrations of ligand 1. The fluorescence spectral tracings were
recorded at each time point after the addition of an aliquot of a ligand (0.15 µM). Panels A, B, C,
D, E, and F correspond to telomeric, c-MYC, BCL2, c-KIT1, c-KIT2 and VEGF G4 DNA
substrates respectively. Panel G shows relative TO displacement for telomeric, c-MYC, BCL2, c-
1
5

ACCEPTED MANUSCRIPT
KIT1, c-KIT2 and VEGF G-quadruplex substrates as a function of increasing concentration of the
specified ligand (as indicated in each panel).
Table 2. DC values (ligand concentrations for the displacement of 50% of the TO from G4
50
DNA-TO complex) for the different G-quadruplexes as determined from fluorescence
a
intercalator displacement assays .
DC
(µM)
5
0
Ligand
Telomere
2.79
c-MYC
BCL2
c-KIT1
c-KIT2
VEGF
ct-DNA
1
1.63
0.85
>3
2.86
1.29
>3
1.62
0.98
>3
1.34
0.74
>3
1.42
0.63
>3
5.28
5.52
>10
2
1.30
3
>3
4
5
1.15
0.98
0.83
0.38
1.44
1.39
0.43
0.88
0.86
0.32
0.66
0.65
0.35
0.75
0.78
0.31
6.49
6.94
0.87
1.13
TMPyP4
0.47
a
The results are the average of two independent experiments and estimated error values are
within ± 5% of each other.
2
.5. Thermal stability of telomere and oncogene promoter G-quadruplexes
The data from fluorescence titrations and the FID assay suggest that ligands 1-5 bind
efficiently to both telomeric and oncogene promoter (c-MYC, c-KIT1, c-KIT2 and VEGF) G-
quadruplexes. To further validate these observations, we examined the thermal stability (∆T ) of
m
telomeric and oncogene promoter (c-MYC, c-KIT1, c-KIT2 and VEGF) G-quadruplexes as a
function of ligands 1-5 using CD spectroscopy. Toward this end, the samples (in 10 mM Tris-
HCl buffer, pH 7.4) containing 5 µM G-quadruplex DNA, 25 mM KCl and 0.1 mM EDTA and
different ligands were heated followed by slow cooling to room temperature. The variable
1
6

ACCEPTED MANUSCRIPT
temperature CD spectra were recorded on a Jasco J-810 spectropolarimeter. Since
benzimidazole‒carbazole conjugates induce a parallel topology in telomeric G4 DNA, melting
studies were carried out at 264 nm.
The Fig. 5A shows thermal melting profiles for telomeric G4 DNA in the absence or
presence of ligands 1-5. In the absence of any ligand, the telomeric G-quadruplex showed a Tm
of 55.3 °C. In the presence of ligands 1-5, the T increased up to 21.2 °C. As shown in Table 3,
m
the ∆T
m
values are quite high for ligands 2, 4 and 5 at 16.8 °C, 21.1 °C and 21.2 °C respectively.
values are lower for ligands 1 and 3 at 9.7 °C and 7.1 °C respectively (Table 3). Thus,
The ∆T
m
these experiments reveal that ligands 1-5 confer thermal stability on the telomeric G-quadruplex
structure.
Next, we investigated the effect of ligands on stabilization of oncogene promoter G-
quadruplexes. As these promoter G-quadruplexes are known to adopt a parallel topology in the
+
presence of K ion, the melting studies were followed at 262 nm. The CD melting measurements
showed significant thermal stabilization of promoter G-quadruplex DNAs in the presence of each
ligand (Fig. 5B-5F). The thermal stability provided by ligands 1-5 to promoter G-quadruplexes
m
followed the order c-KIT1 ≈ c-KIT2 > VEGF > c-MYC ≈ BCL2. The ∆T values for various G-
quadruplexes are in the range 6.5‒17 °C for ligand 1; 5.6-16.8 °C for ligand 2; 3.3‒7.5 °C for
ligand 3; 8.9‒25.9 °C for ligand 4 and 7.4‒24.7°C for ligand 5 (Table 3). The ligands 1, 2, and 3
provided slightly less thermal stabilization compared to ligands 4 and 5. However, there was no
clear order of magnitude in the stabilization provided by various ligands to different G-
quadruplexes. The ligands 4 and 5, having three protonatable side chains with either piperazine
or dimethylamine residues, showed enhanced stabilization compared to ligand 1 (having only
two side chains). The ligand 2, with two morpholine residues, exhibited moderate stabilization
1
7

