Copper(II) complexes of substituted salicylaldehyde dibenzyl semicarbazones: Synthesis, cytotoxicity and interaction with quadruplex DNA

Article abstract of DOI:10.1039/c3dt52297k

A series of substituted salicylaldehyde dibenzyl semicarbazones [RC ₆H₃(OH)CH=N-NHCON(CH₂Ph)₂] and their copper(II) complexes were synthesized and characterized. The chloridocopper(II) complexes of the 4-OH and 5-OH substituted ligands (complexes 9 and 7) show modest affinity and good selectivity (over duplex DNA) for the quadruplex formed from the 22AG human telomeric (HTelo) DNA sequence. Substitution of the chlorido ligands of these two complexes with pyridine yielded derivatives (7-py and 9-py) with increased affinity for HTelo. These derivatives also show good selectivity for HTelo over calf-thymus DNA (170- and 211-fold, respectively). The X-ray crystal structures of 9 and 9-py were determined. Molecular docking studies based on these structures show that the complexes stack on the 5′-end of the HTelo quadruplex, with the hydroxyl group forming a hydrogen bond with a guanine residue. Complexes 7, 9, 7-py and 9-py display significant cytotoxicity against MOLT-4 human leukaemia cells. Interestingly, they have low to negligible cytotoxicity against the non-cancerous IMR-90 human fibroblasts. The Royal Society of Chemistry 2014.

Full text of DOI:10.1039/c3dt52297k

View Article Online
View Journal
Dalton
Transactions
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: S. Munira-Haidad-Ali, Y. K. Yan, P. P. F.
Lee, K. Z. X. Khong, M. Alam Sk, K. H. Lim, B. Klejevskaja and R. Vilar, Dalton Trans., 2013, DOI: 10.1039/C3DT52297K.
This is an Accepted Manuscript, which has been through the RSC Publishing peer
review process and has been accepted for publication.
Dalton
Transactions
Volume 39 | Number 3 | 21 January 2010 | Pages 657–964
Accepted Manuscripts are published online shortly after acceptance, which is prior
An international journal of inorganic chemistry
www.rsc.org/dalton
to technical editing, formatting and proof reading. This free service from RSC
Publishing allows authors to make their results available to the community, in
citable form, before publication of the edited article. This Accepted Manuscript will
be replaced by the edited and formatted Advance Article as soon as this is available.
To cite this manuscript please use its permanent Digital Object Identifier (DOI®),
which is identical for all formats of publication.
More information about Accepted Manuscripts can be found in the
Information for Authors.
Please note that technical editing may introduce minor changes to the text and/or
graphics contained in the manuscript submitted by the author(s) which may alter
content, and that the standard Terms & Conditions and the ethical guidelines
that apply to the journal are still applicable. In no event shall the RSC be held
responsible for any errors or omissions in these Accepted Manuscript manuscripts or
any consequences arising from the use of any information contained in them.
ISSN 1477-9226
COMMUNICATION
PAPER
Bu et al.
Manzano et al.
Zinc(ii)-boron(iii)-imidazolate
Experimental and computational study framework (ZBIF) with unusual
of the interplay between C–H/p and
anion–p interactions
pentagonal channels prepared from
deep eutectic solvent
1477-9226(2010)39:1;1-K
www.rsc.org/dalton
Registered Charity Number 207890

Page 1 of 31
Dalton Transactions
View Article Online
DOI: 10.1039/C3DT52297K
Copper(II) Complexes of Substituted Salicylaldehyde Dibenzyl Semicarbazones:
Synthesis, Cytotoxicity and Interaction with Quadruplex DNA
Siti Munira Haidad Alia, YawꢀKai Yana,*, Peter P. F. Leea, Kenny Zhi Xiang Khonga,
Mahasin Alam Skb, Kok Hwa Limb, Beata Klejevskajac, Ramon Vilarc,
a Natural Sciences & Science Education, National Institute of Education, Nanyang
Technological University, 1 Nanyang Walk, Singapore 637616
b Division of Chemical & Biomolecular Engineering, School of Chemical & Biomedical
Engineering, Nanyang Technological University, Singapore
c Department of Chemistry, Imperial College London, London SW7 2AZ, UK
Abstract
A series of substituted salicylaldehyde dibenzyl semicarbazones [RC6H3(OH)CH=Nꢀ
NHCON(CH2Ph)2] and their copper(II) complexes were synthesized and characterized. The
chloridocopper(II) complexes of the 4ꢀOH and 5ꢀOH substituted ligands (complexes 9 and 7)
show modest affinity and good selectivity (over duplex DNA) for the quadruplex formed
from the 22AG human telomeric (HTelo) DNA sequence. Substitution of the chlorido
ligands of these two complexes with pyridine yielded derivatives (7-py and 9-py) with
increased affinity for HTelo. These derivatives also show good selectivity for HTelo over
calfꢀthymus DNA (170ꢀ and 211ꢀfold, respectively). The Xꢀray crystal structures of 9 and 9-
py were determined. Molecular docking studies based on these structures show that the
complexes stack on the 5′ꢀend of the HTelo quadruplex, with the hydroxyl group forming a
hydrogen bond with a guanine residue. Complexes 7, 9, 7-py and 9-py display significant
cytotoxicity against MOLTꢀ4 human leukaemia cells. Interestingly, they have low to
negligible cytotoxicity against the nonꢀcancerous IMRꢀ90 human fibroblasts.
Keywords: copper(II); semicarbazones; quadruplex DNA; cytotoxicity; molecular docking.
* Correspondence author. Email: yawkai.yan@nie.edu.sg; Fax: 65ꢀ6896ꢀ9414; Tel: 65ꢀ6790ꢀ
3813
1

Dalton Transactions
Page 2 of 31
View Article Online
DOI: 10.1039/C3DT52297K
1
Introduction
Gꢀquadruplex DNA structures are fourꢀstranded, guanineꢀrich DNA structures
containing coplanar square quartets of guanine residues held together by hydrogen bonding. 1ꢀ
3
Such guanineꢀrich sequences can exist in telomeres and in promoter regions of some
oncogenes in the human genome.4 The activity of telomerase, an enzyme that is crucial for
the immortalization of cancerous cells,5 can be inhibited if the single stranded telomeric DNA
is folded into a quadruplex structure.6 Formation of quadruplex structures in the promoter
regions of oncogenes like cꢀmyc and cꢀkit has been proposed to regulate the transcription of
these genes.7ꢀ11 Molecules that bind to, and stabilize, quadruplex DNA are therefore of
interest as potential anticancer agents.12ꢀ20
We have reported a square planar copper(II) complex of salicylaldehyde N,Nꢀdibenzyl
semicarbazone (1, Scheme 1) that is strongly cytotoxic against a range of human cancer cell
lines.21 This complex induces apoptosis on MOLTꢀ4 (human leukaemia) cells and
downregulates proteins that are associated with transcription, translation, protein folding,
glucose metabolism, signal transduction and mitosis.21 It also intercalates with duplex DNA
and cleaves it via the generation of reactive oxygen species.22 With its square planar
geometry and aromatic ligand, complex 1 potentially has the ability for quadruplex binding.
Hence, we decided to determine the binding strength of 1 and its derivatives (Scheme 1) for
quadruplex DNA, and their selectivity for quadruplex over duplex DNA. The results are
reported in this paper, together with the cytotoxicity of these compounds towards MOLTꢀ4
human leukaemia cells and IMRꢀ90 human fibroblasts (nonꢀcancerous).
2

Page 3 of 31
Dalton Transactions
View Article Online
DOI: 10.1039/C3DT52297K
Scheme 1
2
Experimental
2.1
Materials and apparatus
Chemical reagents, unless otherwise stated, were obtained commercially and were
used as received without any preꢀtreatment. All chemical reactions involving airꢀ and/or
moisture sensitive compounds were performed under nitrogen atmosphere using standard
Schlenk techniques. Solvents used for the reactions were of reagent grade and toluene,
dichloromethane and diethyl ether were distilled under nitrogen and vacuumꢀdegassed before
use. Elemental analyses (C, H and N) were performed using a Elementar Vario Micro Cube
elemental analyzer. Solidꢀstate infrared spectra (KBr pellet) were recorded using a Perkinꢀ
Elmer Spectrum 100 FTIR spectrometer of operating region 4000 to 400 cm−1. Proton NMR
spectra were recorded in (CD3)2SO on a Bruker DRX400 spectrometer at room temperature
at 400 MHz. Chemical shifts are quoted in ppm downfield of internal tetramethylsilane.
3

