Synthesis, characterization and antimicrobial activity of Fe(II), Zn(II), Cd(II) and Hg(II) complexes with 2,6-bis(benzimidazol-2-yl) pyridine ligand

Article abstract of DOI:10.1016/j.ejmech.2006.09.023

2,6-Bis(benzimidazol-2-yl)pyridine (L) ligand and complexes [M(L)Cl₂] and [Fe(L)₂](ClO₄)₂ (M = Zn, Cd, Hg) have been synthesized. The geometries of the [M(L)Cl₂] complexes were derived from theoretical calculation in DGauss/DFT level (DZVP basis set) on CACHE. The central M(II) ion is penta-coordinated and surrounded by N₃Cl₂ environment, adopting a distorted trigonal bipyramidal geometry. The ligand is tridentate, via three nitrogen atoms to metal centre and two chloride ions lie on each side of the distorted benzimidazole ring. In the [Fe(L)₂](ClO₄)₂ complex, the central Fe(II) ion is surrounded by two (3N) units, adopting a octahedral geometry. The elemental analysis, molecular conductivity, FT-Raman, FT-IR (mid-, far-IR), ¹H, and ¹³C NMR were reported. The antimicrobial activities of the free ligand, its hydrochloride salt, and the complexes were evaluated using the disk diffusion method in dimethyl sulfoxide (DMSO) as well as the minimal inhibitory concentration (MIC) dilution method, against 10 bacteria and the results compared with that for gentamycin. Antifungal activities were reported for Candida albicans, Kluyveromyces fragilis, Rhodotorula rubra, Debaryomyces hansenii, Hanseniaspora guilliermondii, and the results were referenced against nystatin, ketaconazole, and clotrimazole antifungal agents. In most cases, the compounds tested showed broad-spectrum (Gram positive and Gram negative bacteria) activities that were either more effective than or as potent as the references. The binding of two most biologically effective compounds of zinc and mercury to calf thymus DNA has also been investigated by absorption spectra.

Full text of DOI:10.1016/j.ejmech.2006.09.023

European Journal of Medicinal Chemistry 42 (2007) 205e213
http://www.elsevier.com/locate/ejmech
Original article
Synthesis, characterization and antimicrobial activity of Fe(II),
Zn(II), Cd(II) and Hg(II) complexes with 2,6-bis(benzimidazol-2-yl)
pyridine ligand
b
a
c
c
d
Naz M. Aghatabay a, , A. Neshat , T. Karabiyik , M. Somer , D. Haciu , B. Dulger
*
¨
a Department of Chemistry, Fatih University, Bu¨yu¨kc¸ekmece, 34500 Istanbul, Turkey
b Department of Chemistry, MS 602, University of Toledo, Toledo, OH 43606, USA
c Department of Chemistry, Koc¸ University, Rumelifeneri Yolu, Sariyer, 34450 Istanbul, Turkey
d Department of Biology, Canakkale Onsekiz Mart University, Canakkale, Turkey
Received 9 February 2006; received in revised form 28 July 2006; accepted 28 September 2006
Available online 26 December 2006
Abstract
2,6-Bis(benzimidazol-2-yl)pyridine (L) ligand and complexes [M(L)Cl2] and [Fe(L)2](ClO4)2 (M ¼ Zn, Cd, Hg) have been synthesized. The
geometries of the [M(L)Cl2] complexes were derived from theoretical calculation in DGauss/DFT level (DZVP basis set) on CACHE. The cen-
tral M(II) ion is penta-coordinated and surrounded by N3Cl2 environment, adopting a distorted trigonal bipyramidal geometry. The ligand is
tridentate, via three nitrogen atoms to metal centre and two chloride ions lie on each side of the distorted benzimidazole ring. In the [Fe(L)2]
(ClO4)2 complex, the central Fe(II) ion is surrounded by two (3N) units, adopting a octahedral geometry. The elemental analysis, molecular
conductivity, FT-Raman, FT-IR (mid-, far-IR), 1H, and 13C NMR were reported. The antimicrobial activities of the free ligand, its hydrochloride
salt, and the complexes were evaluated using the disk diffusion method in dimethyl sulfoxide (DMSO) as well as the minimal inhibitory
concentration (MIC) dilution method, against 10 bacteria and the results compared with that for gentamycin. Antifungal activities were reported
for Candida albicans, Kluyveromyces fragilis, Rhodotorula rubra, Debaryomyces hansenii, Hanseniaspora guilliermondii, and the results were
referenced against nystatin, ketaconazole, and clotrimazole antifungal agents. In most cases, the compounds tested showed broad-spectrum
(Gram positive and Gram negative bacteria) activities that were either more effective than or as potent as the references. The binding of two
most biologically effective compounds of zinc and mercury to calf thymus DNA has also been investigated by absorption spectra.
Ó 2006 Elsevier Masson SAS. All rights reserved.
Keywords: Antifungal; Antimicrobial; Equipotent; Metabolites; Tridentate; Trigonal bipyramidal
1. Introduction
complexes have been extensively investigated [4e6]. One of
the most attractive features of these ligands in the field of bi-
ological investigation has been their structural similarities with
the common pyrimidine and purine type nucleobases. In the
imidazole class, nucleosides of amino derivatives are biologi-
cally active site, while in the bisbenzimidazole class it is
nucleosides of the ring substitutions and bridging sections that
are becoming well known bioactive site [7e9]. The interaction
of metal ions with biologically active ligands, for instance in
drugs, is a subject of great interest. Some of the biologically
active complexes do act via chelation or according to prefer-
ences dictated by the hardesoft theory of acids and bases,
Since the use of cisplatin e [Pt(NH3)2Cl2] as an effective
anticancer drug, the interest toward transition metal complexes
containing N-donor ligands has increased in order to obtain
metal-based drugs exhibiting a high biological activity to-
gether with a reduced toxicity [1e3]. In this respect, benz-
imidazole derivatives together with their transition metal
* Corresponding author. Tel.: þ90 212 8663300; fax: þ90 212 8663402.
