Coordination susceptibilities of cinnamaldehyde and cuminaldehyde derived Schiff bases towards the fac-[Re(CO)3]+ core: Formation, computational and DNA interaction studies

Inorganica Chimica Acta 477 (2018) 257–269

Contents lists available at ScienceDirect

Inorganica Chimica Acta

journal homepage: w ww.elsevier.com/locate/ica

Research paper

Coordination susceptibilities of cinnamaldehyde and cuminaldehyde

derived Schiff bases towards the fac-[Re(CO)₃]⁺core: Formation,

computational and DNA interaction studies

⇑

Muhammed Bilaal Ismail, Irvin Noel Booysen , Matthew Piers Akerman

School of Chemistry and Physics, University of KwaZulu-Natal, Private Bag X01, Scottsville 3209, Pietermaritzburg, South Africa

a r t i c l e i n f o

a b s t r a c t

Article history:

Herein, we explore the coordination behaviours of Schiff bases derived from constituents of essential oils

(viz. cinnamaldehyde and cuminaldehyde) and aromatic amines (viz. 2-aminophenol and 4-aminoan-

tipyrene) towards the facial tricarbonylrhenium(I) core. The resultant rhenium(I) compounds:

(m-X)₂[Re(CO)₃]₂{where X = 2-{[3-phenylprop-2-en-1-ylidene]amino}phenol (ciap) (for 1) or 2-{[4-(pro-

pan-2-yl)benzylidene]amino}phenol (cuap) (for 2), fac-[Re(CO)₃(apy)Cl] (apy = 4-aminoantipyrene) (3)

and fac-[Re(CO)₃(cinap)Cl] (cinap = 1, 5-dimethyl-2-phenyl-4-{[3-phenylprop-2-en-1-ylidene]amino}-

1,2-dihydro-3H-pyrazol-3-one) (4) were formed. The metal compounds were spectroscopically charac-

terized and structural elucidations were conﬁrmed by single crystal X-ray diffraction. DFT studies were

utilized to rationalize the contrasting coordination behaviours of the structural analogues, 1,5-

dimethyl-2-phenyl-4-{[4-(propan-2-yl)benzylidene]amino}-1,2-dihydro-3H-pyrazol-3-one (cumap) and

cinap in the formation of 3 and 4, respectively. These metal compounds showed afﬁnities towards

Calf-Thymus DNA based on UV-Vis DNA binding titrations and molecular docking studies revealed that

they are groove binders. In addition, gel electrophoresis experiments indicated that steric factors has

an inﬂuence on the DNA cleavage activities of the mono- and dinuclear rhenium(I) compounds.

Received 24 January 2018

Received in revised form 15 March 2018

Accepted 16 March 2018

Available online 27 March 2018

Keywords:

Rhenium

Schiff base

DNA interaction

Molecular docking

Gel electrophoresis

1. Introduction

Metal complexes of cinnamaldehyde and cuminaldehyde are

scarcely found in literature possibly because they lack electron-

Essential oils are natural, pungent and volatile compounds

which are formed by plants as secondary metabolites. They are

extracted from various ﬂorae originating from temperate and trop-

ical countries where they represent an important part of the tradi-

tional pharmacopoeia [1]. In this study, constituents of essential

oils, cinnamaldehyde and cuminaldehyde Fig. 1, have been utilized

partly due to their inherent biological activities [2,3]. Under in vivo

experimental conditions, cinnamaldehyde has shown to decrease

plasma glucose, glycosylated haemoglobin, cholesterol and

increase insulin levels while similar traits are observed for cumi-

naldehyde with hypoglycaemic effects being in the majority [4].

Moreover, both organic compounds have revealed anti-inﬂamma-

tory and anticancer properties [5–8]. Such complementary proper-

ties may allow these natural occurring compounds and their

derivatives to form chemotherapeutic compounds which minimize

the traditional side-effects of current anticancer agents.

donating atoms in order to form metal-stabilizing bidentate or tri-

dentate ligands. However, these bio-aldehydes can become

extended pi-aromatic systems by the formation of Schiff bases.

An example thereof is N, N⁰–bis(

a

-methyl-trans-cinnamadehyde)

ethylenediimine (L) which was isolated from the 1:2 M condensa-

tion reaction between -methyl-trans-cinnamaldehyde and

a

ethylenediamine. This diimine, L, afforded metal(II) complexes

with general formula of [ML₂Cl₂] (where M = Ni or Co) which dis-

played high anticandidal activity accompanied with limited toxic-

ity towards vertebrate cells [9]. Additionally, biologically relevant

Schiff bases derived from thiosemicarbazone with trans-cin-

namaldehyde (Htcin) and cuminaldehyde (Htcum) coordinate as

bidentate monoanionic chelates to nickel(II) and copper(II) cen-

tres; viz. [Ni(X)₂]ꢀH₂O and [Cu(X)(H₂O)Cl] where X = tcin or tcum.

These transition metal complexes exhibit properties which are

practical for in vivo antileukemic exploration [10].

In this research study, two dinuclear facial tricarbonylrhenium

(I) compounds were isolated from a ‘‘one-pot” synthetic method

of equimolar ratios (1:1:1 stoichiometry) featuring chloropen-

tacarbonylrhenium(I), 2-aminophenol and one of the relevant

⇑

Corresponding author.

E-mail address: Booyseni@ukzn.ac.za (I.N. Booysen).

https://doi.org/10.1016/j.ica.2018.03.032

258

M.B. Ismail et al. / Inorganica Chimica Acta 477 (2018) 257–269

25. Melting points were determined using a Stuart SMP3 melting

point apparatus. The conductivity measurements were determined

at 295 K on a Radiometer R21M127 CDM 230 conductivity and pH

meter. Toluene was dried with sodium wire and stored in the pres-

ence of dry 4 Å microsieves. The redox properties of the metal com-

plexes were investigated by cyclic and square wave voltammetry.

The concentration of the metal complexes was 2 mM and tetra-

butylammoniumtetraﬂuoroborate (0.1 M) was used as the sup-

porting electrolyte. A Metrohm Autolab Potentiometer was used

for measurements equipped with a three electrode system: pseudo

Ag|AgCl reference electrode, a glassy carbon working electrode

(GCE) and Pt auxiliary counter electrode.