Dalton Transactions
Page 4 of 31
View Article Online
DOI: 10.1039/C3DT52297K
Fluorescence spectra were recorded using a Perkin Elmer L50 photoluminescence
spectrometer, while UVꢀvis spectra were recorded on a Perkin Elmer 25 UVꢀvis
spectrophotometer.
All oligonucleotides used were purchased from either AITBiotech (Singapore) or
Eurogentec S.A. (Belgium). The quadruplexꢀforming strands (5′ꢀAGGꢀGTTꢀAGGꢀGTTꢀ
AGGꢀGTTꢀAGGꢀGꢀ3′) and (5′ꢀGGꢀGAGꢀGGTꢀGGGꢀGAGꢀGGTꢀGGGꢀ3′) were used for
human telomeric (HTelo) and c-myc studies respectively. For the duplex DNA studies, the 17
baseꢀpair (ds17) complementary strands (5′ꢀCCAꢀGTTꢀCGTꢀAGTꢀAACꢀCCꢀ3′ and 5′ꢀGGGꢀ
TTAꢀCTAꢀCGAꢀACTꢀGGꢀ3′) and 26 baseꢀpair (ds26) selfꢀcomplementary strand (5′ꢀCAAꢀ
TCGꢀGATꢀCGAꢀATTꢀCGAꢀTCCꢀGATꢀTGꢀ3′) were used. Calf thymus DNA (ctꢀDNA) was
obtained from Rockland (MBꢀ102ꢀ0100). Oligonucleotide concentrations were calculated
using the following extinction coefficients (L cm−1 mol−1) at 260 nm given by the supplier:
228500 (HTelo), 206800 (cꢀmyc), 253200 (ds26), 160900 (ds17 strand 1), 167400 (ds17
strand 2). The extinction coefficient of ctꢀDNA at 260 nm was taken to be 13100 L cm−1 per
mole of nucleotide base pair.23
2.2
Syntheses
The synthesis procedures for the substituted salicylaldehyde dibenzyl semicarbazones
and their respective copper(II) complexes were adapted from previouslyꢀreported methods.24
2.2.1 Dibenzyl semicarbazide, (PhCH2)2NCONHNH2. Dibenzyl amine (960 ꢁL, 5.0
mmol) was dissolved in dry dichloromethane (8 mL) in a Schlenk tube. Triphosgene (0.519
g, 1.75 mmol) was dissolved in dichloromethane (8 mL) in a separate Schlenk tube and
pyridine (810 ꢁL, 10 mmol) was added to this solution. After placing a magnetic stirrer bar
into the triphosgene solution, the two flasks were stoppered tightly. The triphosgene solution
4

Page 5 of 31
Dalton Transactions
View Article Online
DOI: 10.1039/C3DT52297K
was placed in an ice bath on top of a magnetic stirrer. Using a cannula, the amine solution
was added dropwise to the triphosgene solution to yield an orange solution. The reaction
mixture was left to stand at room temperature for three days under nitrogen to give a yellow
solution of dibenzyl carbamic chloride.
To this solution was added diethyl ether (21 mL), and the resultant mixture was added
dropwise to a stirred solution of hydrazine monohydrate (970 ꢁL, 20 mmol) in ethanol (10
mL). The reaction mixture was stirred for 30 min before the solvents were removed under
reduced pressure. Dichloromethane (5 mL) was added, and the resultant mixture was washed
three times with water (5, 3, 3 mL). Evaporation of the organic layer under reduced pressure
gave the semicarbazide, which was redissolved in ethanol (1.7 mL) for reaction with the
substituted salicylaldehydes. For the synthesis of the 6ꢀOHꢀsubstituted semicarbazone, the
washed dichloromethane solution was dried with anhydrous sodium sulfate before the solvent
was removed, and the semicarbazide residue was redissolved in toluene (5 mL).
2.2.2 Substituted salicylaldehyde dibenzyl semicarbazones. For the salicylaldehyde
starting materials that were soluble in toluene (R = 5ꢀCl, 5ꢀBr, 5ꢀOMe), an equimolar amount
of substituted salicylaldehyde dissolved in 4 mL of toluene was added to the solution of the
semicarbazide. For the sparinglyꢀsoluble salicylaldehydes, 10 mL of toluene was added to the
semicarbazide solution, after which the solid salicylaldehyde was added in small portions
with vigorous stirring. The resultant mixture was stirred at room temperature for 30 min (12 h
for the 6ꢀOH derivative) to give a yellow suspension. A catalytic amount of pꢀ
toluenesulfonic acid was added if precipitation did not occur within the first 5 min of stirring.
The precipitate was isolated by filtration, washed with cold toluene, and dried in vacuo. All
the semicarbazones were recrystallised from acetoneꢀwater mixture.
5

Dalton Transactions
Page 6 of 31
View Article Online
DOI: 10.1039/C3DT52297K
C6H4(OH)CH=NNHCON(CH2Ph)2. Yield: (1.08 g, 3.01 mmol, 60%). Anal. Calcd. for
C22H21N3O2: C, 73.6; H, 5.9; N, 11.6. Found: C, 73.5; H, 5.9; N, 11.7. IR (cm−1): ν(C=O)
1
1645 vs. H NMR [(CD3)2SO, ppm]: 11.52 (1H, s, phenol OꢀH), 10.85 (1H, s, hydrazine Nꢀ
H), 8.35 (1H, s, imine CꢀH), 7.2ꢀ7.4 (12H, m, PhꢀH), 6.88 (2H, m, PhꢀH), 4.52 (4H, s, CH2).
(5-Cl)C6H3(OH)CH=NNHCON(CH2Ph)2. Yield: (1.18 g, 3.00 mmol, 60%). Anal. Calcd. for
C22H20N3O2Cl: C, 67.1; H, 5.1; N, 10.7. Found: C, 66.9; H, 5.3; N, 10.3. IR (cm−1): ν(C=O)
1
1644 vs. H NMR [(CD3)2SO, ppm]: 11.51 (1H, s, phenol OꢀH), 10.95 (1H, s, hydrazine Nꢀ
H) 8.31 (1H, s, imine CꢀH), 7.2ꢀ7.4 (12H, m, PhꢀH), 6.90 (1H, d, J= 9.1 Hz, PhꢀH), 4.52 (4H,
s, CH2).
(5-Br)C6H3(OH)CH=NNHCON(CH2Ph)2. Yield: (1.38 g, 3.15 mmol, 63%). Anal. Calcd. for
C22H20N3O2Br: C, 60.5; H, 4.5; N, 9.6. Found: C, 60.2; H, 4.2; N, 9.4. IR (cm−1): ν(C=O)
1
1643 vs. H NMR [(CD3)2SO, ppm]: 11.51 (1H, s, phenol OꢀH), 10.97 (1H, s, hydrazine Nꢀ
H), 8.31 (1H, s, imine CꢀH), 7.2ꢀ7.6 (10H, m, PhꢀH), 6.96 (2H, d, J= 9.6 Hz, PhꢀH), 6.85
(1H, d, J= 8.6 Hz, PhꢀH), 4.52 (4H, s, CH2).
(5-NO2)C6H3(OH)CH=NNHCON(CH2Ph)2. Yield: (1.25 g, 3.09 mmol, 62%). Anal. Calcd.
for C22H20N4O4: C, 65.3; H, 5.0; N, 13.9. Found: C, 65.0; H, 5.1; N, 13.8. IR (cm−1): ν(C=O)
1
1651 vs. H NMR [(CD3)2SO, ppm]: 11.10 (1H, s, phenol OꢀH), 8.45 (2H, s, hydrazine NꢀH
and imine CꢀH), 8.11 (1H, s, PhꢀH), 7.2ꢀ7.4 (11H, m, PhꢀH), 7.06 (1H, d, J= 10 Hz, PhꢀH),
4.52 (4H, s, CH2).
(5-OMe)C6H3(OH)CH=NNHCON(CH2Ph)2. Yield: (1.03 g, 2.64 mmol, 53%). Anal. Calcd.
for C23H23N3O3: C, 71.0; H, 6.0; N, 10.8. Found: C, 71.0; H, 6.0; N, 11.0. IR (cm−1): ν(C=O)
6