E-mail address: natabay@fatih.edu.tr (N.M. Aghatabay).
0223-5234/$ - see front matter Ó 2006 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ejmech.2006.09.023

206
N.M. Aghatabay et al. / European Journal of Medicinal Chemistry 42 (2007) 205e213
but for most, little is known about how metal coordination
influences their activity [10e12]. The most successful
compounds seem to be those that interfere with the construc-
tion of the bacterial cell wall, the process of protein synthesis,
and replication or transcription of DNA. In this respect, the
metal oxidation state, the type and the number of donor atoms,
as well as their relative disposition within the ligand are major
factors determining structureeactivity relationship of metal
complexes. These studies are also important for the probes
of nucleic acid structure and in determining the mechanism
of metal ion toxicity. There has been substantial interest in un-
derstanding the binding properties of metal complexes, with
biomolecules such as nucleic acids, proteins and lipids [13].
As borderline elements the iron(II) and zinc(II) are the most
abundant ubiquitous and essential trace elements in biological
system they serve more biological roles than any other metal
ions. They have been identified as catalytic components of
many enzymes and play several vital structural roles in a large
number of proteins, transportations and transcription factors.
Their binding properties with the biomolecules, particularly
to the oxygen and nitrogen donor centres, serve their bio-
functionality. On the other hand, nonessential cadmium(II)
and mercury(II) are very soft acids with very high toxicity.
Their salts have a long history of use as antibacterial agents.
One of the best illustrations of these metal ions principle in
biochemistry is provided by the metallothionein. Metallothio-
nein is a generic name for a super family of ubiquitous low
molecular weight metalloproteins possessing a unique type
of sulphur-based metal cluster. Vital roles for this pleiotropic
protein result in its involvement in homoeostasis of essential
trace metals such as zinc, or sequestration of the environmen-
tal toxic metals namely cadmium and mercury. Herein, we de-
scribe the structural, biological activity and DNA binding
properties of this very important planar tridentate ligand and
its transition metal complexes. Only few information has
been reported for the [Fe(L)2](ClO4)2 and [Cd(L)Cl2] com-
plexes lacking any detailed spectroscopic study or biological
investigation [14e16].
general procedures for the preparation of these compounds
are given in Section 6. The spectroscopic data from the
FT-Raman, FT-IR, 1H and 13C NMR together with molar con-
ductivity measurements do provide useful information for their
formation and structural characterization. Optimized struc-
tures of the ligand and [Zn(L)Cl2] complex are presented in
Figs. 1 and 2. Required analytical data and physical properties
are summarized in Table 1.
3. Theoretical calculation
The structure of the ligand and the [Zn(L)Cl2] complex was
refined by performing a geometry optimization calculation in
DGauss/DFT (density functional theory) available on CACHE
work system pro version 6.1.10 package program. The
B88-PW91 with the widely-used DZVP basis set, uses Beck’s
[18] exchange functional and Perdew’s [19] correlation func-
tional of the generalized gradient approximation (GGA). The
theoretical calculation of the off-resonance 13C NMR spec-
trum of the ligand was carried out using NMR simulation
gNMR 4.1 program at version level 4.1.
4. Biological data
Gentamycin inhibits protein synthesis by binding to ribo-
somal subunit, nystatin works by binding to sterols in the fun-
gal cellular membrane altering the permeability to allow
leakage of the cellular contents and destroying the fungus,
ketoconazole inhibits the growth of fungal organisms by interfer-
ing with the formation of the fungal cell wall and clotrimazole
kills fungi and yeasts by interfering with their cell membranes
and causing essential constituents of the fungal cells leakage.
MuellereHinton media, nutrient broth and malt extract broth
are purchased from Difco and yeast extracts are obtained
from Oxoid.
5. Results and discussions
5.1. Chemistry
2. Chemistry
5.1.1. Nuclear magnetic resonance
Ligand (L) is prepared by literature procedures [5,17]. Syn-
theses of the complexes are carried under reflux in EtOH. Full
1H and 13C spectral data of the free ligand and [M(L)Cl2]
(M ¼ Zn, Cd, Hg) are presented in Table 2. The H NMR
1
Fig. 1. Structure of the ligand (L).

N.M. Aghatabay et al. / European Journal of Medicinal Chemistry 42 (2007) 205e213
207
complexes. These strong downfield resonances on these car-
bon atoms are most probably caused by de-shielding effect
of the nitrogen atoms, which tend to support coordination
via nitrogen atoms (Table 2).
5.1.2. Infrared and Raman spectra
The prominent vibrational spectral data for free ligand and
the [M(L)Cl2] complexes together with their assignments are
presented in Table 3 and the spectra are shown in Figs. 3e7.
The NeH stretching and bending vibrations of the ligand at
3193 and 1460 cmꢀ1, remain either unperturbed or undergo
a slight shift in the complexes, suggesting that the imine pro-
ton remains attached at the N-1 position. The broadness of the
stretching vibration is very likely the result of the intramolec-
ular hydrogen bonding. These modes didn’t appear in the
Raman spectra, due to their weak intensity. The pure charac-
teristic n(CeH) modes of the ring residues are observed as ex-
pected in the wave regions 3100 cmꢀ1 both in IR and Raman
spectra. Slight but specific differences between the spectra of
the free ligand and the coordination compounds in this region
point to the formation of new complexes. The presence of
a sharp medium band at 1602 cmꢀ1 in IR spectrum and
weak band at 1625 cmꢀ1 in Raman spectrum, owing to
n(CeC) (ring) vibration in the free ligand shifts to higher fre-
quency at ca. 1628 cmꢀ1. The appearance of sharp strong
bands at w1600 cmꢀ1 in both IR and Raman spectra, owing
to the nas(C]N) (imine moiety) vibrations in the complexes,
indicate the bonding of imine nitrogen atoms with the metal
ions. Such vibration in the free ligand appears as a weaker
band at 1589 cmꢀ1. The comparison of bands for nas(CeN)
(pyridine moiety) vibrations of the ligand and complexes is
somewhat difficult owing to appearance of this band along
with bands of the nitrogen of the benzimidazole residue. How-
ever, presence of strong bands at w1580 cmꢀ1 in Raman and
at w1590 cmꢀ1 in IR spectra in the complexes could represent
the respective nas(CeN) stretches.