Fig. 1. Structures of cinnamaldehyde (a) and cuminaldehyde (b).

natural oils, cinnamaldehyde or cuminaldehyde to afford (m-X)₂[Re

(CO)₃]₂{where X = 2-{[3-phenylprop-2-en-1-ylidene]amino}phe-

nol (ciap) (for 1) or 2-{[4-(propan-2-yl)benzylidene]amino}phenol

(cuap) (for 2)} Scheme 1. Mononuclear rhenium(I) complexes,

fac-[Re(CO)₃(apy)Cl] (apy = 4-aminoantipyrene) (3) and fac-[Re

(CO)₃(cinap)Cl] (cinap = 1, 5-dimethyl-2-phenyl-4-{[3-phenyl-

prop-2-en-1-ylidene]amino}-1,2-dihydro-3H-pyrazol-3-one) (4)

were formed from exploring the coordination behaviours of

the Schiff bases derived from the respective natural oils and

4-aminoantipyrene Scheme 2. The structural elucidations were

conﬁrmed by single crystal X-ray diffraction and supported by

spectroscopic characterization. Interestingly for 3, its solid-state

structure showed that cumap underwent hydrolysis and as a result

only the apy moiety coordinated to the fac-[Re(CO)₃Cl] unit as

opposed to 4 where the cinap Schiff base moiety remained intact.

The contrasting coordination behaviours of the aforementioned

structural analogues, cumap and cinap were rationalized using

computational studies at the DFT level. Furthermore, biological

interaction studies of 1–4 were conducted in the form of CT-DNA

titrations, gel electrophoresis and supported by molecular docking

simulations.

2.2. 1,5-Dimethyl-2-phenyl-4-{[4-(propan-2-yl)benzylidene]amino}-

1,2-dihydro-3H-pyrazol-3-one (cumap)

Equimolar amounts of 4-aminoantipyrene (0.25 g; 1.23 mmol)

and cuminaldehyde (0.87 cm³; 1.23 mmol) were added to metha-

nol (10 cm³) and heated at a reﬂux for 3 h. Thereafter, the light

orange solution was concentrated to an oily residue. Upon cooling

to room temperature, the substance precipitated. Afterwards, 15

cm³of petroleum ether was added to the precipitate and the resul-

tant mixture was stirred for an hour. Then a cream-coloured pre-

cipitate was ﬁltered, washed with cold methanol (3 cm³) and

hexane (3 cm³), then allowed to dry in a desiccator. Yield = 68%;

M.P. 191–196 °C; IR (

m m(C@O) 1654 (s), m(C@N) 1595

_max/cm^ꢁ1):

(s),

m

(C@C) 1447 (m). ¹H NMR (295 K/ppm, see Fig. 2): 9.57 (s,

1H, H1), 7.74 (d, 2H, H14, H18), 7.54 (t, 2H, H15, H17), 7.42–7.32

(m, 5H, H3, H4, H9, H10, H16), 3.18 (s, 3H, H19, H19⁰, H19⁰⁰), 2.95

(m, 1H, H6), 2.46 (s, 3H, H20, H20⁰, H20⁰⁰), 1.24 (d, 6H, H7, H7⁰,

H7⁰⁰, H8, H8⁰, H8⁰⁰). UV–Vis (DCM, k_max, M^ꢁ1cm^ꢁ1)): 332 nm

(e

(48,374).

2. Experimental

2.1. Materials and methods

2.3. 1,5-Dimethyl-2-phenyl-4-{[3-phenylprop-2-en-1-ylidene]

amino}-1,2-dihydro-3H-pyrazol-3-one (cinap)

Rhenium pentacarbonyl chloride, electrochemical grade tetra-

butylammoniumtetraﬂuoroborate, 2-aminophenol, 4-aminoan-

tipyerene, cinnamaldehyde, cuminaldehyde, phosphate buffered

saline (PBS) tablets and calf thymus (CT) DNA were obtained from

Sigma Aldrich. DNA chips, DNA 1 K gel and stain, DNA ladder and

loading buffer were all obtained within a DNA 1 K Analysis Kit

from Bio-Rad. All solvents were obtained from Merck SA. These

chemicals were used without any further puriﬁcation. The infrared

spectra were recorded on a Perkin–Elmer Spectrum 100 from 4000

to 650 cm^ꢁ1. ¹H NMR spectra were obtained using a Bruker Avance

400 MHz spectrometer. All NMR spectra were recorded in DMSO-

d₆. UV-Vis spectra were recorded using a Perkin-Elmer Lambda

Using the same experimental procedure as described in Sec-

tion 2.2, the 1:1 M condensation reaction of 4-aminoantipyrene

(0.25 g; 1.23 mmol) with cinnamaldehyde (0.16 cm³; 1.23 mmol)

was conducted methanol (10 cm³) which resulted in the isolation

of a deep orange precipitate. Yield = 63%; M.P. 188–192 °C; IR

(m m(C@O) 1637 (s), m(C@N) 1570 (s), m(C@C) 1524 (m).

_max/cm^ꢁ1):

¹H NMR (295 K/ppm): 9.42 (d, 1H, H12), 7.64 (d, 2H, H2, H6),

7.54 (t, 2H, H3, H5), 7.44–7.31 (m, 6H, H13, H14, H16, H20, H17,

H19), 7.15–6.98 (m, 2H, H18, H4), 3.18 (s, 3H, H11, H11⁰, H11⁰⁰),

2.41 (s, 3H, H10, H10⁰, H10⁰⁰). UV-Vis (DCM, k_max, M^ꢁ1cm^ꢁ1)):

(e

367 nm (38,980).

Scheme 1. Synthesis of the dinuclear rhenium(I) compounds 1 and 2.

M.B. Ismail et al. / Inorganica Chimica Acta 477 (2018) 257–269

259

Scheme 2. Synthesis of the mononuclear rhenium(I) complexes 3 and 4.

2H, H6, H6⁰), 3.51 (s, 2H, H14, H14⁰), 2.88 (s, 6H, H15A, H15B,

H15C, H15A⁰, H15B⁰, H15C⁰), 1.60 (s, 6H, H16A, H16B, H16C, H16A⁰,

H16B⁰, H16C⁰); UV-Vis (MeOH, k_max

(e,

M^ꢁ1cm^ꢁ1)): 282 nm

(11810); 338 nm (5793); 390 nm (5081).