Page 7 of 31
Dalton Transactions
View Article Online
DOI: 10.1039/C3DT52297K
1
1645 vs. H NMR [(CD3)2SO, ppm]: 10.90 (1H, s, phenol OꢀH), 10.81 (1H, s, hydrazine Nꢀ
H) 8.31 (1H, s, imine CꢀH), 7.2ꢀ7.4 (10H, m, PhꢀH), 6.90 (1H, d, J= 2.5 Hz, PhꢀH), 6.70ꢀ6.80
(2H, m, PhꢀH), 4.52 (4H, s, CH2), 3.70 (3H, s, OCH3).
C10H6(OH)CH=NNHCON(CH2Ph)2. Yield: (1.15 g, 2.81 mmol, 56%). Anal. Calcd. for
C26H23N3O2: C, 76.3; H, 5.6; N, 10.1. Found: C, 76.3; H, 5.7; N, 10.3. IR (cm−1): ν(C=O)
1
1633 vs. H NMR [(CD3)2SO, ppm]:12.90 (1H, s, phenol OꢀH), 10.90 (1H, s, hydrazine Nꢀ
H), 9.20 (1H, s, imine CꢀH), 8.10 (1H, s, NpꢀH), 7.86 (2H, m, NpꢀH), 7.54 (1H, m, NpꢀH),
7.35 (4H, m, PhꢀH), 7.29 (6H, m, PhꢀH), 7.20 (2H, m, NpꢀH), 4.52 (4H, s, CH2).
(5-OH)C6H3(OH)CH=NNHCON(CH2Ph)2. Yield: (1.12 g, 2.98 mmol, 60%). Anal. Calcd.
for C22H21N3O3: C, 70.4; H, 5.6; N, 11.2. Found: C, 70.5; H, 5.5; N, 11.2. IR (cm−1): ν(C=O)
1630 vs. 1H NMR [(CD3)2SO, ppm]: 10.80 (1H, br s, phenol OꢀH), 8.95 (1H, br s, phenol Oꢀ
H), 8.88 (1H, s, hydrazine NꢀH), 8.27 (1H, s, imine CꢀH), 7.2ꢀ7.4 (10H, m, PhꢀH), 6.82 (2H,
m, PhꢀH), 6.68 (1H, m, PhꢀH). 4.52 (4H, s, CH2).
(3-OH)C6H3(OH)CH=NNHCON(CH2Ph)2. Yield: (0.96 g, 2.56 mmol, 51%). Anal. Calcd.
for C22H21N3O3: C, 70.4; H, 5.6; N, 11.2. Found: C, 70.5; H, 5.5; N, 11.2. IR (cm−1): ν(C=O)
1
1629 vs. H NMR [(CD3)2SO, ppm]: 8.30 (1H, s, imine CꢀH), 7.2ꢀ7.4 (10H, m, PhꢀH), 6.80
(2H, m, PhꢀH), 6.69(1H, m, PhꢀH), 4.52 (4H, s, CH2).
(4-OH)C6H3(OH)CH=NNHCON(CH2Ph)2. Yield: (1.05 g, 2.80 mmol, 56%). Anal. Calcd.
for C22H21N3O3: C, 70.4; H, 5.6; N, 11.2. Found: C, 70.1; H, 5.7; N, 11.0. IR (cm−1): ν(C=O)
1639 vs. 1H NMR [(CD3)2SO, ppm]:11.66 (1H, br s, phenol OꢀH), 10.60 (1H, s, hydrazine Nꢀ
7

Dalton Transactions
Page 8 of 31
View Article Online
DOI: 10.1039/C3DT52297K
H), 9.82 (1H, s, phenol OꢀH), 8.22 (1H, s, imine CꢀH), 7.2ꢀ7.4 (10H, m, PhꢀH), 7.15 (1H, d,
J= 9.1 Hz, PhꢀH), 6.26ꢀ6.32 (2H, m, PhꢀH), 4.52 (4H, s, CH2).
(6-OH)C6H3(OH)CH=NNHCON(CH2Ph)2. Yield: (1.08 g, 2.81 mmol, 56%). Anal. Calcd.
for C22H21N3O3. ½ H2O: C, 68.7; H, 5.7; N, 10.9. Found: C, 68.9; H, 5.7; N, 10.5. IR (cm−1):
1
ν(C=O) 1626 vs. H NMR [(CD3)2SO, ppm]: 11.10 (2H, s, phenol OꢀH), 10.80 (1H, s,
hydrazine NꢀH), 8.70 (1H, s, imine CꢀH), 7.2ꢀ7.4 (10H, m, PhꢀH), 7.0 (1H, m, PhꢀH), 6.35
(2H, m, PhꢀH), 4.52 (4H, s, CH2).
2.2.3 Preparation of copper(II) complexes of dibenzyl semicarbazone. Substituted
salicylaldehyde dibenzyl semicarbazone (0.50 mmol) was dissolved in 35 mL of
dichloromethane. A solution of CuCl2.2H2O (0.50 mmol) in methanol (10 mL) was added
dropwise to the stirred semicarbazone solution, after which the mixture was stirred for
another 15 min. About threeꢀquarters of the solvent was removed under vacuum. Diethyl
ether was added dropwise with stirring until a trace amount of precipitate could be seen. After
standing at −4 ºC overnight, the precipitate formed was collected by filtration and washed
with diethyl ether prior to drying under vacuum.
Complex 1. Green colour. Yield: (0.14 g, 0.29 mmol, 58%). Anal. Calcd. for
C22H20ClCuN3O2.H2O: C, 55.5; H, 4.6; N, 8.8. Found: C, 55.5; H, 4.5; N, 9.0. IR (cm−1):
ν(C=O) 1603 vs.
Complex 2. Green colour. Yield: (0.16 g, 0.31 mmol, 62%). Anal. Calcd. for
C22H19Cl2CuN3O2.H2O: C, 51.8; H, 4.1; N, 8.2. Found: C, 51.8; H, 3.8; N, 8.2. IR (cm−1):
ν(C=O) 1606 vs.
Complex 3. Green colour. Yield: (0.15 g, 0.28 mmol, 56%). Anal. Calcd. for
C22H19BrClCuN3O2: C, 49.2; H, 3.5; N, 7.8. Found: C, 48.9; H, 3.6; N, 7.6. IR (cm−1):
ν(C=O) 1610 vs.
8