Fig. 2. Structure of the [Zn(L)Cl2] complex.
spectrum of the ligand exhibits a singlet at 13.1 ppm owing to
the imine proton atom NeH. The resonances corresponding to
the two equivalent benzimidazole groups are located at 7.33t,
7.40t, 7.79d and 7.84d ppm. In addition the pyridine moiety
displays two signals, a triplet at 8.22 and a doublet at
8.38 ppm due to the (1H-40, t) and (2H-30, 50, d) protons. Apart
from imine proton atoms, all the other proton signals are also
observed at nearly identical positions in the complexes baring
different coupling patterns compared with the free ligand.
However, the signal due to NeH proton in the mercury com-
plex is observed as a broad band in the 13e10 ppm region,
while the signals for the other complexes are not observed.
These protons either resonate along with aromatic protons or
may appear beyond 15 ppm [20].
The 13C NMR spectrum of the benzimidazole moiety in the
ligand, exhibits six signals for the benzene ring carbon atoms.
The nitrogen atom of pyridine unit is capable to form intramo-
lecular hydrogen bonding with NeH group of benzimidazole
moiety resulting the inhibition of fluxional behaviour of
N1eHeN3 unit (Fig. 1), creating an asymmetric o-substituted
molecule and therefore as expected the 13C spectrum of the
uncoordinated ligand exhibits 10 signals in the range
112.12e150.82 ppm (Table 2). The assignments of the signals
are made with the aid of theoretical calculation of the off-res-
onance 13C spectrum of the ligand and as well as in compar-
ison with the reported values [21].
On complexation, very strong band appears at w1450 cmꢀ1
both in Raman and IR and are interpreted as ns(C]N) frequen-
cies of the coordinated ligand, through its imine nitrogen atoms.
However, this mode is not characteristic and is strongly coupled
with several other vibrations, including n(CeH), n(NeH) and
n(CeC) (Table 3, Figs. 3 and 4).
Of particular interest is the lower frequency region, charac-
teristic for the metalechloride and metalenitrogen stretching
vibrations. The Raman spectra of the anhydrous solid metal
The spectral pattern will change in the complex due to the
coordination with lowering of the electron density in the li-
gand system, resulting possible downfield chemical shifts of
the carbon signals. Significant chemical shifts are observed
for C-2, 150.82 / 165.94; C-9, 144.50 / 148.15; C-8,
134.07 / 147.36 in the [Hg(L)Cl2] complex. Similar shifts
are also observed for the [Cd(L)Cl2] and [Zn(L)Cl2]
Table 1
The analytical data and physical properties of the ligand and complexes
Compound
Found (calculated)
C
Colour
Yield %
M.p./dec
Lm
H
N
C
19H13N5 (L)
e
e
e
White
P-y
75
68
63
65
70
195
>300
>300
>300
>320
e
ZnC19H13N5Cl2 [A]
CdC19H13N5Cl2 [B]
HgC19H13N5Cl2 [C]
[Fe(L)2](ClO4)2 [D]
51.0 (49.7)
46.1 (45.3)
39.1 (38.7)
e
2.9 (3.1)
2.6 (2.8)
2.2 (2.4)
e
15.7 (15.3)
14.2 (13.9)
12.0 (11.7)
e
6.1
5.7
7.8
e
White
White
R-b
Lm, Molar conductivity Uꢀ1 cm2 molꢀ1; r-b, reddish-brown; p-y, pale-yellow; dec, decompose.

208
N.M. Aghatabay et al. / European Journal of Medicinal Chemistry 42 (2007) 205e213
Table 2
1H chemical shifts (ppm) for the ligand (L) and complexes
Compound
4 (2H)
5 (2H)
6 (2H)
7 (2H)
NH (2H)
13.1s
30 (1H)
40 (1H)
50 (1H)
L
7.79, 7.84m
7.4m, 7.33m
7.44m, 7.32m
7.48m, 7.35m
7.45m, 7.23m
7.4, 7.33m
7.79, 7.74m
7.84m, 7.72m
7.88m, 7.76m
7.8m, 7.72m
8.38d
8.22t
8.38d
[A]
[B]
[C]
7.84m, 7.72m
7.88m, 7.76m
7.82m, 7.72m
7.44m, 7.32m
7.48m, 7.35m
7.45, 7.23m
8.6e8.2m
8.6e8.2m
8.5e8.25m
8.6e8.2m
8.6e8.2m
8.5e8.25m
8.6e8.2m
8.6e8.2m
8.5e8.25m
13e10
Observed and theoretical 13C chemical shifts (ppm) for the ligand (L) and complexes
Comp.
2
4
5
6
7
8
9
20
30
40
50
60
(L)a
(L)b
[A]
[B]
[C]
a
150.82
150.13
165.51
165.54
165.94
120.06
116.65
122.95
123.19
124.52
124.08
123.48
125.91
125.69
126.44
122.55
123.25
124.99
124.5
112.12
116.11
116.6
134.07
138.89
147.9
144.5
140.08
148.2
148.08
150.88
149.46
149.6
121.71
117.68
124.57
124.27
124.88
139.55
138.84
138.42
141.11
140.92
121.71
117.68
124.57
124.27
124.88
148.08
150.88
149.46
149.6
115.89
123.3
148.1
147.36
149.45
148.15
125.77
149.98
149.98
Experimental data.