2.6. fac-[Re(CO)₃(apy)Cl] (3)

The 1:1 M coordination reaction of cumap (92.2 mg; 276

lmol)

with [Re(CO)₅Cl] (100 mg; 276 mol) were done in toluene (20

l

cm³). The reaction mixture was heated until reﬂux for 4 h under

inert conditions. The resultant solution was cooled in ice bath fol-

lowed by ﬁltration of a brown precipitate. This precipitate was

washed with diethyl ether (3 cm³) and allowed to dry in a desicca-

tor. Then, the precipitate was dissolved in tetrahydrofuran and lay-

ered with n-hexane to afford rectangular crystals after 3 days.

Yield = 31%; M.P. > 350 °C; Conductivity (DMF, 10^ꢁ3M) = 12.22

Fig. 2. The atomic numbering scheme of cumap.

2.4. (m-ciap)₂[Re(CO)₃]₂(1)

A 1:1 stoichiometric ratio of cinnamaldehyde (0.05 cm³; 415

X

^ꢁ1cm^ꢁ2mol^ꢁ1

2019 and 1872;

;

IR

v

(m v(NAH) 3112 w; v(C„O)_fac

_max/cm^ꢁ1):

(C@O) 1608 m; ¹H NMR (295 K/ppm): 7.68–

l

mol) and 2-aminophenol (45.2 mg; 415 mol) was added to

toluene (20 cm³) and heated to reﬂux for 5 h. To the cooled solu-

tion, a molar equivalent of [Re(CO)₅Cl] (150 mg; 415 mol) was

l

7.62 (m, 2H, H11, H13), 7.58 (t, 1H, H12), 7.46–7.41 (m, 2H, H10,

H14), 6.55 (d, 1H, N1-H1A), 5.65 (d, 1H, N1-H1B), 2.43 (s, 3H,

H8A, H8B, H8C), 2.10 (s, 3H, H7A, H7B, H7C); UV-Vis (MeOH, k_max

l

added and the reaction mixture was heated to reﬂux for a further

3 h under an inert N₂atmosphere. Thereafter, the cooled solution

was layered with n-hexane and after progressive slow diffusion,

orange crystals suitable for X-ray analysis were collected. Yield =

(e

, M^ꢁ1cm^ꢁ1)): 258 nm (46,274); 276 nm (38,815); 329 nm (9222).

2.7. fac-[Re(CO)₃(cinap)Cl] (4)

24%; M.P. > 350 °C; Conductivity (DMF, 10^ꢁ3M) = 22.04

X

^-1cm^ꢁ2

-

mol^ꢁ1; IR (

(m),

v

_max/cm^ꢁ1):

v(C„O)_fac2018 and 1874 (s), v(C@N) 1610

A solution of cinap (87.8 mg; 276 lmol) and [Re(CO)₅Cl] (100

mg; 276 l

mol) in toluene (20 cm³) was stirred at reﬂux tempera-

v

(C@C) 1565 (m); ¹H NMR (CDCl₃, 295 K/ppm): 8.52 (d, 2H,

H7, H7⁰), 7.65–7.57 (m, 4H, H2, H2⁰, H5, H5⁰), 7.50–7.38 (m, 8H,

H3, H3⁰, H4, H4⁰, H8, H8⁰, H9, H9⁰), 7.23 (d, 2H, H11, H11⁰), 7.24–

7.15 (m, 4H, H12, H12⁰, H13, H13⁰), 7.02 (d, 2H, H15, H15⁰), 6.91

ture under N₂for 4 h. The resultant reaction mixture was allowed

to cool to room temperature. Then several aliquots of the reaction

mixture were layered with n-hexane. After six days of slow diffu-

sion of hexane into the reaction mixture, X-ray quality crystals

were obtained. Yield = 22%; M.P. > 350 °C; Conductivity (DMF,

(t, 2H, H14, H14⁰); UV-Vis (MeOH, k_max, M^ꢁ1cm^ꢁ1)): 283 nm

(e

(23,501); 306 nm (21,822); 345 nm (17,068), 461 nm (7396).

10^ꢁ3M) = 14.55

2012 and 1888 s;

X

v

^ꢁ1cm^ꢁ2mol^ꢁ1

(C@O) 1600 m;

;

v

IR

(m v(C„O)_fac

max/cm^ꢁ1):

(C@N) 1584 m; ¹H NMR (295

K/ppm): 8.71 (d, 1H, H12), 7.82 (d, 1H, H2), 7.72–7.62 (m, 5H, H6,

H5, H3, H13, H14), 7.58–7.48 (m, 6H, H4, H16, H17, H18, H19,

H20), 3.47 (s, 3H, H11A, H11B, H11C), 2.76 (s, 3H, H10A, H10B,

2.5. (m-(cuap)₂[Re(CO)₃] (2)

The same experimental procedure was adopted as in Section 2.4,

the following reagents were utilized: cuminaldehyde (0.06 cm³;

H10C); UV-Vis (MeOH, k_max(e

, M^ꢁ1cm^ꢁ1)): 235 nm (3675); 241

nm (3347); 288 nm (3597); 350 nm (4681).