Page 9 of 31
Dalton Transactions
View Article Online
DOI: 10.1039/C3DT52297K
Complex 4. Green colour. Yield: (0.14 g, 0.28 mmol, 56%). Anal. Calcd. for
C22H19ClCuN4O4: C, 52.5; H, 3.8; N, 11.1. Found: C, 52.6; H, 3.8; N, 11.2. IR (cm−1):
ν(C=O) 1609 vs.
Complex 5. Brown colour. Yield: (0.13 g, 0.25 mmol, 50%). Anal. Calcd. for
C23H20ClCuN3O3.2H2O: C, 52.7; H, 4.4; N, 8.0. Found: C, 52.4; H, 4.3; N, 7.9. IR (cm−1):
ν(C=O) 1612 vs.
Complex 6. Light green colour. Yield: (0.14 g, 0.28 mmol, 56%). Anal. Calcd. for
C26H22ClCuN3O2: C, 61.5; H, 4.3; N, 8.3. Found: C, 61.5; H, 4.4; N, 8.3. IR (cm−1): ν(C=O)
1603 vs.
Complex 7.
Brown colour. Yield: (0.18 g, 0.36 mmol, 72%). Anal. Calcd. for
C22H20ClCuN3O3.1.75H2O: C, 52.3; H, 4.7; N, 8.3. Found: C, 52.0; H, 4.6; N, 8.7. IR (cm−1):
ν(C=O) 1603 vs
Complex 8. Green colour. Yield: (0.19 g, 0.39 mmol, 78%). Anal. Calcd. for
C22H20ClCuN3O3.H2O: C, 53.7; H, 4.5; N, 8.6. Found: C, 53.5; H, 4.7; N, 8.6. IR (cm−1):
ν(C=O) 1607 vs.
Complex 9. Green colour. Yield: (0.15 g, 0.32 mmol, 64%). Anal. Calcd. for
C22H20ClCuN3O3: C, 55.8; H, 4.2; N, 8.9. Found: C, 55.6; H, 4.2; N, 8.5. IR (cm−1): ν(C=O)
1606 vs.
Complex 10. Green colour. Yield: (0.12 g, 0.24 mmol, 48%). Anal. Calcd. for
C22H20ClCuN3O3.H2O: C, 53.4; H, 4.5; N, 8.5. Found: C, 53.4; H, 4.3; N, 8.3. IR (cm−1):
ν(C=O) 1611 vs.
2.2.4 Preparation of pyridine derivatives of (dihydroxybenzaldehyde dibenzyl
semicarbazone)copper(II) complexes. The corresponding chlorido complex (0.0384 g,
0.081 mmol) was dissolved in 20 mL of methanol in a Schlenk tube. One molar equivalent of
silver nitrate (0.0138 g, 0.081 mmol) was added and the solution (shielded from light) was
stirred vigorously for 1 h. The resultant mixture was filtered through Celite and evaporated
9

Dalton Transactions
Page 10 of 31
View Article Online
DOI: 10.1039/C3DT52297K
completely under reduced pressure. Pyridine (10 mL) was added to dissolve the residue, and
the solution was stirred at r.t. for 1 h. The dark green solution was evaporated under reduced
pressure, and the residue was redissolved in 10 mL of methanol. About 10 mL of diethyl
ether was added, and the mixture was stored overnight at 4°C. The precipitate formed was
isolated by filtration and washed with diethyl ether.
7-py. Brown colour. Yield: (0.025 g, 0.040 mmol, 50%). Anal. Calcd. for
CuC27H25N5O6.2.25 H2O: C, 52.3; H, 4.8; N, 11.3. Found: C, 52.1; H, 4.4; N, 11.1. IR (cm−1):
ν(C=O) 1618 vs, ν(NO3) 1384 vs.
9-py. Green colour. Yield: (0.037 g, 0.064 mmol, 79%). Anal. Calcd. for CuC27H25N5O6: C,
56.0; H, 4.4; N, 12.1. Found: C, 55.6; H, 4.5; N, 11.7. IR (cm−1): ν(C=O) 1610 vs, ν(NO3)
1384 vs.
2.3
X-Ray crystallography
Crystals of 9 and 9-py were grown by slow evaporation of solutions of the complexes
in methanol at room temperature. The crystals were mounted on glass fibres for data
collection at 298(2)K. The data were collected in the θ/2θ mode using a Siemens P4
diffractometer with Mo Kα radiation (λ = 0.71073 Å), and were corrected for absorption
effects using ψꢀscan data. The structures were solved by direct methods and refined by fullꢀ
matrix leastꢀsquares on F2. All nonꢀhydrogen atoms were refined anisotropically. Hydrogen
atoms of the molecule were introduced in calculated positions. All hydrogen atoms were
allowed to ride on their carrier atoms with idealized bond lengths and angles. Crystal and
refinement data and geometric parameters of the crystal structures are given in Table 1.
10

Page 11 of 31
Dalton Transactions
View Article Online
DOI: 10.1039/C3DT52297K
Table 1
Crystallographic data and structure refinement for compounds 9 and 9-py.
Compound
9
C22H20ClCuN3O3
473.40
9-py
C27H25CuN5O6
579.06
Empirical formula
Formula weight
Crystal system
Space group
a /Å
Triclinic
Monoclinic
P2(1)/n
Pꢀ1
9.516(2)
9.587(1)
b / Å
13.668(3)
17.620(2)
75.658(4)
84.462(7)
70.26(2)
16.270(2)
17.130(3)
90
c/ Å
α(°)
β (°)
101.44(1)
90
γ (°)
3
2089.7(7)
4
2618.8(6)
4
V/Å
Z
Crystal size/mm
Theta range (°)
Max. and min. transmission
0.30 x 0.20 x 0.10
1.78 to 25.00
0.9863 and 0.8699
1.202
0.30 x 0.30 x 0.20
2.26 to 25.00
0.9847 and 0.9042
0.886
1
ꢁ/mm−
Reflns. collected
8742
5931
Independent reflns. (Rint)
R1, wR2 [I > 2σ(I)]
7303 ( 0.0290)
0.0449,0.0919
4618 (0.0208)
0.0397, 0.0832
2.4
Fluorescent Intercalator Displacement (FID) Assay25
Sample Preparation: The oligonucleotides were dissolved in MilliꢀQ water to give 100 ꢁM
stock solutions. These were diluted in MilliꢀQ water to give 20 ꢁM solutions, which were
further diluted using 10 mM potassium cacodylate (pH 7.4) / 50 mM potassium chloride (60
mM K+) buffer to give working solutions of 0.50 ꢁM concentration. Prior to use in the FID
assay, the DNA strands were annealed by heating to 90 °C for 5 min and cooling to room
temperature overnight. The test compounds and thiazole orange (TO) were first dissolved in
DMSO to give 1 mM stock solutions, which were then diluted using 10 mM potassium
cacodylate (pH 7.4) / 50 mM potassium chloride (60 mM K+) buffer to give working
solutions of various concentrations: 100 ꢁM (test compounds), 3.0 ꢁM (TO for ds26) and 2.5
11

Dalton Transactions
Page 12 of 31
View Article Online
DOI: 10.1039/C3DT52297K
ꢁM (TO for HTelo, cꢀmyc and ds17). Procedure: In a 3ꢀmL quartz cuvette, 750 ꢁL of
oligonucleotide solution was mixed with 375 ꢁL of thiazole orange solution. Various
volumes of solutions of the test compounds (corresponding to final concentrations of 0 – 2.5
ꢁM) were then added, followed by the appropriate amount of K+ buffer to bring the total
volume to 1.5 mL. After an equilibration time of 3 min, the emission spectrum was recorded
between 510 and 750 nm with an excitation wavelength of 501 nm. The fluorescence
intensity (peak area) was calculated using the “trapezium rule” method and corrected for the
buffer contribution by subtraction. The percentage TO displacement was calculated using the
following formula:
fluorescence intensity with compound
% TO displacement ꢀ 100 ‒ 100 ꢁ ꢂ
ꢃ
fluorescence intensity without compound
To accurately determine the DC50 value (concentration of test compound causing 50%
TO displacement) for compounds that interact strongly with the DNA (DC50 < 2.5 ꢀM), the
intensity data were plotted using the SternꢀVolmer equation,26 I0/I = 1 + kc, where
I0 = intensity without test compound, I = intensity with test compound, c = concentration of
test compound, and k is a constant. Plots of I0/I against c appear to be linear (Figure S1).
When c = DC50, I0/I = 2, hence DC50 can be obtained from the reciprocal of the gradient of
the fitted line (constrained to intercept the vertical axis at 1). The average DC50 value was
calculated after obtaining three sets of reproducible data within 10 % standard error. For
compounds that cause less than 50% TO displacement at 2.5 ꢀM, triplicate measurements of
I were made at c = 2.5 ꢁM, and the DC50 value was estimated from the twoꢀpoint (c = 0 and
2.5 ꢀM) SternꢀVolmer plot.
2.5
UV-vis spectrophotometric titration
Sample preparation: Solutions of HTelo (2 ꢀ 3 mM) and ctꢀDNA (6 ꢀ 10 mM) were prepared
by dissolving HTelo and ctꢀDNA respectively in 50 mM TrisꢀHCl (pH 7.4) / 100 mM KCl
12