Theoretical data.
b
halides MeCl (Hg, Cd, Zn) are illustrated in Fig 5. The most
intense bands appearing at 314 (Hg), 229 (Zn) and 221 cmꢀ1
(Cd) represent the n(metaleCl) vibration frequencies. How-
ever, the position of the four-fold coordinating chloride atoms
in these salts is only bridging while in the title compounds
both (metaleCl) bonds are terminal. Therefore, the values of
the (metaleCl) stretches taken from the tetrahedral [M(pyridi-
ne)2Cl2] complex series are more helpful for comparison.
These vibrations have been located in 350e250 cmꢀ1 region
[22,23]. The two metalechloride bonds in the [M(L)Cl2] com-
plexes give rise to a pair of bands in both IR and Raman spec-
tra which are interpreted as nas(MeCl) and ns(MeCl)
vibrations (Table 3, Figs. 6 and 7).
in 250e180 cmꢀ1 region. However, in the title compounds the
nitrogen ligands are not ‘‘free’’ but integrated in a chelate ring
(Fig. 2). The bands at 203, 179 in the IR and 205, 192 cmꢀ1 in
the Raman spectra for the [Hg(L)Cl2] complex may be assign-
able to these two vibration modes, respectively. Similar values
are expected also for the [Cd(L)Cl2] and [Zn(L)Cl2]
complexes, since the effect of the metal ions will be small
on metalenitrogen stretching vibrations (Table 3).
5.1.3. Electronic spectra
The [Zn(L)Cl2] and [Hg(L)Cl2] complexes show absor-
bance at 315 and 312 nm, respectively, which may be assign-
able to n / p* and p / p* ligand field transitions.
Nitrogen / metal charge transfer for zinc complex is
observed as shoulders at 346 and 366 nm, whereas for the
mercury complex the same transitions occur at 324 and 350 nm
Based on the vibrational spectroscopic data for the different
Znepyridine [23e25] both the antisymmetric and symmetric
vibrations of metalenitrogen stretches are expected to appear
Table 3
Prominent IR (mid and far) and Raman bands for the ligand Bzp and complexes
Compound
(L)
FT-IR (cmꢀ1
)
Raman (cmꢀ1
)
3193vs n(NeH), 3063s n(CeH), 1602m n(CeC), 1589m nas(CeN),
1574s, 1460vs [ns(CeN), d(NeH), n(CeH)], 1435vs, 1317vs,
1278vs [n(CeN), n(CeC)], 1231m, 820s, 742vs, 694m, 451m, 419s,
402s, 350s, 322m, 287s, 277sh, 217m, 174m.
3075m n(CeH), 1625w n(CeC), 1596s n(CeN), 1575vs, 1538s,
1455s [ns(CeN), d(NeH), n(CeH)], 1282s, 1267s [ns(CeN),
n(CeC)], 1008w, 996s, 870w, 804w, 651w, 626w, 291w, 196m.
[A]
[B]
[C]
3442m n(HeOeH), 3180s n(NeH), 3100 n(CeH), 1629m n(CeC),
1608 n(CeN), 1589s, 1495s, 1482s, 1460vs [ns(CeN), d(NeH),
n(CeH)], 1427s, 1317vs, 1147m, 820s, 744vs, 666m, 583m, 488s,
432s, 397m, 332s, 289s, 269mb, 221m.
3073m n(CeH), 1630s n(CeC), 1611sn n(CeN), 1579s, 1554s,
1539vs, 1447vs [ns(CeN), d(NeH), n(CeH)], 1267vs [n(CeN),
n(CeC)], 1154m, 1017s, 973m, 808w, 585m, 533s, 333s,
316m n(ZneCl), 268s n(ZneCl), 207s n(ZneN), 195s n(ZneN).
3068m n(CeH), 1629m, n(CeC), 1601s n(CeN), 1582s, 1547vs,
1533vs, 1447vs [ns(CeN), d(NeH), n(CeH)], 1268vs [n(CeN),
n(CeC)], 1156s, 1013s, 974m, 628w, 582m, 531m, 330s, 315m
n(CdeCl), 262s n(CdeCl), 205s n(CdeN), 193s n(CdeN).
3069m n(CeH), 1627m n(CeC), 1600s n(CeN), 1581s, 1544vs,
1529vs, 1447vs [ns(CeN), d(NeH), n(CeH)], 1294m, 1267vs
[n(CeN), n(CeC)], 1154m, 1008vs, 972w, 807w, 740w d(CeH),
626w, 580m, 530m, 323s n(HgeCl), 266s n(HgeCl), 205s
n(HgeN), 192s n(HgeN).
3448m n(OeH), 3177sh n(NeH), 3104vs n(CeH), 1628m n(CeC),
1605s n(CeN), 1593s, 1851s, 1495s, 1475s, 1456e1447vs [ns(CeN),
d(NeH), n(CeH)], 1429vs, 1315vs, 1302s, 820s, 756vs d(CeH),
745vs d(CeH).
3450m n(OeH), 3175sh n(NeH), 3108sb n(CeH), 1623sh n(CeC),
1601s n(CeN), 1593s, 1581s, 1495m, 1477s, 1454vs [ns(CeN),
d(NeH), n(CeH)], 1443vs, 1429vs, 1315vs, 1300vs [n(CeN),
n(CeC)], 1233m, 993m, 820s, 743vs d(CeH), 678m, 663m 423s,
352m, 337m n(HgeCl), 276m n(HgeCl), 203s n(HgeN),
179s n(HgeN).
[D]
3415vs, 3237m, 3075m, 1642vs, 1618vs, 1608vs, 1477s, 1412s,
1365s, 1305s, 1147vs, 1118vs, 1109vs, 1083vs, 813s, 763s, 742,
629s, 432m.