415

(150 mg; 415

were isolated. Yield = 32%; M.P. > 350 °C; Conductivity (DMF,

10^ꢁ3M) = 20.95 ^ꢁ1cm^ꢁ2mol^ꢁ1; IR ( _max/cm^ꢁ1):

(C„O)_fac2022

and 1882 (s), (C@N) 1686 (m),

(C@C) 1610 (s); ¹H NMR (295

lmol), 2-aminphenol (45.2 mg; 415

lmol) and [Re(CO)₅Cl]

lmol). Bright yellow, parallelogram-shaped crystals

2.8. X-ray crystallography

X

v

The X-ray data for 1–4 were recorded on a Bruker Apex Duo

equipped with an Oxford Instruments Cryojet operating at 100 K

and an Incoatec microsource operating at 30 W power. Crystal

and structure reﬁnement data for 3 and 4 are given in Table 1 while

K/ppm): 8.71 (s, 2H, H7, H7⁰), 7.88 (d, 2H, H1, H1⁰), 7.85 (d, 2H,

H4, H4⁰), 7.42 (d, 2H, H9, H9⁰), 7.38 (d, 2H, H11, H11⁰), 7.32 (d, 2H,

H10, H10⁰), 7.21 (t, 2H, H5, H5⁰), 7.04 (d, 2H, H13, H13⁰), 6.93 (t,

260

M.B. Ismail et al. / Inorganica Chimica Acta 477 (2018) 257–269

their selected bond lengths and angles are given in Tables 2 and 3,

respectively. Only low resolution structures could be attained for

the dinuclear compounds 1 and 2, see Figs. S10 and S11. The data

was collected with Mo Ka (k = 0.71073 Å) radiation at a crystal-to-

detector distance of 50 mm. The following conditions were used

for the Bruker data collection: omega and phi scans with exposures

taken at 30 W X-ray power and 0.50° frame widths using APEX2

[11]. The data were reduced with the programme SAINT [11] using

outlier rejection, scan speed scaling, as well as standard Lorentz

and polarisation correction factors. A SADABS [11] semi-empirical

multi-scan absorption correction was applied to the data [12].

Direct methods, SHELXS-97 [13] and WinGX [14] were used to

solve the structure. All non-hydrogen atoms were located in the

difference density map and reﬁned anisotropically with SHELXL-

97 [13]. All hydrogen atoms were included as idealised contribu-

tors in the least squares process. Their positions were calculated

using a standard riding model with C-H_aromaticdistances of 0.93 Å

Table 1

Crystal data and structure reﬁnement data for 3ꢀC₄H₈O and 4ꢀCH₂Cl_2.

3ꢀC₄H₈O

4ꢀCH₂Cl₂

Chemical formula

Formula weight

Temperature(K)

Crystal system

C₁₈H₂₁ClN₃O₅Re

581.03

100(2)

Monoclinic

P2₁/c

C₂₅H₂₃Cl₅N₃O₄Re

792.91

100(2)

Triclinic

P-1

and U_iso= 1.2 U_eq.

Space group

2.9. Computational details

Unit cell dimensions (Å, °)

a = 11.0029(6)

b = 17.4560(10)

c = 12.2221(8)

b = 116.710(2)

a = 9.3823(5)

b = 9.6238(5)

c = 16.2075(9)

Gaussian 09 W was used to conduct the computational experi-

ments [15]. Geometry optimizations of the rhenium complexes

were achieved through DFT calculations using the B3LYP func-

tional, with an accompanying hybrid basis set viz. the 6-311G⁺⁺

(d, p) basis set was applied to all the C, H, N, O, and Cl atoms and

the LANL2DZ basis set applied to the metal centre [16,17]. The

crystal structures were used as starting structures with solvents

of recrystallization removed prior to any calculations. Using the

optimized structures of the metal complexes, frequency calcula-

tions conﬁrmed that the respective structures are at global minima

on the potential energy surfaces and from these, the vibrational

energies were attained [18]. The close correlation between the

optimized and geometrical parameters of compounds 3 and 4 is

emphasized by the low calculated RMSD values (3: 0.0986 Å and

4: 0.5700 Å) Fig. 3. The larger RMSD value of 4 is mainly attributed

to the ﬂexible degree of rotation of the CAC single bonds within its

chelating ligand. The molecular docking simulations were con-

ducted using Patchdock Beta version 1.3 and the different data sets

were reﬁned using Firedock [19]. All data processing was con-

ducted with Yasara view [20]. The optimized conformers of the

respective rhenium compounds were used as the small molecules

while the water molecules of recrystallization were removed from

the B-DNA crystal structure (LOX) (PDB ID: 1F8N) and the resultant

conformer was used as the receptor.

a

= 75.865(3)

b = 84.750(2)

= 86.911(2)

c

V(Å³)

2097.0(2)

4

1.840

5.955

1128

1412.41(13)

2

1.864

4.812

Z

Density (calc.) (Mg/m³)

Absorption coefﬁcient (mm^ꢁ1

F(0 0 0)

)

772

h range for data collection (°)

Index ranges

2.072–27.385

ꢁ14/14;

ꢁ22/22;

ꢁ15/8

1.300–28.220

ꢁ8/12;

ꢁ12/12;

ꢁ21/21

28,980

Reﬂections measured

18,242

Observed data [I > 2

r

(I)]

4678

6944

Data/restraints/parameters

Goodness of ﬁt on F²

R, wR²

4678/4/255

1.044

0.0191, 0.0388

6944/0/345

1.050

0.0237, 0.0419

Table 2

Selected experimental and optimized bond lengths [Å] and bond angles [°] of 3.

Experimental

Optimized

Re(1)-N(1)

Re(1)-O(1)

Re(1)-Cl(1)

Re(1)-C(15)

Re(1)-C(16)

Re(1)-C(17)

N(2)-C(9)

N(2)-C(3)

N(3)-C(6)

O(1)-Re(1)-(N1)

O(1)-Re(1)-C(15)

N(1)-Re(1)-C(16)

Cl(1)-Re(1)-C(17)

2.244(2)

2.198(2)

2.4819(6)

1.901(2)

1.926(3)

1.910(2)

1.433(3)

1.374(3)

1.375(3)

80.03(7)

176.91(9)

171.19(9)

173.89(7)

2.3386

2.2680

2.5131

1.9021

1.9143

1.9105

1.4264

1.3745

1.3891

76.88

2.10. UV-Vis DNA binding titrations

Calf thymus (CT)-DNA was prepared in phosphate-buffered sal-

ine (PBS) at a pH of 7.2 and the DNA binding studies of the metal

complexes were carried out in MeOH. The DNA solution afforded

a ratio of UV absorbance at 260 and 280 nm of approximately

1.9:1 which indicates that the DNA was sufﬁciently free of protein

[21]. In addition, the CT-DNA concentration per nucleotide was

determined by UV absorption spectroscopy using the molar

absorption coefﬁcient (e₂₆₀@ 6600 M^ꢁ1cm^ꢁ1) [21]. The resultant

stock DNA solution was stored at a temperature of 4 °C and used

within 2 days after preparation. Prior to any additions of CT-DNA

aliquots, separate methanolic standard solutions of 1–4 were mon-

itored spectrophotometrically during various time intervals over

24 h to conﬁrm the stabilities of the respective rhenium com-

pounds in methanol. Afterwards, the respective metal complexes

and CT-DNA solutions were incubated for 24 h at 25 °C prior to

any UV-Vis measurements [22]. Furthermore, the UV-Vis spectra

were attained for a solution of metal complex (of known concen-

tration in methanol) with varying nucleotide concentrations (0–

174.14

170.89

172.05

Table 3

Selected experimental and optimized bond lengths [Å] and bond angles [°] of 4.