Page 13 of 31
Dalton Transactions
View Article Online
DOI: 10.1039/C3DT52297K
buffer. The solutions gave ratios of absorbance values at 260 and 280 nm above 1.8:1,
indicating that the DNA preparations were free from protein contamination.27 The copper
complexes were first dissolved in DMSO to give 1 mM stock solutions, which were then
diluted using 50 mM TrisꢀHCl (pH 7.4) / 100 mM KCl buffer to obtain 1 mL solutions of 20
ꢁM concentration. Procedure: Solutions of the copper complexes were titrated with aliquots
(2 ꢁL for HTelo, 10 ꢁL for ctꢀDNA) of the corresponding DNA solutions until the apparent
molar extinction coefficient at the λmax of the complex remained constant for three
consecutive additions of the DNA solution. The binding constants (Kb) of the complexes
were determined by fitting the data to a plot of [DNA]/(εf εa) versus [DNA] using the
following equation: 28
[DNA]/(εf εa) = [DNA]/(εf εb) + 1/[Kb(εf εb)]
where εf and εb correspond to the molar extinction coefficients of free and DNAꢀbound
complex respectively, and εa is the apparent molar extinction coefficient (absorbance at λmax
of complex / concentration of complex).
2.6
Circular Dichroism (CD) Studies
Sample Preparation: HTelo DNA was dissolved in MilliꢀQ water to give a 100 ꢁM stock
solution. The solution was then diluted using 50 mM TrisꢀHCl /100 mM KCl buffer (pH 7.4)
or 50 mM TrisꢀHCl (pH 7.4) to 5 ꢁM. Prior to use in the CD assay, the DNA solution was
annealed by heating the solution to 90 ºC for 5 min, followed by cooling to room temperature
overnight. The test compounds were dissolved in DMSO to give 10 mM stock solutions. All
solutions were stored at −20 ºC, and were thawed and diluted to 5 mM using DMSO
immediately before use. Procedure: Measurements were performed with 1:1, 1:2 and 1:4
oligo/compound ratios by adding the appropriate amount of solution of the test compound (5
mM) to 600 ꢀL of the HTelo solution (5 ꢀM) in a strainꢀfree 10 mm × 2 mm rectangular
quartz cuvette. After an equilibration time of 5 min, the CD spectra (200ꢀ450 nm) were
13

Dalton Transactions
Page 14 of 31
View Article Online
DOI: 10.1039/C3DT52297K
obtained at 20°C on a JASCOꢀ715 spectrometer using the following parameters: bandwidth,
2 nm; stepꢀsize, 0.1 nm; timeꢀpepꢀpoint, 4.0 s. The buffer contribution was subtracted
manually.
2.7
Cytotoxicity Assay
The cancer cell line, MOLTꢀ4 (human Tꢀlymphoblastic leukaemia), and nonꢀcancerous cell
line, IMRꢀ90 (human foetal lung fibroblast), were obtained from the American Type Culture
Collection. An RPMIꢀ1640 based medium containing 10 % foetal bovine serum was used to
culture MOLTꢀ4 cells, while Eagle’s Minimum Essential Medium supplemented with 10 %
foetal bovine serum was used for IMRꢀ90 cells. All media also contained 100 units/mL of
penicillin and 100
incubator.
ꢁg/mL of streptomycin. The cells were maintained at 37 °C in a 5 % CO2
Cytotoxicity of the compounds was determined using the MTT [3(3,4ꢀdimethylthiazolꢀ2ꢀ
yl)2,5ꢀdiphenyltetrazolium bromide] assay29 on microtitre plates. Both MOLTꢀ4 and IMRꢀ90
cells were seeded at 40,000 cells per well in 70 ꢀL of culture medium. While the suspended
MOLTꢀ4 cells were exposed to test compounds immediately after seeding, the adherent IMRꢀ
90 cells were cultured for 24 h prior to treatment with compounds.
All compounds were dissolved in absolute DMSO to give 100 mM stock solutions, which
were then serially diluted to give working solutions in 10% v/v DMSO in culture medium.
Six serial dilutions were made for each compound in order to generate a doseꢀresponse curve,
and six replicate test wells were set up for each concentration of the compound. Ten
microliters of the appropriate working solution was added to each test well, hence the final
DMSO concentration was 1.25 % v/v in each well. Each plate also contained a blank well
(cellꢀfree mediumꢀonly well), solvent control wells which contained cells and 1.25 % v/v
14

Page 15 of 31
Dalton Transactions
View Article Online
DOI: 10.1039/C3DT52297K
DMSO, drug colour control wells which contained test compound and medium only, and
growth control wells which contained only cells in medium.
Following incubation of cells with test compounds for 24 h at 37 °C and 5 % CO2, 20 ꢂL of a
5 mg/mL solution of MTT was added to each well (final MTT concentration 1 mg/mL).
Three hours later, 100 ꢂL of lysing solution (20 % sodium dodecyl sulfate dissolved in 50 %
N,Nꢀdimethylformamide, pH adjusted to 4.7 with acetic acid) was added to each well. After
the microtitre plate was left to stand overnight, the absorbance of the solution in each well
was read at 570 nm. The percent inhibition of growth for each concentration of compound
was calculated from the absorbance29 and plotted against the concentration to give a graph
from which the IC50 value (concentration of compound required to inhibit the growth of the
cells by 50 %) was determined.
2.8
Molecular docking studies
The AutoDock Vina30 docking software was used to model the binding of complexes
9 and 9-py with the parallel quadruplex31 formed from the 22AG HTelo sequence. The
crystal structure of the quadruplex was obtained from the Protein Data Bank (PDB code
1KF1). All input structures were prepared from the respective crystal structures using
AutoDock Tool (version 1.5.4): the nitrate anion was removed from 9-py, polar hydrogen
atoms were added to the quadruplex structure, and the crystallographic water molecules and
nonꢀintercalated potassium ion were removed from the quadruplex. The rotatable bonds of 9
and 9-py are indicated in yellow in Figure 1.
15

Dalton Transactions
Page 16 of 31
View Article Online
DOI: 10.1039/C3DT52297K
Figure 1. Structures of 9 and 9-py with rotatable bonds (yellow color).
3
Results and discussion
Crystal structure of 9
3.1
There are two independent molecules of the complex (Figure 2) in the unit cell. These
molecules are similar in their gross features, except that the benzyl groups are in syn
conformation in one molecule and antiꢀconformation in the other. The semicarbazone ligand
acts as a tridentate ligand that binds the copper atom via the imine nitrogen [N(1)/N(4)],
carbonyl oxygen [O(3)/O(6)] and phenoxo oxygen [O(1)/O(4)] atoms. A chloride ligand
occupies the fourth coordination site on each Cu(II) ion. The coordination geometry can be
described as distorted square planar, with the angles between cisꢀligands on each copper ion
(Table 2) summing up to 361º. In addition, the three angles about the terminal nitrogen atoms
of both the molecules, C(8)ꢀN(3)ꢀC(9), C(8)ꢀN(3)ꢀC(16), C(9)ꢀN(3)ꢀC(16) and C(30)ꢀN(6)ꢀ
C(38), C(30)ꢀN(6)ꢀC(31), C(31)ꢀN(6)ꢀC(38), sum up to 360º each, indicating a trigonal
planar geometry at N(3) and N(6). This implies that the terminal nitrogen atoms are sp2
hybridised and conjugated with the carbonyl group. Overall, the bond lengths and angles of 9
(Table 2) are in close agreement with those reported for other copper(II) salicylaldehyde
semicarbazone complexes.21, 24
16