ACCEPTED MANUSCRIPT
while compound 3 (containing an additional morpholine moiety that, however, blocks the
piperazine side-chain) exhibited a much weaker stabilizing ability and selectivity for all the
investigated quadruplexes. The standard G-quadruplex DNA binder TMPyP4 conferred much
lower thermal stability to G-quadruplex DNA than ligand 4 or 5.
(
A)
(B)
(C)
c-MYC G4 alone
c-MYC G4 + 1
c-MYC G4 + 2
c-MYC G4 + 3
c-MYC G4 + 4
c-MYC G4 + 5
c-MYC G4 + TMPyP4
BCL2 G4 alone
BCL2 G4 + 1
Telomere G4 alone
Telomere G4 + 1
Telomere G4 + 2
Telomere G4 + 3
Telomere G4 + 4
Telomere G4 + 5
Telomere G4 + TMPyP4
1.0
0.8
0.6
0.4
0.2
0.0
1.0
0.8
0.6
0.4
0.2
0.0
1.0
BCL2 G4 + 2
BCL2 G4 + 3
BCL2 G4 + 4
0.8
BCL2 G4 + 5
BCL2 G4 + TMPyP4
0.6
0.4
0.2
0.0
30
40 50 60 70 80 90 100
30 40 50 60 70 80 90 100
30 40 50 60 70 80 90 100
O
O
O
Temperature
(
C
)
Temperature
( )
C
Temperature
( )
C
(
D)
(E)
(F)
KIT2 G4 alone
KIT2 G4 + 1
VEGF G4 alone
VEGF G4 + 1
KIT1 G4 alone
KIT1 G4 + 1
1
0
0
0
0
0
.0
KIT2 G4 + 2
VEGF G4 + 2
KIT1 G4 + 2
1.0
1.0
0.8
0.6
0.4
0.2
0.0
KIT2 G4 + 3
VEGF G4 + 3
KIT1 G4 + 3
.8
.6
.4
.2
.0
KIT1 G4 + 4
KIT2 G4 + 4
VEGF G4 + 4
0.8
0.6
0.4
0.2
0.0
KIT2 G4 + 5
VEGF G4 + 5
KIT1 G4 + 5
KIT2 G4 + TMPyP4
VEGF G4 + TMPyP4
KIT1 G4 + TMPyP4
30
40 50 60 70 80 90 100
O
30
40 50 60 70 80 90 100
O
30 40 50 60 70 80 90 100
O
Temperature
(
C)
Temperature
( )
C
Temperature
( C)
Fig. 5. Benzimidazole‒carbazole ligands stabilize telomeric, c-MYC, BCL2, c-KIT1, c-KIT2 and
VEGF G-quadruplex structures. Thermal denaturation studies of G-quadruplex in the absence or
presence of ligands 1, 2, 3, 4, 5 or TMPyP4. Black, red, blue, dark cyan, pink, olive and dark
blue spectral tracings correspond to the thermal denaturation profiles in the absence and presence
of ligand 1, 2, 3, 4, 5 and TMPyP4 respectively. Panels A, B, C, D, E, and F correspond to
telomeric, c-MYC, BCL2, c-KIT1, c-KIT2 and VEGF G-quadruplex substrates, respectively.
1
8

ACCEPTED MANUSCRIPT
Table 3. Thermal stability of various G-quadruplex DNA structures with different
a
benzimidazole‒carbazole conjugates measured by CD melting studies .
∆Tm (°C)
Ligand
Telomere
c-MYC
BCL2
c-KIT1
c-KIT2
VEGF
1
9.74
8.63
8.65
4.95
8.86
7.35
10.97
6.53
5.54
3.26
9.50
9.54
14.27
13.18
12.20
3.58
17.00
16.55
7.53
12.79
14.18
4.35
2
16.86
7.14
3
4
5
21.17
21.26
15.63
25.88
24.69
15.14
21.96
23.27
15.54
14.15
11.28
12.35
TMPyP4
a
∆
Tm values were obtained from the differences in the melting temperatures of the ligand bound
and G4 DNA alone in 10 mM Tris−HCl buffer (pH 7.4) containing 25 mM KCl and 0.1 mM
EDTA. The results are the average of two independent experiments and are within ± 0.5 °C of
each other.
Our previous studies with telomeric G-quadruplex showed relatively weaker stabilization by
ligands in the presence of NaCl compared with G-quadruplex stabilized by KCl [42, 43]. In the
+
presence of Na ion, the ligands failed to bring about a conformational change in telomeric G4
+
DNA from antiparallel basket to parallel type; however, in the case of K ion, the ligands bring
about a conformational change and they favor parallel conformation. Thus, the higher
+
denaturation temperatures in the presence of K ion are most likely due to the topological
transformation from hybrid to a more stable parallel structure. This suggests that the parallel
form of G-quadruplexes meets the structural requirements for maximum non-covalent
interactions with the ligands. Collectively, the results from CD melting studies suggest that
ligands 1-5 preferentially stabilize parallel the G-quadruplex conformation.
1
9