Experimental

Optimized

Re(1)-N(2)

Re(1)-O(1)

Re(1)-Cl(1)

Re(1)-C(21)

Re(1)-C(22)

Re(1)-C(23)

N(2)-C(12)

C(12)-C(13)

C(13)-C(14)

N(6)-C(1)

N(5)-C(9)

N(6)-C(7)

O(1)-Re(1)-(N2)

O(1)-Re(1)-C(23)

N(2)-Re(1)-C(21)

Cl(1)-Re(1)-C(22)

2.254(2)

2.165(2)

2.4648(6)

1.914(3)

1.936(2)

1.893(3)

1.304(3)

1.429(3)

1.341(3)

1.437(3)

1.354(3)

1.355(3)

78.57(6)

174.14(9)

169.6(1)

173.86(7)

2.3032

2.2658

2.5101

1.9188

1.9149

1.8956

1.3018

1.4280

1.3543

1.4247

1.3921

1.3674

75.68

200 lM) in PBS. The data attained from the absorption titration

experiments were ﬁtted to the following equation to obtain the

intrinsic binding constant (K_b):

175.66

169.28

174.39

½DNAꢂ=ðꢀ_a

ꢁ

ꢀ_fÞ ¼ ½DNAꢂ=ðꢀ_b

ꢁ

ꢀ_fÞ þ 1=K_bðꢀ_b

ꢁ

ꢀ_f

Þ

M.B. Ismail et al. / Inorganica Chimica Acta 477 (2018) 257–269

261

Fig. 3. Overlay structures of the optimized (red) and crystal structures (blue) for compounds 3 (A) and 4 (B). (For interpretation of the references to colour in this ﬁgure

legend, the reader is referred to the web version of this article.)

In this equation, [DNA] is the concentration of DNA in base

pairs, e_ais the extinction coefﬁcient of the observed absorption

band at the given DNA concentration (corresponds to A_obs/[com-

plex]), e_fis the extinction coefﬁcient of the complex free in solu-

tion, and e_bis the extinction coefﬁcient of the complex when

the ESP surface of cumap shows that its phenylimino moiety is

approaching negative ESP values whereas the 1-((imino)ethenyl)

benzene moiety of cinap generally exhibits ESP neutrality. This

implies that the imino bond of cumap is weaker than that of cinap.

Therefore, the main computational factor that rationalises the

experimental instability of cumap is denoted to cumap’s phenylim-

ino moiety which generates a higher inductive inﬂuence as com-

pared to the 1-((imino)ethenyl)benzene moiety of cinap. In turn,

the experimental stability of cinap is mainly emphasized by its

computed frontier orbitals which shows a more extended pi-delo-

calized system over its 1-((imino)ethenyl)benzene moiety; ulti-

mately leading to its propensity not to undergo hydrolysis [25,26].

The low molar conductivity values of 1–4 (12.22–20.95 X^ꢁ1

cm^ꢁ2mol^ꢁ1) afﬁrm that the metal compounds exist as neutral

molecules in solution. Additionally, these metal compounds dis-

play moderate solubility in alcohols and mid to high solubility in

polar aprotic solvents. The low resolution X-ray structures for 1

and 2 are nearly identical whereby two geometrically equivalent

imino moieties coordinate in a bidentate manner via N_imino, bridg-

ing O_phenolatemoieties. Consequently, the ¹H NMR spectra of these

metal compounds essentially show non-distinguishable proton

signals for their chelators due to chemical equivalence, see

Figs. S3, S5, S7 and S9. The proton spectra for 1–4 show well

resolved peaks, illustrating the diamagnetic nature of each metal

compound. Imine signals are observed at 8.52 (doublet), 8.71 (sin-

glet) and 7.82 ppm (doublet) for 1, 2 and 4, respectively. An upﬁeld

shift of the imine peak is noted between the free-ligand cinap

(9.42 ppm) and its corresponding metal complex (4). A comparison

between the free-ligand cumap and its metal complex (3) afﬁrms

that this Schiff base has cleaved given that the free-ligand’s imino

singlet at 9.57 ppm is not evident in the proton spectrum of 3. A

similar conclusion is drawn when comparing the IR spectra of 3

and its free-ligand, cumap, where NAH vibrational bands in the

region of 3100 cm^ꢁ1are visible for the metal complex but not for

the free-ligand. Mutual for all solid-state infrared spectra of the

metal compounds are the high-intensity carbonyl stretches

between 1872 and 2022 cm^ꢁ1, see Figs. S2, S4, S6 and S8. In addi-

tion, the aminoantipyrene ketonic bonds vibrate at cumap, cinap, 3

and 4 at 1654, 1637, 1608 and 1600 cm^ꢁ1, respectively, with the

vibrational bands shifts between free ligands and their respective

complexes afﬁrming that coordination occurred. Additionally,

fully bound to DNA. A plot of [DNA]/[e_a

ꢁ

e

_f] versus [DNA] gave a

slope 1/[e_{a f}] and Y intercept equal to 1/K_b[e_be

ꢁe

ꢁ

_f]. The intrinsic

binding constant (K_b) is estimated to be the ratio of the slope to

the intercept [22].

Gel electrophoresis experiments were conducted using a Bio-

Rad Experion^TMAutomated Electrophoresis System which includes

the Experion^TMelectrophoresis station with its accompanying

peripheral devices, priming station and vortex station II. Data cap-

turing and manipulation were achieved with the aid of the Expe-

rion software. A set of 2 mM CT-DNA solutions prepared via the

same method as speciﬁed before were incubated with varying con-

centrations of the respective metal compounds for a period of 24 h.