Page 17 of 31
Dalton Transactions
View Article Online
DOI: 10.1039/C3DT52297K
Figure 2.
Table 2
The independent molecules in the crystal structure of 9 (40% probability
ellipsoids).
Selected bond lengths (Å) and angles (º) for Complexes 9 and 9-py
Complex 9 (Molecule 1)
Complex 9 (Molecule 2)
Complex 9-py
Cu(1)ꢀO(1)
Cu(1)ꢀO(1)
1.897(2)
1.935(3)
1.957(2)
2.228(1)
92.7(1)
Cu(2)ꢀO(4)
1.894(2)
1.920(3)
1.988(3)
2.227(1)
92.6(1)
1.868(2)
1.930(2)
1.954(2)
1.979(2)
92.75(8)
Cu(1)ꢀN(1)
Cu(2)ꢀN(4)
Cu(1)ꢀN(1)
Cu(1)ꢀO(3)
Cu(2)ꢀO(6)
Cu(1)ꢀO(3)
Cu(1)ꢀCl(1)
Cu(2)ꢀCl(2)
Cu(1)ꢀN(4)
O(1)ꢀCu(1)ꢀN(1)
O(1)ꢀCu(1)ꢀO(3)
N(1)ꢀCu(1)ꢀO(3)
O(1)ꢀCu(1)ꢀCl(1)
O(4)ꢀCu(2)ꢀN(4)
O(4)ꢀCu(2)ꢀO(6)
N(4)ꢀCu(2)ꢀO(6)
O(4)ꢀCu(2)ꢀCl(2)
O(1)ꢀCu(1)ꢀN(1)
172.4(1)
81.0(1)
170.7(1)
80.8(1)
O(1)ꢀCu(1)ꢀO(3) 173.32(8)
N(1)ꢀCu(1)ꢀO(3)
O(1)ꢀCu(1)ꢀN(4)
81.38(8)
91.82(9)
95.34(8)
92.46(7)
N(1)ꢀCu(1)ꢀCl(1) 168.93(9)
N(4)ꢀCu(2)ꢀCl(2) 173.34(9)
N(1)ꢀCu(1)ꢀN(4) 173.46(9)
O(3)ꢀCu(1)ꢀCl(1)
C(8)ꢀN(3)ꢀC(9)
C(8)ꢀN(3)ꢀC(16)
C(9)ꢀN(3)ꢀC(16)
91.52(8)
123.0(3)
118.4(3)
117.4(3)
O(6)ꢀCu(2)ꢀCl(2)
C(30)ꢀN(6)ꢀC(31)
C(30)ꢀN(6)ꢀC(38)
C(31)ꢀN(6)ꢀC(38)
94.68(8)
123.4(3)
120.0(3)
116.3(3)
O(3)ꢀCu(1)ꢀN(4)
C(8)ꢀN(3)ꢀC(9)
C(8)ꢀN(3)ꢀC(16)
C(9)ꢀN(3)ꢀC(16)
93.78(8)
123.0(2)
118.8(2)
118.0(2)
17

Dalton Transactions
Page 18 of 31
View Article Online
DOI: 10.1039/C3DT52297K
3.2
Crystal structure of 9-py
The semicarbazone ligand acts as a tridentate ligand that binds the copper atom via
the imine nitrogen [N(1)], carbonyl oxygen [O(3)] and phenoxo oxygen [O(1)] atoms. A
pyridine ligand occupies the fourth coordination site of the copper atom (Figure 3). There
appears to be a very weak interaction between the copper atom and O(6) of the nitrate
counterion [Cu(1)⋅⋅⋅O(6) 2.78 Å]. This interaction has negligible effect on the coordination
geometry of the cationic complex, which remains essentially square planar, with the angles
between the cis ligands on the copper atom summing up to 360º. As in complex 9, the three
angles about the terminal nitrogen atom, C(8)ꢀN(3)ꢀC(16), C(8)ꢀN(3)ꢀC(9) and C(9)ꢀN(3)ꢀ
C(16), also sum up to 360º, indicating a trigonal planar geometry and hence conjugation of
N(3) with the carbonyl group. The mean plane of the pyridine ring makes an angle of 25.6°
with the coordination plane of the Cu(II) ion.
Figure 3.
Crystal structure of 9-py (40% probability ellipsoids).
18

Page 19 of 31
Dalton Transactions
View Article Online
DOI: 10.1039/C3DT52297K
3.3
Fluorescence Intercalator Displacement assay (FID). The ability of the copper(II)
salicylaldehyde semicarbazone complexes to bind to different oligonucleotides via
intercalation or endꢀstacking was evaluated using the FID assay. Two quadruplex forming
sequences, HTelo and c-myc, and two duplex forming sequences, ds26 and ds17, were used
to explore the quadruplex versus duplex selectivity. The DNA affinity is measured by the
concentration of the test compound required to displace thiazole orange from the DNA matrix
by 50% (DC50), the values of which are given in Table 3.
Table 3
Affinity and selectivity of copper(II) salicylaldehyde semicarbazone
complexes for various oligonucleotides, determined by the FID assay.
DC50 (µM)a
Selectivity for HTelo b
Complex
HTelo
c-myc
ds17
ds26
vs. ds17 vs. ds26 vs. c-myc
20.00
(0.05)
11.1
(0.1)
5.6
(0.1)
12.50
(0.07)
0.28
1.1
0.63
1.8
0.56
3.6
1
2
9.2
(0.1)
33.2
(0.1)
10.2
(0.2)
16.7
(0.1)
15.06
(0.09)
22.7
(0.2)
20.2
(0.1)
28.7
(0.2)
1.34
0.17
0.18
1.45
2.21
1.12
3.33
0.90
3.28
10
1.91
0.41
0.34
0.62
5.62
0.57
2.30
1.38
6.21
26
1.51
1.53
2.42
1.12
5.55
0.58
5.34
1.37
5.30
22
3
39.21
(0.08)
60.0
(0.1)
6.52
(0.08)
16.13
(0.07)
4
33.33
(0.07)
80.6
(0.2)
5.56
(0.06)
11.6
(0.1)
5
22.94
(0.08)
25.6
(0.2)
33.33
(0.08)
14.29
(0.09)
6
2.40
(0.08)
13.32
(0.03)
5.30
(0.05)
13.50
(0.05)
7
26.33
(0.07)
15.3
(0.1)
29.4
(0.1)
15.10
(0.09)
8
2.10
(0.02)
11.22
(0.03)
7.00
(0.06)
4.84
(0.04)
9
15.1
(0.1)
20.7
(0.2)
13.5
(0.1)
20.9
(0.1)
10
7-py
9-py
1.98
(0.03)
10.50
(0.06)
6.50
(0.07)
12.30
(0.06)
0.55
(0.03)
12.20
(0.05)
5.50
(0.02)
14.3
(0.1)
b
a Standard errors (N = 3) are shown in parentheses. Ratio of DC50 for oligonucleotide to
DC50 for HTelo.
19

Dalton Transactions
Page 20 of 31
View Article Online
DOI: 10.1039/C3DT52297K
The FID screening showed most of the complexes investigated to have low affinity
(DC50 > 2.5 ꢀM) for DNA. Amongst the nonꢀ and 5ꢀsubstituted complexes (1 – 7), only the
hydroxylꢀsubstituted complex 7 shows some affinity for HTelo DNA. Its HTeloDC50 value
(2.4 ꢁM) is comparable to those of monoꢀmetallic complexes with reasonable binding affinity
and selectivity for HTelo DNA.16ꢀ19 This prompted us to study analogous complexes (8 – 10)
with the hydroxyl group at other positions on the salicylaldehyde ring. Of these, only the 4ꢀ
hydroxyl analogue, 9, exhibits some affinity for HTelo. Interestingly, replacement of the
chloride ligands of 7 and 9 with pyridine results in an increase in affinity and selectivity for
HTelo, especially for the 4ꢀhydroxyl analogue. The increase in affinity is attributed to the
presence of a positive charge on 7-py and 9-py, which enhances electrostatic binding with the
electronꢀrich guanine groups, and the increase in π surface area for stacking with the Gꢀ
tetrads. Another interesting feature of complex 9-py is that it not only displays selectivity for
HTelo over duplex DNA, but also over another quadruplex structure, namely cꢀmyc.
3.4
UV-vis Spectrophotometric DNA titration. The binding constants of complexes 7,
9, 7-py and 9-py towards HTelo and ctꢀDNA were determined by UVꢀvis spectrophotometric
DNA titration. Addition of HTelo to the complexes resulted in considerable hypochromicity
and bathochromic shift of the absorption bands of the complexes. These spectral changes
indicate an endꢀstacking binding mode for the complexes.32 A typical set of spectra in the
presence of increasing amounts of HTelo are shown in Figure 4. The DNA binding constants
derived from plots of [DNA]/(εf εa) vs. [DNA] are given in Table 4. These data are
consistent with the trends in affinity and selectivity observed in the FID assay. It is
noteworthy that the HTeloKb / ctꢀDNAKb ratios of 7-py and 9-py exceed that of telomestatin12
(70) and those of HTeloꢀselective copper(II)ꢀterpyridine16 and salphen19 complexes (2 – 135).
20