ACCEPTED MANUSCRIPT
2
.6. Polymerase stop assay
Having established that ligands 1-5 interact with and stabilize promoter G-quadruplex
structures, we asked whether such stabilized G-quadruplexes present obstacles to DNA synthesis
by DNA polymerases. For this purpose, a 77-mer c-MYC G-quadruplex forming sequence was
3
2
annealed to P-labeled primers through heating followed by slow cooling in 10 mM Tris-HCl
buffer (pH 7.4) containing 5 mM KCl. The concentration of the primer used for annealing was
1
.5-fold higher than the template and targeted towards a complementary region at the 3′ end of
the template. The primer extension reaction was initiated by the addition of Taq DNA
polymerase. The reaction was performed at 48 °C for 30 min. As expected, the reaction
performed in the absence of any added ligand generated a full-length product (Fig. 6A, lane 2).
In theory, if the ligand stabilizes a G-quadruplex structure in the c-MYC template, the
primer extension by Taq DNA polymerase would pause at the G-quadruplex-forming (pause)
site. On adding increasing concentrations of ligand 5, enhanced pausing was observed at the site
coinciding with the formation of G-quadruplex in the 77-mer c-MYC G-rich promoter sequence
(
Fig. 6A, lanes 3 to 6). These results suggest that ligand 5 enhances Taq DNA polymerase
pausing by stabilizing the G-quadruplex structure. At high concentrations of ligand 5, there was
increased premature termination resulting in complete inhibition of the formation of full-length
products.
2
0

ACCEPTED MANUSCRIPT
(A)
(B)
Ligand 5
2.5
Ligand 5
2.5
(L/Template)
0
0.5
1
5
(L/Template)
0
0.5
1
5
Full length
product
Pause
product
Primer
1
2
3
4
5
6
1
2
3
4
5
6
Figure. 6. Taq DNA polymerase stop assay performed on the c-MYC DNA template sequence in
the presence of increasing concentrations ligand 5 and 1. The position of primer, full-length
product and the pause product is indicated on the right-hand side. (A) Primer extension reaction
in the presence of ligand 5 at 48 °C. (B) Primer extension reaction in the presence of ligand 1 at
4
0 °C.
To further substantiate these findings, Taq DNA polymerase primer extension reactions
were carried out using ligand 1 at 40 °C. As shown in Fig. 6B (lane 1), in the absence of ligand 1,
a small amount of ligand-independent DNA synthesis arrest was seen at the guanine tract. In
contrast, ligand 1 generated the same premature chain termination products as ligand 5,
confirming that these ligands are responsible for blocking DNA synthesis. Furthermore, the
extent of premature chain termination observed was directly proportional to the concentration of
ligand added to the reaction (Fig. 6B, lanes 3-6). The pattern and extent of ligand-dependent
arrest of DNA synthesis is consistent with the notion of stabilization of G-quadruplex DNAs
formed by oncogene promoter G-rich sequences.
2
1

ACCEPTED MANUSCRIPT
2
.7. Computational studies
To investigate the most probable modes of binding of these ligands to various G-
quadruplex structures at the telomeric and oncogene promoter regions, we performed docking
studies using the AutoDock 4.0 software [64]. In addition, the extent of binding of these ligands
to various G4 DNAs was determined, thus substantiating the biophysical results described above.
For this purpose, two representative benzimidazole‒carbazole conjugates were taken for docking
studies namely: ligand 1 and 2. Initially, the energy optimization of ligands 1 and 2 was
performed and the most stable energy state structures were obtained. These were then docked
with the structures of human c-KIT (2O3M), c-KIT2 (2KQH), VEGF (2M27), HTelo (1KF1), c-
MYC (1XAV), and BCL2 (2F8U) G-quadruplexes (Fig. 7). Both the ligands were found to
interact with different G-quadruplexes through efficient stacking and/or H-bond formation.
Primarily the central benzimidazole‒carbazole pharmacophore interacted with the planar G-
tetrad; the side chains were found to interact with the negatively charged phosphate backbone in
the groove and loop region. Taken together, these results corroborate the spectroscopic analysis
and DNA synthesis arrest studies showing that ligands 1 and 2 efficiently interact with the G-
quadruplex structures formed by both telomere and oncogene promoter G-rich sequences.
2
2