Utilizing the DNA 1 K Analysis Kit and the protocols applied to chip

preparation, the respective samples were loaded and run on the

instrument.

3. Results and discussion

3.1. Synthesis, characterization and computational studies

The formation of 1 and 2 is a typical illustration of template

synthesis where condensation and coordination reactions occur

in a complementary manner [23,24]. Characteristic to this class

of synthetic approach, the isolation of the free Schiff bases, viz. ciap

or cuap could not be achieved in sufﬁcient purity. The bridging

bidentate chelates of these dinuclear rhenium(I) compounds act

as bidentate monoanionic moieties. In the case of the mononuclear

rhenium(I) complexes 3 and 4, the cumap Schiff base hydrolysed

upon coordination leading to formation of the neutral bidentate

coordination mode of N_aminoO_ketonefor the 4-aminoantipyrine moi-

ety (in 3) whilst cinap remained intact and coordinated through its

neutral N_iminoO_ketonedonor pair (in 4). The diverse coordination pat-

terns of the cinap and cumap ligands were rationalized through

DFT studies (see Table 4).

Although, the optimized structure of cinap has a higher com-

puted total energy than cumap, ligand cinap has a lower calculated

band-gap energy. In addition, the bond angle C_apy-N_imino-C_imino

(where apy = aminoantipyrene moiety) of cinap is closer to the ide-

alized bond angle of 120° for a bond containing a bridging imino

nitrogen. More interestingly, the Natural Population Analysis

(NPA) charges of the imino carbon and nitrogen atoms suggest that

the isopropylbenzene moiety of cumap induces a stronger induc-

tive effect as compared to the 1-ethenylbenzene moiety of cinap.

This is further evident from visual inspection of the Electrostatic

Potential (ESP) surfaces of the two Schiff base conformers whereby

medium to strong v(C@N) and v(C@C) stretches are common to 1

and 2 [between 1565 and 1684 cm^ꢁ1].

The electronic spectra for 1–4 display several absorption bands

between 235 and 461 nm. The electronic transitions between 200

and 350 nm are ascribed as intraligand

p

?

p

⁄ transitions while

those above 350 nm are ascribed to metal-to-ligand charge trans-

fer bands [27]. Common to most facial tricarbonylrhenium(I) com-

pounds, no metal-based electronic transitions are observed due to

their low-spin d⁶-electronic conﬁguration. Facial tricarbonylrhe-

262

M.B. Ismail et al. / Inorganica Chimica Acta 477 (2018) 257–269

Table 4

Summary of the computational data for cumap and cinap.

Ligand cumap

Ligand cinap

Total Energy (eV)

ꢁ2.8681 ꢃ 10⁴

ꢁ0.473

0.121

ꢁ2.7565 ꢃ 10⁴

ꢁ0.418

0.106

NPA charge of N_imino

NPA charge of C_imino

Band-gap energy (eV)

Electrostatic potential surfaces

4.04

3.40

HOMO surfaces

LUMO surfaces

C_apy-N_imino-C_imino(°)

121.047

120.535

nium(I) compounds tend to exhibit the redox behaviour: Re(I) ?

Re(II) + e^ꢁ[28] and these irreversible processes are observed for

1–4 at

E_pa= 0.96 V, 1.04 V, 1.09 V and 1.11 V, respectively

Figs. S12–S16. The values are within range for literature reported

rhenium(I) compounds: 0.83–1.60 V vs Ag|AgCl [29,30]. Further

irreversible processes at E_pa= 0.68 V (for 1), E_pa= 0.58 V (for 2),

E_pc= 0.44 V (for 3) and E_pc= 0.64 V (for 4) are tentatively assigned

as ligand induced redox processes. Additionally, the linearity of the

ﬁgure insets [plots of I_pa(anodic peak current) vs v^1/2(square root

of scan rate)] conﬁrm that the redox processes adhere to diffusion

controlled behaviour.

3.2. Description of crystal structures 3.C₄H₈O and 4.2(CH₂Cl₂)

Compound 3 co-crystallizes with a tetrahydrofuran molecule of

recrystallization in the space group P2_1/cwhereas 4 displays an

inverted symmetry (P-1 space group) and is accompanied with

two dichloromethane molecules of recrystallization Figs. 4 and 5.

Both metal compounds are monomeric, neutral and the crystal lat-

tice of 3.C₄H₈O is stabilized by a series classical hydrogen bonding

patterns. In fact, Cl1 H1A-N1 [2.3117 Å] intermolecular hydrogen-

bond between neighbouring molecules of 3 allows them to pack in

columns aligned with the [c]-axis, see Fig. S17. Ultimately this

affords a polymeric network structure which is further supported

by classical hydrogen bonding interactions between the other

amino hydrogens and its solvent molecules [O5ꢀ ꢀ ꢀH1B = 2.3117

Å]. In the case of 4, typical pi-stacking interactions (below 3.5 Å)

are observed between the C15-C20 phenyl ring and sp²-hybridized

C13 and C14 atoms with respective centroid to carbon distances of

3.458 and 3.383 Å. These interactions allow the molecules of 4 to

afford pi-stacked columns parallel to [b]-axis and these columns

are supported by non-classical hydrogen bonding with the bridg-

ing dichloromethane molecules.

Fig. 4. ORTEP view of compound 3ꢀC₄H₈O showing 50% probability displacement

ellipsoids and the atom labelling.