Page 21 of 31
Dalton Transactions
View Article Online
DOI: 10.1039/C3DT52297K
8E-09
7.5E-09
7E-09
6.5E-09
6E-09
5.5E-09
5E-09
y = 1.06E-04x + 5.85E-10
R² = 9.99E-01
4.5E-09
4E-09
4E-05
5E-05
6E-05 7E-05
[DNA] / M
(a)
(b)
Figure 4. Titration of complex 9 (20 ꢁM) with HTelo: (a) UVꢀvis spectral changes with
increasing amounts of HTelo; (b) plot of [DNA]/(εf εa) vs. [DNA].
Table 4
Binding constants, Kb (M−1), of copper(II) salicylaldehyde semicarbazone
complexes with HTelo and ctꢀDNA, determined by UVꢀvis spectrophotometric
titration. Standard errors for Kb values (N = 3) are shown in parentheses.
HTeloKb
ct-DNAKb
Selectivity a
20.7
Complex (λmax/nm)
7 (377)
5.77 (0.03) × 104
1.76 (0.05) × 105
2.79 (0.07) × 103
3.18 (0.05) × 103
9 (353)
7-py (399)
9-py (352)
55.3
9.46 (0.03) × 104
3.42 (0.04) ×105
5.55 (0.06) ×102
1.62 (0.04) × 103
170
211
a Ratio of Kb for HTelo to Kb for ctꢀDNA.
3.5
Circular Dichroism (CD) Studies. CD spectroscopic studies were carried out to
investigate the interaction of complexes 7, 9, 7-py and 9-py with HTelo. In the presence of
potassium ions (100 mM), HTelo adopts a mixture of conformations,33 as demonstrated by
the positive band at 295 nm, the shoulder band between 270ꢀ280 nm, and the weaker positive
band at 260 nm (Figure 5). A slight decrease in magnitude of molar ellipticity between 230
and 320 nm was typically observed upon addition of increasing amounts of the test
compound, although the overall shape of the spectrum was retained (Figure 5). There is
21

Dalton Transactions
Page 22 of 31
View Article Online
DOI: 10.1039/C3DT52297K
therefore negligible change in the conformation of the HTelo quadruplex when it binds to the
copper complexes. The CD spectrum of HTelo in potassiumꢀfree buffer showed negligible
changes in the presence of the copper complexes (Figure 6). This indicates that the
complexes do not template quadruplex formation from the unfolded HTelo sequence. The
lack of templating effect of the complexes, despite their selective binding to the HTelo
quadruplex in the presence of potassium ions, is consistent with the moderate magnitude of
the binding constants. Given that the potassium concentration in cells is in the region of 100
mM, the ability of the complexes to recognize preꢀformed quadruplex structures is still
biologically relevant.
Figure 5. CD spectra of HTelo quadruplex DNA (5 ꢁM) in TrisꢀHCl 50 mM buffer
supplemented with KCl (100 mM) with increasing amounts of complex 7.
22

Page 23 of 31
Dalton Transactions
View Article Online
DOI: 10.1039/C3DT52297K
Figure 6. CD spectra of HTelo DNA (5 ꢁM) in 50 mM TrisꢀHCl buffer (no KCl added) with
increasing amounts of complex 7.
3.6
Molecular docking studies. Molecular docking studies were conducted to investigate
the interaction of complexes 9 and 9-py with the parallel quadruplex31 formed from the 22AG
HTelo sequence. Twenty binding configurations were generated for each complex (Figures
S2 and S3). While the majority of these configurations involve the stacking of the complexes
on the 3′ or 5′ face of the quadruplex, there are also several configurations wherein the
salicylimine moiety occupies one of the channels formed between the TTA loops and Gꢀ
tetrads. In agreement with the empirical data, the calculated binding affinities (Table 5)
indicate that 9-py binds more strongly than 9 with the quadruplex DNA.
In the most stable binding configuration for 9 (Figure 7), the complex is stacked on
the 5′ side of the quadruplex, with its planar core overlapping one of the guanine bases and
the benzyl groups being directed into the mouth of a channel formed between the TTA loops
and the G tetrads. The Cu atom is ca. 3.32 Å above the mean plane of the guanine base. A
weak hydrogen bond is indicated between the carbonyl oxygen of another guanine base in the
tetrad and the 4ꢀOH group of 9 (H⋅⋅⋅O distance ~2.89Å). Correspondingly, the contribution
23

Dalton Transactions
Page 24 of 31
View Article Online
DOI: 10.1039/C3DT52297K
of hydrogen bonding to the overall binding affinity is calculated to be −0.27 kcal/mol. The
calculated total hydrophobic contributions to the binding affinity for configurations 1
(described above) and 12 (also stacked, but with benzyl groups pointing away from the
quadruplex surface, see Figure S2) are −0.19 and −0.09 kcal/mol, respectively. This indicates
that there are hydrophobic interactions between the benzyl groups and the aforesaid channel
in configuration 1.
Table 5. Calculated binding affinity of 9 and 9-py with DNA Gꢀquadruplex.
Binding affinity (kcal/mol)
Configurations
9
9-py
−7.5
−7.3
−7.3
−7.2
−7.1
−7.1
−7.1
−6.8
−6.7
−6.7
−6.7
−6.7
−6.7
−6.6
−6.5
−6.5
−6.4
−6.4
−6.3
−6.1
1
2
−6.9
−6.6
−6.6
−6.4
−6.3
−6.2
−6.2
−6.2
−6.1
−6.1
−6.1
−6.1
−6.1
−5.9
−5.9
−5.9
−5.8
−5.8
−5.8
−5.7
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
24

Page 25 of 31
Dalton Transactions
View Article Online
DOI: 10.1039/C3DT52297K
Figure 7. (a) Molecular docking model illustrating the interaction between 9 and DNA Gꢀ
quadruplex, (b) top and (c) side view of top surface of DNA Gꢀquadruplex containing 9.
Complex 9-py is also 5′ꢀstacked with its benzyl groups inside a channel in its most favoured
configuration (Figure 8). The pyridine ring overlaps the imidazole ring of a guanine base,
with a centroid (pyridine) – mean plane (guanine) distance of ca. 3.63 Å. As in the case of 9,
a weak hydrogen bond (albeit stronger than that for 9) exists between a guanine carbonyl
oxygen and the 4ꢀOH group of 9-py (H⋅⋅⋅O distance ~2.39Å). The calculated hydrogen
bonding contribution to the binding affinity is −0.40 kcal/mol.
In summary, the docking results indicate that the higher HTelo affinity of 9-py is due
to the additional π−π interaction afforded by the pyridine ring, and stronger hydrogen
bonding with the guanine oxygen. The results also suggest that it may be possible to enhance
the interaction of the copper complex with the channel and/or groove of the HTelo
quadruplex by functionalizing the benzyl groups and/or pyridine ring of 9-py. This may
result in new complexes with higher affinity and selectivity for HTelo.
25