3074m n(CeH), 1667m, 1605vs n(CeC), 1577vs n(CeN), 549,
430vs [ns(CeN), d(NeH), n(CeH)], 1304m, 1272s [n(CeN),
n(CeC)], 1148m, 1027s, 1010s, 582, 526m, 464s, 364m, 326m,
255m n(FeeN), 204m n(FeeN).
n, Stretching; d, bending; u, wagging; m, medium; s, strong; w, weak; sh, shoulder; b, broad.

N.M. Aghatabay et al. / European Journal of Medicinal Chemistry 42 (2007) 205e213
209
(a)
(a)
(b)
(b)
(c)
(d)
(c)
(d)
600
3000
2500
2000
1500
1000
3500
3000
2500
2000
1500
1000
500
Raman shift (cm–1
)
Wavenumber (cm–1
)
Fig. 4. Raman spectrum of (a) ligand, (b) [Hg(L)Cl2], (c) [Cd(L)Cl2], (d)
[Zn(L)Cl2] in the 3200e600 cmꢀ1 region.
Fig. 3. FT-IR spectrum of (a) ligand, (b) [Hg(L)Cl2], (c) [Cd(L)Cl2], (d)
[Zn(L)Cl2] in the 3500e400 cmꢀ1 region.
(Figs. 8 and 9). Complexation of metal ions with DNA bases is
expected to bring about marked changes in its electronic spec-
trum. However, no marked changes have been observed in the
electronic spectrum of these two complexes with the addition
of DNA, ruling out the possibility of their direct coordinative
binding to DNA donor atoms. With the addition of DNA, these
two complex systems exhibit only small changes in their
electronic spectrum. Such a small change may arise with
groove binding, leading to small perturbations. This hyper-
chromism in the absorption intensity may probably be due
to the dissociation of ligand aggregates [26] or due to its exter-
nal contact (surface binding) with the duplex [27].
(a)
(b)
5.2. Microbial activity
The results concerning in vitro antimicrobial activities of
the ligand, their hydrochloride salts, the metal salts and the
complexes together with the inhibition zone (mm) and
(MIC) values of compared antibiotic and antifungal activities
are presented in Tables 4e6. All the compounds tested exhibit
strong or moderate antimicrobial activity. Of all the test com-
pounds attempted, [Zn(L)Cl2] and [Hg(L)Cl2] complexes
showed the highest activities against most Gram positive and
Gram negative bacteria and as well as yeast cultures. As an ex-
ample, 25 (Hg), 21 (Zn), 29 (Hg), 26 (Zn) and 30 (Hg), 26 mm
(Zn) inhibition zone values of the these complexes on
(c)
400
350
300
250
200
)
150
100
Raman shift (cm–1
Fig. 5. Raman spectrum of (a) HgCl2, (b) CdCl2, (c) ZnCl2 in the 400e
100 cmꢀ1 region.

210
N.M. Aghatabay et al. / European Journal of Medicinal Chemistry 42 (2007) 205e213
(a)
(a)
(b)
(b)
(c)
(c)
(d)
500
450
400
350
300
250
200
Wavenumber (cm–1
)
600
500
400
300
200
Fig. 6. The 500e175 cmꢀ1 region FT-IR spectrum of (a) (L), (b) [Hg(L)Cl2],
(c) [Zn(L)Cl2].
Raman shift (cm–1
)
Fig. 7. Raman spectrum of (a) (L), (b) [Hg(L)Cl2]2, (c) [Cd(L)Cl2]2, (d)
[Zn(L)Cl2] in the 600e100 cmꢀ1 region.
Escherichia coli (Gram negative), Staphylococcus aureus
(Gram positive) and Proteus vulgaris (Gram negative) organ-
isms are exceptionally effective compared with most of the
reference antibiotics (Table 4). Similarly these two complexes
show strong antifungal activity with the inhibition zone values
of 27 (Hg), 24 (Zn) and 22 (Hg), 20 (Zn), 21 (Hg), 15 mm (Zn)
inhibition on Kluyveromyces fragilis, Candida albicans and
Rhodotorula rubra organisms, respectively (Table 5). These
values are consistent and compatible with all the antifungal ac-
tivities in comparison tests. The MIC values in Table 6 indi-
cate that all the compounds tested exhibit moderate to strong
antimicrobial activity on the tested microorganisms. Once
again the microbial data indicate that the mercury and zinc
complexes have a very strong and penetrating activity against
most Gram positive and Gram negative bacteria and as well as
yeast cultures. For instance, these two complexes showed su-
perior activity against S. aureus (Gram positive) organism
experimental conditions on Mycobacterium smegmatis and
Listeria monocytogenes microorganisms.
The inhibition activity seems to be governed in certain degree
by the facility of coordination at the metal centre, and therefore
the complexes show stronger activity against the tested
(MIC ¼ 0.78 mg mLꢀ1
)
than the reference gentamycin
(MIC ¼ 25 mg mLꢀ1) or against P. vulgaris (Gram negative)
organism (MIC ¼ 0.78 (Hg), 1.56 mg mLꢀ1 (Zn)) than the ref-
erence gentamycin (MIC ¼ 6.25 mg mLꢀ1). Similarly the mer-
cury and zinc complexes show improved antifungal activity
compared to the other test compounds on most of the organ-
isms, particularly against K. fragilis (0.78 (Hg), 1.56 mg mLꢀ1
(Zn)) and C. albicans (3.125 (Hg), 3.125 mg mLꢀ1 (Zn)) than
the reference nystatin (6.25 mg mLꢀ1) and (3.125 mg mLꢀ1),
respectively (Table 6). Surprisingly all of the test compounds
attempted have either very weak or no effect under the given
Fig. 8. UVevis spectra of CT-DNA-[Zn(L)Cl2] complex.