constrained bite angles for 3 and 4, respectively: O1-Re1-N1 =

80.03(7)° and O1-Re1-N2 = 78.57(6)° which are substantially smal-

ler than the expected bond angles of 90°. This distortion induces

deviation from linearity for the O1-Re1-C15 [176.91(9)°], N1-

Re1-C16 [171.09(9)°], Cl1-Re1-C17 [173.89(7)°] bond angles for 3

and O1-Re1-C23 [174.14(9)°], N2-Re1-C21 [169.6(1)°], Cl1-Re1-

C22 [173.86(7)°] for 4. The ﬂexibility in the aliphatic N2-C12-

C13-C14 moiety in 4 induces the C15-C20 phenyl ring to form a

dihedral angle of 14.49° with respect to the C21C23N2O1 basal

The central cause for the deviation from perfect octahedral

symmetry stems from 5-membered chelate rings which affords

M.B. Ismail et al. / Inorganica Chimica Acta 477 (2018) 257–269

3.3. DNA interaction studies

263

3.3.1. UV-Vis DNA binding of 1–4

Steady state absorption spectroscopy has been used to study

the binding mode and the strength of the binding interaction

between the rhenium(I) complexes and CT-DNA. An intercalating

binding mode generally results in hypochromism coupled with a

distinct red shift of the absorption band [40]. Upon the addition

of CT-DNA, compounds 1, 2 and 4 display 7–19% hyperchromism

of their absorption maxima (1: 283 nm, 2: 282 nm and 4: 350

nm) with negligible red or blue shifts, implying a lack of signiﬁcant

structural contusions of the DNA Figs. 6–9. For 3, a slight blue shift

(261–258 nm) is seen upon the ﬁrst CT-DNA aliquot addition, sig-

nifying an initial destabilization of the DNA helix [41] but there-

after, no further shifting is observed during hyperchromic steps.

Isosbestic points are noted at 310 and 378 nm for 1 and 279 and

380 nm for 4. Hence, the electronic spectral changes for 1–4 are

consistent with metal complexes of a groove-binding nature [42–

44]. Furthermore, the calculated K_bvalues [1: 8.48 ( 0.42) x10³

M^ꢁ1, 2: 5.76 ( 0.73) ꢃ 10³M^ꢁ1, 3: 4.38 ( 0.39) ꢃ 10⁴M^ꢁ1, 4: 1.10

( 0.55) ꢃ 10⁵M^ꢁ1] are within literature range for other Schiff base

containing, groove-binding metal complexes (10³M^ꢁ1> K_b> 10⁶

M^ꢁ1) [45–48].

Fig. 5. ORTEP view of 4.2 (CH₂Cl₂) showing 50% probability displacement ellipsoids

and the atom labelling.

3.3.2. Molecular docking

Molecular docking simulations of 1–4 were conducted to fur-

ther ascertain the compounds’ modes of interactions to DNA. The

most stable adducts of 1–4 with B-DNA afﬁrm their groove-binding

nature where all four compounds dock within the minor groove

(base pair region: AT, GC, CG, GC, CG) Figs. 10–13. The large dinu-

clear structures of 1 and 2 (van der Waal’s radii: 10.270 Å and

10.113 Å respectively) prevents manoeuvrability within B-DNA

grooves and attests for their low intrinsic binding constants. The

mononuclear compounds 3 and 4 (van der Waal’s radii: 7.212 Å

and 9.835 Å respectively) display similarity where the bidentate

chelators interact with the hydrophobic B-DNA interior and the

more polar carbonyl co-ligands situate towards the hydrophilic

exterior. Compounds 3 and 4 are stabilized primarily by hydrogen

bonding [3: antipyrene C_methylatom with a guanine N atom (2.472

Å) and 4: antipyrene C_methylatom with a cytosine O_ketoneatom

(2.477 Å)]. Further close contacts are observed for 4 [Cl atom with

an O_phosphateatom (2.488 Å); axial O_carbonylwith a carbon-phosphate

backbone C atom (2.476 Å) and an alkyl C atom with a guanine

N_amineatom (2.450 Å)]. This stronger bonding pattern for 4 results

in a lower total global energy than that of 3 (3: ꢁ42.79 kJ.mol^ꢁ1

and 4: ꢁ52.34 kJ.mol^ꢁ1) and corroborates 4⁰s higher intrinsic bind-

ing constant (K_b) quantiﬁed in the previous section.

plane. The optimized conformer of 4 has similar bond distances for

N2-C12 [1.304(3) Å for experimental and 1.3018 Å for optimized],

C12-C13 [1.429(3)° Å for experimental and 1.4280 Å for optimized]

and C13-C14 [1.341(3)° Å for experimental and 1.3543 Å for opti-

mized] which supports the sp²-hybridized nature of the atoms. A

similar phenomenon was found in the optimized conformer of 3

where a plane-to-plane angle of 21.68° was found between the

C21C23N2O1 basal plane and C15-C20 phenyl ring. The bond order

of the N2-C9 (for 3) and N6-C1 (for 4) is clearly established based

on their bond lengths and these single bonds [1.433(3) Å for 3 and

1.437(3) for 4] allow wider dihedral angles [43.95 for 3 and 74.79°

for 4] between pyrazole and appending phenyl rings. The same also

accounts for the Schiff base functionality in 4 [N2-C12 = 1.304(3)

Å] which is shorter than the intracyclic pyrazole CAN bonds of 3

[N2-C3 = 1.374(3) Å and N3-C6 = 1.375(3) Å] and 4 [N5-C9 =

1.354(3) Å and N6-C7 = 1.355(3) Å].

Metal compounds 3 and 4 coordinate in a similar manner via

ketonic oxygen atoms and nitrogen donors [amino (for 3) or imino

(for 4)] donor atoms from their respective antipyrene moieties. In

comparison to 3, a similar coordination mode was observed for the

rhenium(V)

compound:

cis-[ReBr₂(pap)(H₂pap)(PPh₃)](ReO₄)

(H₂pap = 4-aminoantipyrene) where the 4-aminoantipyrene moi-

ety acts a neutral bidentate chelator [31]. The Re-N_amino[3: 2.244

(2) Å] and Re-O_ketonic[3: 2.198(2)] bond lengths for 3 are both

slightly longer than the equivalent bonds of the cited rhenium(V)

compound [2.19(1) and 2.102(9) Å] and is attributed to the less

acidic nature of the fac-[Re(CO)₃]⁺core. The coordination bonds

for 3 and 4 [4: Re-N_imino= 2.254(2) Å; 4: Re-O_ketonic= 2.165(2) Å]

compare well with analogous tricarbonyl rhenium(I) bonds found

3.3.3. Gel electrophoresis

The virtual gel stains of compounds 1 and 2 indicate that both

compounds cleave CT-DNA best at lowest concentration (lanes 1

and 4) with 2 having a similar cleaving activity at 10 and 50 lM

concentrations (lanes 4 and 5) Fig. 14. A similar scenario is

observed for 3 and 4 Fig. 15 however, for 3 at lowest concentration

(lane 1), the cleaving ability is far greater than that observed for

compounds 1, 2 and 4. This is possibly due to a combination of

its smaller size (less steric effects) and higher afﬁnity for DNA than

the dinuclear complexes 1 and 2 or alternatively, 3 is able to pro-

duce a higher concentration of DNA-damaging radical species than

the other metal compounds at lower concentrations [46]. Further

to this, a trend of decreasing cleaving activity is noted for all four

Re(I) compounds with increasing concentration, possibly due to

aggregation at elevated concentrations [47]. The difference

observed in the lanes 7 representing the CT-DNA only, is ascribed

to variable stirring rates employed when the commercially

attained CT-DNA ﬁbres were dissolved in PBS buffer.