Dalton Transactions
Page 26 of 31
View Article Online
DOI: 10.1039/C3DT52297K
Figure 8. (a) Molecular docking model illustrating the interaction between 9-py and DNA Gꢀ
quadruplex, (b) top and (c) side view of top surface of DNA Gꢀquadruplex containing 9-py.
3.7
Cytotoxicity. Most of the copper(II) salicylaldehyde semicarbazone complexes
tested show high cytotoxicity against MOLTꢀ4 human leukaemia cells, with IC50 values in
the low micromolar region (Table 6). In addition, these complexes inhibit MOLTꢀ4 cells
selectively over nonꢀcancerous IMRꢀ90 human fibroblasts, especially for 9-py. It is also
noteworthy that, whilst 7-py is not exceptional in its toxicity towards MOLTꢀ4 cells, it has
negligible toxicity towards the fibroblasts. Interestingly, there is good correlation between
the selectivity for MOLTꢀ4 cells and that for HTelo amongst the complexes that bind more
strongly to HTelo (Table 6). This might indicate that the mode of action of these complexes
involve their binding to telomeric DNA. Generation of DNAꢀcleaving reactive oxygen
species may occur after DNA binding, leading to cell death. The ability of copper(II)
salicylaldehyde semicarbazone complexes to cleave DNA via the generation of reactive
oxygen species has been documented.22
26

Page 27 of 31
Dalton Transactions
View Article Online
DOI: 10.1039/C3DT52297K
On the whole, the complexes with low affinity for HTelo appear to be more active
than their strongerꢀbinding counterparts against both MOLTꢀ4 and IMRꢀ90 cells. The
weaklyꢀbinding complexes are unlikely to act via interaction with telomeric DNA, however.
Correspondingly, there is no discernible correlation between their selectivity for HTelo and
their selectivity for MOLTꢀ4 cells. Mediation of oxidative damage might be the dominant
mode of action instead for these complexes.
Table 6
Cytotoxicity and selectivity of copper(II) salicylaldehyde semicarbazone
complexes against MOLTꢀ4 and IMRꢀ90 cells. Standard errors for IC50 values
(N = 6) are given in parentheses. Selectivity values of the complexes between
HTelo and ds26 oligonucleotides are included for comparison.
IC50 (µM)
Selectivity
IMR-90IC50 /
ds26DC50 /
HTeloDC50
Complex
MOLT-4
IMR-90
MOLT-4IC50
Strong HTelo binders
3.1 (0.2)
18.1 (0.5)
5.0 (0.2)
30.2 (0.4)
> 125
9.7
> 6.9
6.3
26
6.2
5.6
2.3
9-py
7-py
31.5 (0.5)
30.5 (0.8)
7
8.03 (0.08)
3.8
9
Weak HTelo binders
1.0 (0.4)
1.1 (0.1)
1.93 (0.08)
2.1 (0.3)
1.9 (0.1)
15.3 (0.4)
5.5 (0.3)
5.9 (0.2)
11.7 (0.5)
7.3 (0.8)
10.3 (0.5)
7.3 (0.1)
5.03 (0.08)
30.1 (0.2)
6.8 (0.4)
6.8 (0.4)
12
0.62
0.41
0.57
0.63
0.34
1.4
6
4
6.6
5.3
3.5
2.6
2.0
1.2
1.2
8
1
5
10
2
1.8
1.9
3
27

Dalton Transactions
Page 28 of 31
View Article Online
DOI: 10.1039/C3DT52297K
4
Summary and Conclusion
A series of copper(II) complexes containing substituted salicylaldehyde dibenzyl
semicarbazone ligands have been prepared. Amongst the chloridocopper(II) complexes,
only those of the 4ꢀ and 5ꢀhydroxylꢀsubstituted ligands (complexes 9 and 7) show affinity for
HTelo quadruplex DNA. Complexes 7 and 9 also show good selectivity for HTelo over cꢀ
myc and duplex DNA. The chlorido ligands of 7 and 9 can be substituted with pyridine,
yielding derivatives (7-py and 9-py) with improved affinity and much higher selectivity for
HTelo DNA. Molecular docking studies show that 9 and 9-py stack on the Gꢀtetrad at the 5′ꢀ
end of the HTelo quadruplex, with the benzyl groups directed into a channel formed between
the TTA loops and the Gꢀtetrads, and the hydroxyl group forming a hydrogen bond with a
guanine residue. Most of the complexes prepared in this work show high cytotoxicity against
MOLTꢀ4 human leukaemia cells. Complexes 7, 9, 7-py and 9-py also exhibit high selectivity
for MOLTꢀ4 cells over nonꢀcancerous IMRꢀ90 human fibroblasts.
5
Acknowledgements
Financial support from the National Institute of Education, Singapore (grants RS 12/10 YYK
and RI 10/10 PL) and from the Engineering and Physical Sciences Research Council, UK are
gratefully acknowledged. The authors also thank Ms YingꢀChang Ting, HuiꢀXin Ting, and
WanꢀYen Lee for technical assistance.
6
Supplementary Information
Crystallographic data (excluding structure factors) for complexes 9 and 9-py may be obtained
from the Cambridge Crystallographic Data Centre (CCDC, 12 Union Road, Cambridge CB2
1EZ, UK; fax: _/44ꢀ1223ꢀ336ꢀ033; eꢀmail: deposit@ccdc.cam.ac.uk or www:
28

Page 29 of 31
Dalton Transactions
View Article Online
DOI: 10.1039/C3DT52297K
http://www.ccdc.cam.ac.uk) on quoting the depository numbers CCDC 955520 and 955521,
respectively.
REFERENCES
1.
Burge, S.; Parkinson, G. N.; Hazel, P.; Todd, A. K.; Neidle, S. Nucleic Acids Res. 2006,
34, 5402.
2.
3.
4.
5.
6.
Parkinson, G. N.; Cuenca, F.; Neidle, S. J Mol.Bio. 2008, 381,1145.
Neidle, S.; Parkinson, G. N. Biochimie. 2008, 90, 1184.
Jiang, Y. L.; Liu, Z. P. Mini Rev. Med. Chem. 2010, 10, 726.
Blasco, M. A. Eur. J. Cell Biol. 2003, 82, 441.
De Cian, A.; Lacroix, L.; Douarre, C.; TemimeꢀSmaali, N.; Trentesaux, C.; Riou, J. F.;
Mergny, J. L. Biochimie. 2008, 90, 131.
7.
8.
Rangan, A.; Fedoroff, O. Y.; Hurley, L. H. J. Biol. Chem. 2001, 276, 4640.
SiddiquiꢀJain, A.; Grand, C. L.; Bearss, D. J.; Hurley, L. H. Proc. Natl. Acad. Sci.
U. S. A. 2002, 99, 11593.
9.
Fernando, H.; Reszka, A. P.; Huppert, J.; Ladame, S.; Rankin, S.; Venkitaraman, A. R.;
Neidle, S.; Balasubramanian, S. Biochemistry. 2006, 45, 7854.
10. Rankin, S.; Reszka, A. P.; Huppert, J.; Zloh, M.; Parkinson, G. N.;Todd, A. K.;
Ladame, S.; Balasubramanian, S.; Neidle, S. J. Am. Chem. Soc. 2005, 127, 10584.
11. Qin, Y.; Hurley, L. H. Biochimie. 2008, 90, 1149.
12. Monchaud, D.; TeuladeꢀFichou, M.ꢀP. Org. Biomol. Chem. 2008, 6, 627.
13. Ou, T.ꢀM.; Lu, Y.ꢀJ.; Tan, J.ꢀH.; Huang, Z.ꢀS.; Wong, K.ꢀY.; Gu, L.ꢀQ.
ChemMedChem 2008, 3, 690.
14. Reed, J. E.; White, A. J. P.; Neidle, S.; Vilar, R. Dalton Trans. 2009, 2558.
15. Barry, N. P. E.; Abd Karim, N. H.; Vilar, R.; Therrien, B. Dalton Trans. 2009, 10717.
16. Suntharalingam, K.; White, A. J. P.; Vilar, R. Inorg. Chem. 2010, 49, 8371.
17. Suntharalingam, K.; Gupta, D.; Sanz Miguel, P.J.; Lippert, B,; Vilar, R. Chem.Eur. J.
2010, 16, 3613.
29