ACCEPTED MANUSCRIPT
(A)
(C)
(E)
(B)
I
I
I
II
II
II
I
I
I
II
II
II
(D)
(F)
Figure. 7. The docked models of ligands 1 and 2 with (A) telomeric, (B) c-MYC, (C) BCL2, (D)
c-KIT1, (E) c-KIT2 and (F) VEGF G-quadruplex structures. PDB code of the telomeric, c-MYC,
BCL2, c-KIT1, c-KIT2 and VEGF G-quadruplex structure used for the docking are 1KF1, 1XAV,
2
F8U, 2O3M, 2KQH, and 2M27 respectively. Panel I represent docking with ligand 1 and
subpanel II with ligand 2.
2
3

ACCEPTED MANUSCRIPT
2
.8. Benzimidazole‒carbazole ligands inhibit telomerase activity and oncogene expression,
which in turn shows cancer cell specific cytotoxicity
The foregoing studies demonstrate that telomeric and oncogene promoter G-quadruplex
structures can be stabilized by benzimidazole-carbazole ligands. Next, we sought to investigate
their ability to inhibit telomerase activity; to bind and repress the transcription of target oncogene
promoters. To this end, the effect of ligand 5 on telomerase activity was tested using a modified
telomeric repeat amplification assay (TRAP-LIG) [65]. Here, ligand 5 was used in the
concentration range of 0.01 to 2 µM. As shown in Fig. 8A, it inhibited telomerase activity in a
dose-dependent manner, and this was correlated with the decreased intensity and number of
telomeric bands with an IC value of 0.3 µM. We wondered whether ligand 5 can suppress the
5
0
expression of c-MYC and VEGF gene products in cell models. Accordingly, we analyzed their
transcriptional response to ligand 5 in MCF-7 and HeLa cells using qRT-PCR assay. The
expression of the housekeeping gene GAPDH served as a control. We found a dose-dependent
decrease in c-MYC and VEGF transcript levels with no alteration in GAPDH transcription. The
HeLa cells showed 49% and 71% decrease in c-MYC transcription upon treatment with 5 µM and
1
0 µM ligand 5, respectively; however, no corresponding reduction was seen in cells treated with
DMSO (Fig. 8B). Similarly, the VEGF transcript levels were reduced by 44% and 68% upon
treatment with 5 µM and 10 µM of ligand 5 respectively.
Next, we evaluated whether these ligands exhibit anti-proliferative effect against various
cancer cell lines. The cytotoxic efficacy of ligand 5 against human breast cancer cells (MCF-7)
and human cervical cancer (HeLa) cells was carried out using MTT assay. The results showed
that ligand 5 significantly inhibited the proliferation of cancer cell lines in a concentration-
dependent manner with IC50 values of 11.42 µM and 8.29 µM respectively (Fig. 8C). Thus, these
2
4

ACCEPTED MANUSCRIPT
results support the idea that benzimidazole-carbazole ligands could potentially be used for
suppression of oncogene expression in cancer cells, and as suitable candidates in cancer
chemotherapy.
(
A)
+
(B)
(C)
0.01-2 µM
ve -ve
c-MYC
VEGF
1.2
0.8
0.4
0.0
120
00
MCF-7
HeLa
1
8
6
4
2
0
0
0
0
0
Compound 5 (µM)
Compound 5 (µM)
1
2 3 4 5 6 7 8 910111213
Figure 8. Benzimidazole‒carbazole ligand 5 inhibits telomerase activity, represses oncogene
expression and displays cancer cell specific cytotoxicity. (A) Ligand 5 inhibits telomerase
activity in a dose-dependent manner. Telomerase inhibition of ligand 5 was evaluated by
TRAP–LIG assay. Lane 1, positive control; 2, negative control. Lanes 3-13, incubated with
increasing concentrations of ligand 5 (0.01, 0.02, 0.04, 0.08, 0.1, 0.3, 0.6, 0.9, 1.2, 1.5, 2 µM
respectively). (B) Ligand 5 suppresses c-MYC and VEGF transcription in HeLa cells. The HeLa
cells were treated with increasing concentrations of ligand 5 for 24 h and then the transcript
levels of c-MYC and VEGF were measured. The histograms show relative fold change in the c-
MYC and VEGF transcript levels upon ligand treatment. GAPDH was used as an internal control
for normalization. (C) Ligand 5 decreases cell viability and induces cell death. Cytotoxic effect
of ligand 5 in MCF-7 and HeLa cells was evaluated by the MTT assay. The error bars represent
the mean ± SD of three independent experiments performed in triplicate
2
5