in literature, Re-N_amino: 2.196(2) – 2.244(4) Å [32], Re-N_imino

:

2.208(3)–2.258(4) Å [33,34] and Re-O_ketonic: 2.164(5)–2.195(2) Å

[35,36]. Indicative to literature trends, the rhenium(I) carbonyl

bonds [3: Re1-C15 = 1.901(2) Å, Re1-C16 = 1.926(3) Å, Re1-C17 =

1.910(2) Å and 4: Re1-C21 = 1.914(3) Å, Re1-C22 = 1.936(2) Å,

Re1-C23 = 1.893(3) Å] are different which is largely ascribed to

the different degree of trans-inﬂuence imposed on their carbonyl

co-ligands [30,37]. Furthermore, the rhenium(I) to chloride coordi-

native bonds, Re-Cl [3: 2.4819(6) Å, 4: 2.4648(6) Å] are within the

range of other reported axial Re-Cl bond lengths [2.452(4)–2.491

(2) Å] [38,39].

264

M.B. Ismail et al. / Inorganica Chimica Acta 477 (2018) 257–269

Fig. 6. The overlay UV–Vis absorption spectra of compound 1. The dashed line indicates complex 1 free of any DNA and the hypochromic effect (indicated by the arrow) is due

to progressive increments of a 5 mM CT-DNA solution.

Fig. 7. The overlay UV–Vis absorption spectra of compound 2. The dashed line indicates complex 1 free of any DNA and the hypochromic effect (indicated by the arrow) is due

to progressive increments of a 5 mM CT-DNA solution.

Fig. 8. The overlay UV–Vis absorption spectra of compound 3. The dashed line indicates complex 1 free of any DNA and the hypochromic effect (indicated by the arrow) is due

to progressive increments of a 5 mM CT-DNA solution.

M.B. Ismail et al. / Inorganica Chimica Acta 477 (2018) 257–269

265

Fig. 9. The overlay UV–Vis absorption spectra of compound 4. The dashed line indicates complex 1 free of any DNA and the hypochromic effect (indicated by the arrow) is due

to progressive increments of a 5 mM CT-DNA solution.

Fig. 10. Docked position of 1 bound to the minor groove of B-DNA: ball and stick view (left) and DNA molecular surface view (right).

gested that the main computational factor that rationalises the

experimental instability of cumap is ascribed to cumap’s

phenylimino moiety which generates a higher inductive inﬂuence

as compared to the 1-((imino)ethenyl)benzene moiety of cinap. All

the metal compounds were classed as DNA groove binders and

their DNA interaction modes were supported by molecular docking

simulations. Interestingly, the gel electrophoresis experiments

showed that the metal compounds exhibit concentration depen-

dant DNA cleavage activities.

4. Conclusion

The mono- and dinuclear facial tricarbonylrhenium(I) com-

pounds 1–4 were formed and spectroscopically characterized.

Structural elucidations of these metal compounds were conﬁrmed

by single crystal X-ray diffraction. The solid-state structure of 3

revealed that cumap underwent hydrolysis upon coordination to

the rhenium(I) centre which ultimately led to the presence of the

bidentate neutral 4-aminoantipyrine moiety in 3. DFT studies sug-

266

M.B. Ismail et al. / Inorganica Chimica Acta 477 (2018) 257–269

Fig. 11. Docked position of 2 bound to the minor groove of B-DNA: ball and stick view (left) and DNA molecular surface view (right).

Fig. 12. Docked position of 3 bound to the minor groove of B-DNA: ball and stick view (left) and DNA molecular surface view (right).

M.B. Ismail et al. / Inorganica Chimica Acta 477 (2018) 257–269

267

Fig. 13. Docked position of 4 bound to the minor groove of B-DNA: ball and stick view (left) and DNA molecular surface view (right).

Fig. 14. Lane L: DNA Ladder of known base pair lengths, Lane 1: DNA + 10 mM of 1, Lane 2: DNA + 50 mM of 1, Lane 3: DNA + 100 mM of 1; Lane 4: DNA + 10 mM of 2, Lane 5:

DNA + 50 mM of 2, Lane 6: DNA + 100 mM of 2; Lane 7: CT DNA only.

268

M.B. Ismail et al. / Inorganica Chimica Acta 477 (2018) 257–269

Fig. 15. Lane L: DNA Ladder of known base pair lengths, Lane 1: DNA + 10 mM of 3, Lane 2: DNA + 50 mM of 3, Lane 3: DNA + 100 mM of 3; Lane 4: DNA + 10 mM of 4, Lane 5:

DNA + 50 mM of 4, Lane 6: DNA + 100 mM of 4; Lane 7: CT DNA only.

[15] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman,

Acknowledgements

G. Scalmani, V. Barone, B. Mennucci, G.A. Petersson, H. Nakatsuji, M. Caricato,

X. Li, H.P. Hratchian, A.F. Izmaylov, J. Bloino, G. Zheng, J.L. Sonnenberg, M.

We are grateful to the University of KwaZulu-Natal, the

National Research Foundation of South Africa and the Deutscher

Akademischer Austauschdienst (DAAD) for ﬁnancial support.

Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y.

Honda, O. Kitao, H. Nakai, T. Vreven, J.A. Montgomery Jr, J.E. Peralta, F. Ogliaro,

M. Bearpark, J.J. Heyd, E. Brothers, K.N. Kudin, V.N. Staroverov, R. Kobayashi, J.

Normand, K. Raghavachari, A. Rendell, J.C. Burant, S.S. Iyengar, J. Tomasi, M.