N.M. Aghatabay et al. / European Journal of Medicinal Chemistry 42 (2007) 205e213
211
products of analytical grade without any purification. Molar
conductivity of the complexes was measured on a WPA
CMD750 conductivity meter in dimethyl sulfoxide (DMSO)
solution at 25 ꢁC. UVevis spectral measurements of the
complexes, and DNA binding studies were carried out on
Unicam Alpha Helios spectrophotometer at 25 ꢁC. FT-IR
spectra were recorded in KBr pellets (mid, 4000e
400 cmꢀ1) and in polyethylene discs (far, 600e170 cmꢀ1
)
on a Jasco FT/IR-600 Plus Spectrometer. Raman spectra
were obtained from powdered samples placed in a Pyrex
tube using a Bruker RFS 100/S Raman spectrometer in the
range 4000e0 cmꢀ1. The 1064 nm line, provided by a near
infrared Nd:YAG air-cooled laser was used as excitation
line. The output laser power was set to 100e120 mW. Rou-
1
tine H and 13C NMR spectra were recorded at ambient tem-
perature on a 500 MHz NMR Spectrometer in DMSO-d6.
Chemical shifts (d) are expressed in units of part per million
relative to tetramethylsilane.
Fig. 9. UVevis spectra of CT-DNA-[Hg(L)Cl2] complex.
microorganisms, compared to the free ligands. This supports the
argument that some type of bimolecular binding most probably
occurs to the metal ions causing the inhibition of biological syn-
thesis and preventing the organisms from reproducing.
6.1.1. Synthesis
6.1.1.1. 2,6-Bis(benzimidazol-2-yl)pyridine ligand (L). The
ligand (L) was prepared according to the literature procedures
[5,17], by reacting a mixture of pyridine-2,6-dicarboxylic acid
(3.34 g, 20 mmol) and freshly sublimed o-phenylenediamine
(4.32 g, 40 mmol) in 5 M HCl (40 mL) at reflux temperature.
The precipitates were collected and neutralized by aqueous
ammonia.
In classifying the antibacterial activity as Gram positive or
Gram negative, it would generally be expected that a much
greater number would be active against Gram positive than
Gram negative bacteria. However, in this study, the compounds
are active against both types of the bacteria and as well as active
against yeasts, which may indicate broad-spectrum properties.
The results of our study indicate that the complexes as
a whole, but particularly essential zinc complex, have the po-
tential to generate novel metabolites, by displaying moderate
to high affinities for most of the receptors. This compound
could be selected for further pharmacological tests to be eval-
uated as potential drug against many infectious diseases.
6.1.1.2. [Zn(L)Cl2] complex [A]. Hot solution mixture of (L)
(156 mg, 50 mmol) and ZnCl2 (75 mg, 55 mmol) in EtOH
(10 mL) was reacted for several hours. The pale-yellow solid
products were collected, washed with small portions of
EtOH and dried in vacuum.
6.1.1.3. [Cd(L)Cl2] complex [B] [15]. Complex [B] was
synthesized in a similar manner to that used for complex
[A], reacting a mixture of (L) (156 mg, 50 mmol) and CdCl2
(100 mg, 55 mmol) in EtOH (10 mL).
6. Experimental
6.1. Chemistry
Apart from o-phenylenediamine (freshly sublimed), all
other chemicals and solvents were used as commercial
6.1.1.4. [Hg(L)Cl2] complex [C]. Complex [C] was synthe-
sized in a similar manner to that used for complex [A],
Table 4
In vitro antibacterial activities of compounds and standard reagents (inhibition zone, mm)
Microorganisms
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
PEN
SAM
CTX
VAN
OFL
TET
Escherichia coli
14
15
8
16
13
10
10
14
14
9
13
16
9
12
11
8
14
15
12
10
13
14
10
11
11
14
18
12
12
15
17
11
12
12
16
21
12
11
16
21
11
10
11
18
24
14
14
18
24
10
12
13
21
26
14
15
16
26
12
12
14
25
29
15
15
20
30
12
12
16
18
13
18
14
36
10
15
10
8
12
16
14
12
32
16
21
12
10
10
12
13
14
32
18
11
16
54
22
13
22
18
34
20
20
26
10
30
24
28
30
28
28
32
30
44
28
26
30
25
22
26
24
28
34
Staphylococcus aureus
Klebsiella pneumoniae
Bacillus cereus
11
12
11
8
9
12
14
12
8
Micrococcus luteus
Proteus vulgaris
12
10
8
Mycobacterium smegmatis
Listeria monocytogenes
Pseudomonas aeruginosa
10
10
9
12
8
11
10
12
PEN, penicillin G; SAM, ampicillin; CTX, cefotaxime; VAN, vancomycin; OFL, oflaxacin; TET, tetracyclin; (1), (ZnCl2); (2), (HgCl2); (3), (CdCl2); (4),
[Fe(ClO4)2]; (5), (L); (6), [(L)(3HCl)]; (7), [Fe(ClO4)2(L)]; (8), [CdCl2(L)]; (9), [ZnCl2(L)]; (10), [HgCl2(L)].

212
N.M. Aghatabay et al. / European Journal of Medicinal Chemistry 42 (2007) 205e213
Table 5
In vitro antifungal activities of compounds and standard reagents (inhibition
zone, mm)
(DMSO) according to the required concentrations (compounds
are not soluble in water). To ensure that the solvent had no ef-
fect on bacterial growth, a control test was performed with test
medium supplemented with DMSO as the same procedures as
used in the experiments. The discs were impregnated with
20 mL of these solutions. All the bacteria were incubated
and activated at 30 ꢁC for 24 h inoculation into nutrient broth,
and the yeasts were incubated in malt extract broth for 48 h.