ACCEPTED MANUSCRIPT
3
. Conclusions
The biological significance of G-quadruplexes has been well-recognized and they have
emerged as attractive candidates for cancer therapy. The interest in telomere-specific therapeutic
compounds continues, in parallel, there is growing interest in oncogene promoter G-
quadruplexes as potential therapeutic targets to design a new effective anticancer drug. In
contrast to the telomeric G-quadruplex forming sequence, the G-quadruplex-forming sequences
in promoter regions are diverse and can fold into multiple G-quadruplex conformations. Thus,
the discovery and development of small molecules that can interact with genomic G-quadruplex
DNA structures may provide necessary opportunities for targeting promoter sequences and
downstream events. A ligand that stabilizes one structure may interact differently with others.
Therefore, it is necessary to profile the interaction of ligands with different G-quadruplexes to
determine their biological activity as G-quadruplex stabilizing compounds.
Our previous studies have demonstrated that benzimidazole-carbazole ligands preferentially
bind to and stabilize telomeric G-quadruplexes over double-stranded DNA and, consequently,
inhibit telomerase activity [42]. Additionally, these ligands exhibit selectivity towards parallel G-
quadruplex structures. In the current study, benzimidazole-carbazole ligands were found to bind
and stabilize human c-MYC, c-KIT1, c-KIT2, VEGF and BCL2 promoter G-quadruplex structures
in vitro. The spectroscopic experiments suggest that benzimidazole-carbazole ligands induce
topological changes from hybrid to stable parallel G-quadruplex DNAs. Consistent with the
experimental data on thermal stability of G-quadruplexes and FID assays, these ligands
displayed higher affinity to oncogene promoter G-quadruplexes. Furthermore, studies with
ligands 2, 4 and 5 indicate that these are strong oncogene promoter G-quadruplex binding
2
6

ACCEPTED MANUSCRIPT
ligands compared to ligands 1 and 3. Structural modifications among ligands 1‒5 strongly
suggest that the hydroxyethyl substituted bis-benzimidazole residues are responsible for
topological transformation of the G-quadruplex DNA structure. The additional protonatable
residue at the carbazole N-centre (ligand 2, 4, and 5) enhances G-quadruplex DNA affinity
without altering the specificity toward parallel topological structure. Notably, the stabilization of
promoter G-quadruplexes by benzimidazole‒carbazole ligands stall DNA replication and inhibit
oncogene expression in vivo. We close with a point emphasizing the notion that the
transcriptional activation of oncogenes in cancer cells can be suppressed through the stabilization
of G-quadruplexes by benzimidazole‒carbazole ligands.
4
. Experimental section
4
.1 Synthesis of benzimidazole‒carbazole ligands
We designed and synthesized ligand 1, 3 and 5 (scheme 1-3) as previously described [42].
All starting materials were from the commercial sources and used without further purification.
All solvents were from Merck, and they were distilled and/or dried prior to use whenever
necessary. All ligands were found to be at least >95% pure by elemental analysis.
2
7

ACCEPTED MANUSCRIPT
Scheme 1. Reagents, conditions and yields: a) dry DMF, K CO , N -atomosphere, 110 °C, 12 h,
2
3
2
9
5%, b) H /Pd-C, rt, 12 h, 100%, c) NaH, ethylbromide, TBAI, DMF, rt, 12 h, 80%, d) POCl ,
2 3
DMF, ZnCl , 100 °C, 24 h, 75%, (e) Compound 7, Na S O , ethanol, 80 °C, 24 h, 60%.
2
2
2
5
2
8

ACCEPTED MANUSCRIPT
Scheme 2. Reagents, conditions and yields: a) dry DMF, K CO , N -atm. 110 °C, 12 h, 95%, b)
2
3
2
H /Pd-C, room temperature, 12 h, 100%, c) 1, 4-Dibromobutane, acetonitrile, 50 °C, 6 h, 90%;
2
d) Morpholine, room temperature, 12 h, 98%, e) (12), Na S O , EtOH, 80 °C, 12 h, 85%.
2
2
5
2
9