Inoculums containing 106 bacterial cells or 108 yeast cells
per mL were spread on MuellereHinton Agar plates (1 mL in-
oculum for each plate). The discs injected with solutions were
placed on the inoculated agar by pressing slightly and incu-
bated at 35 ꢁC (24 h) and at 25 ꢁC (72 h) for bacteria and
yeast, respectively. On each plate an appropriate reference an-
tibiotic disc was applied depending on the test microorgan-
isms. In each case triplicate tests were performed and the
average was taken as the final reading.
Microorganisms
(1) (2) (3) (4) (5) (6) NY KET CLT
Candida albicans 14 15 15 17 20 22 20
Kluyveromyces fragilis 15 18 21 20 24 27 18
Rhodotorula rubra 12 14 14 18 15 21 18
H. guilliermondii 12 14 12 14 15 18 21
Debaryomyces hansenii 11 13 12 14 15 17 16
21
16
22
24
14
15
18
16
22
18
NY, nystatin; KET, ketaconazole; CLT, clotrimazole; (1), (L); (2), [(L)
(HCl)3]; (3), [Fe(ClO4)2(L)]; (4), [CdCl2(L)]; (5), [ZnCl2(L)]; (6), [HgCl2(L)].
reacting a mixture of (L) (125 mg, 40 mmol) and HgCl2
(120 mg, 44 mmol) in EtOH (10 mL).
6.1.1.5. [Fe(L)2](ClO4)2 complex [D] [16]. Complex [D] was
prepared in a similar manner to that used for complex [A], re-
acting a mixture of (L) (125 mg, 40 mmol) and Fe(ClO4)2
(112 mg, 44 mmol) in EtOH (8 mL).
6.2.2.2. Dilution method. Screening was performed following
the procedure outlined in the Manual of Clinical Microbiology
[6,30]. All the bacteria were incubated and activated at 30 ꢁC
for 24 h inoculation into nutrient broth and the yeasts were in-
cubated in malt extract broth for 48 h. The compounds were
dissolved in DMSO and then diluted using cautiously adjusted
MuellereHinton broth. Two-fold serial concentrations of the
compounds were employed to determine the (MIC) ranging
from 100 mg mLꢀ1 to 0.78 mg mLꢀ1. In each case triplicate
tests were performed and the average was taken as the final
reading.
Cultures are grown at 37 ꢁC (20 h) and the final inoculation
(inoculums) was approximately 106 cfu mLꢀ1. Test cultures
are incubated at 37 ꢁC (24 h). The lowest concentrations of an-
timicrobial agents that result in complete inhibition of micro-
organisms are represented as (MIC) mg mLꢀ1. In each case
triplicate tests were performed and the average was taken as
the final reading.
6.2. Pharmacology
6.2.1. Microorganisms
The antimicrobial activities are evaluated against Gram
positive (S. aureus, Bacillus cereus, M. smegmatis, L. monocy-
togenes, Micrococcus luteus) and Gram negative (E. coli,
Klebsiella pneumoniae, Pseudomonas aeruginosa, P. vulgaris,
Enterobacter aerogenes) bacteria and the yeast cultures (C.
albicans, K. fragilis, R. rubra, D. hansenii and H. guilliermondii)
using disk diffusion method as well as the minimal inhibitory
concentration (MIC) dilution method which are outlined in
Section 6.
6.2.2. Antimicrobial screening
6.2.2.1. Disk diffusion method. Sterilized antibiotic discs
(6 mm) were used following the literature procedure [28,29].
Fresh stock solutions (30 mg mLꢀ1) of the ligand and the com-
plexes were prepared in redistilled dimethyl sulfoxide
6.2.2.3. DNA binding. Calf thymus DNA was purchased from
Worthington Biochemical Corp. and dissolved in 10 mM
Table 6
In vitro antimicrobial activity of compounds and standard reagents (MIC, mg mLꢀ1
)
Microorganisms
(1)
12.5
6.25
(2)
(3)
12.5
6.25
(4)
(5)
(6)
12.5
3.125
(7)
6.25
1.56
50
50
6.25
3.125
50
(8)
3.125
1.56
25
25
6.25
1.56
100
(9)
3.125
0.78
12.5
12.5
6.25
1.56
50
(10)
1.56
0.78
12.5
12.5
3.125
0.78
50
50
6.25
GN
6.25
25
6.25
6.25
25
6.25
12.5
12.5
6.25
NY
Escherichia coli
6.25
25
25
12.5
e
e
e
e
e
e
e
e
e
Staphylococcus aureus
Klebsiella pneumoniae
Bacillus cereus
12.5
50
12.5
50
100
25
50
50
50
50
50
50
25
100
Micrococcus luteus
Proteus vulgaris
12.5
12.5
12.5
50
12.5
25
12.5
12.5
50
12.5
12.5
12.5
25
100
25
6.25
100
50
Mycobacterium smegmatis
Listeria monocytogenes
Pseudomonas aeruginosa
Kluyveromyces fragilis
Rhodotorula rubra
Candida albicans
100
50
100
50
25
100
100
100
100
50
50
50
12.5
25
12.5
14.5
12.5
25
25
12.5
12.5
12.5
12.5
25
100
12.5
50
12.5
50
3.125
12.5
3.125
12.5
3.125
3.125
6.25
1.56
12.5
3.125
0.78
e
e
e
e
e
6.25
6.25
50
3.125
3.125
6.25
12.5
12.5
50
12.5
50
12.5
25
12.5
50
12.5
50
12.5
50
3.125
3.125
H. guilliermondii
Debaryomyces hansenii
12.5
12.5
12.5
12.5
25
25
25
50
50
50
6.25
12.5
GN, gentamycin; NY, nystatin; (1), (ZnCl2); (2), (HgCl2); (3), (CdCl2); (4), [(Fe(ClO4)2]; (5), (L); (6), [(L)(3HCl)]; (7), [Fe(ClO4)2(L)]; (8), [CdCl2(L)]; (9),
[ZnCl2(L)]; (10), [HgCl2(L)].

N.M. Aghatabay et al. / European Journal of Medicinal Chemistry 42 (2007) 205e213
213
[12] V. Ganesan Vaidyanathan, B. Unni Nair, J. Inorg. Biochem. 91 (2002)
405.
TriseHCl buffer at pH 8.0. CT-DNA in the buffer gave a ratio
of UV absorbance at 260 and 280 nm of 1.9:1, indicating that
DNA was free from protein [31]. The DNA concentration per
base pair was determined by absorption spectroscopy using the
molar absorption coefficient (12824 Mꢀ1 cmꢀ1) [32]. The ab-
sorption spectra of the complex [10 mM in buffereDMSO (1:1
v/v)] was obtained in the absence and presence of increasing
amounts of CT-DNA ([DNA] ¼ 0e10 mM]).
[13] S.J. Lippard, J.M. Berg, Principles of Bioinorganic Chemistry, University
Science Books, Mill Valley, CA, 1994, pp. 103e137.
[14] R. Boca, P. Baran, L’ubor Dlhan, H. Fuess, W. Haase, F. Renz, W. Linert,
I. Svoboda, R. Werner, Inorg. Chim. Acta 260 (1997) 129.
[15] Sheng-Gui Liu, Jing-Lin Zuo, Yi-Zhi Li, Xiao-Zeng You, J. Mol. Struct.
705 (2004) 153.
[16] M. Hasegawaa, F. Renzb, T. Haraa, Y. Kikuchia, Y. Fukudab, J. Okuboc,
T. Hoshia, W. Linertd, Chem. Phys. 277 (2002) 21.
[17] A.W. Addison, P.J. Burke, Heterocycl. Chem. 18 (1981) 803.
[18] A.D. Becke, Phys. Rev. A 38 (1988) 3098.
[19] J. Perdew, Phys. Rev. B 33 (1986) 8822.
References
J. Perdew, Phys. Rev. B 34 (1986) 7046.
[20] R.N. Mohanty, V. Chakravorty, K.C. Dash, Polyhedron 10 (1991) 33.
[21] A. Gelling, K.G. Orrell, A.G. Osborne, V. Sik, M.B. Hursthouse,
D.E. Hibbs, K.M.A. Malik, Polyhedron 13e14 (1998) 2141.
[22] K. Nakamoto, Infrared, Raman Spectra of Inorganic Coordination Com-
pounds, Part B, fifth ed. 1997, pp. 23e33.
[1] R. Canetta, M. Rozencweig, R.E. Wittes, L.P. Schacter, Cancer Chemo-
therapy: Challenges for the Future, vol. 5, Excerpta Medica, Tokyo,
1990, p. 318.
[2] B. Rosenberg, B. Lippert, Cisplatin, Chemistry and Biochemistry of
a Leading Anticancer Drug, Verlag Chemie VCH, Basel, 1999, p. 3.
[3] I.H. Krakoff, M. Nicali (Eds.), Platinum and Other Metal Coordination
Compounds in Cancer Chemotherapy: Clinical Application of Platinum
Complexes, Martinus Nijhoff, Boston, 1988, p. 351.
[4] K.K. Monthilal, C. Karunakaran, A. Rajendran, R. Murugesan, J. Inorg.
Biochem. 98 (2004) 322.
[23] D.A. Thornton, Coord. Chem. Rev. 104 (1990) 251.
[24] D. Demertzi, D. Nicholls, Inorg. Chim. Acta (1983) 73.
[25] S. Akyuz, J.E.D. Davis, A.B. Dempster, K.T. Holmes, J. Chem. Soc. Dal-
¨
ton. Trans. (1976) 1746.
[26] Rajamanickam Vijayalakshmi, Mookandi Kanthimathi, Venkatesan
Subramanian, Balachandran Unni Nai, Biochim. Biophys. Acta
1475 (2000) 157.
[5] M. NazAgh-Atabay, B. Dulger, F. Gucin, Eur. J. Med. Chem. 38 (2003)
875.
[27] R. Tamilarasan, D.R. McMillin, Inorg. Chem. 29 (1990) 2798.
[28] Performance Standards for Antimicrobial Disk Suspectibility Tests,
Approved Standard NCCLS Publication M2-A5, Villanova, PA, USA,
1993, pp. 1e32.
[6] M. NazAgh-Atabay, B. Dulger, F. Gucin, Eur. J. Med. Chem. 40 (2005)
1096.
[7] F.J. Schendel, Y.S. Cheng, J.D. Otvos, S. Vehrli, J. Stubbe, Biochemistry
27 (1988) 2614.
[29] C.H. Collins, P.M. Lyre, J.M. Grange, Microbiological Methods, sixth ed.
Butterworths Co. Ltd., London, 1989.
[8] R.V. Devivar, E. Kawashima, G.R. Revankar, J.M. Breitenbach,
E.D. Kreske, J.C. Drach, L.B. Townsend, J. Med. Chem. 37 (1994)
2942.
[30] R.N. Jones, A.L. Barry, T.L. Gaven, J.A. Washington, in: E.H. Lennette,
A. Balows, W.J. Shadomy (Eds.), Manual of Clinical Microbiology,
fourth ed. American Society for Microbiology, Washington DC, 1984,
pp. 972e977.
[9] L.B. Townsend, R.V. Devivar, S.R. Turk, M.R. Nassiri, J.C. Drach, J.
Med. Chem. 38 (1995) 4098.
[10] H. Irving, P.J.R. Williams, J. Chem. Soc. (1953) 3192.
[11] (a) R.G. Pearson, J. Am. Chem. Soc. 85 (1963) 3353;
(b) R.G. Parr, R.G. Pearson, J. Am. Chem. Soc. 105 (1983) 7512.
[31] J. Marmur, J. Mol. Biol. 3 (1961) 208.
[32] D. Suh, J.B. Chanes, Bioorg. Med. Chem. 3 (1995) 723.