Anti-inflammatory, antiproliferative, and radical-scavenging activities of tolfenamic acid and its metal complexes

Article abstract of DOI:10.1002/cbdv.200800120

Some new complexes of tolfenamic acid (=2-[(2-methyl-3-chlorophenyl)amino] benzoic acid; Htolf) with potentially interesting biological activities are described. The complexes [Mn(tolf)₂(H₂O)₂], [Co(tolf)₂(H₂O)₂], [Ni(tolf₂(H ₂O)₂], [Cu(tolf)₂(H₂O)]₂ , and [Zn(tolf)₂(H₂O)] were prepared by the reaction of tolfenamic acid, a potent anti-inflammatory drug, with metal salts. The radical-scavenging activities of the complexes were evaluated using the 1,1-diphenyl-2-picrylhydrazyl (DPPH) free radical-scavenging assay. Their ability to inhibit soybean lipoxygenase, β-glucuronidase, and trypsin-induced proteolysis was studied. Their inhibitory effects on rat paw edema induced by carrageenin was studied and compared with those of tolfenamic acid. The complex [Zn(tolf)₂(H₂O)] exhibited the strongest in vivo inhibitory effect at 0.1 mm/kg Body Weight (BW; 93.0±0.9%), superior than the inhibition induced by tolfenamic acid at the same molar dose (76.0±0.9%). Tolfenamic acid and its metal complexes have been evaluated for antiproliferative activity in vitro against the cells of three human cancer cell lines, MCF-7 (breast cancer cell line), T24 (bladder cancer cell line), and A-549 (non-small cell lung carcinoma), and a mouse fibroblast L-929 cell line. The complexes [Mn(tolf)₂(H₂O)₂] and [Cu-(tolf)₂(H₂O)]₂ have shown selectivity against T24 cell line. The IC₅₀ values of these two complexes against T24 cancer cell lines are in a micromolar range similar or better to that of the antitumor drug cisplatin.

Full text of DOI:10.1002/cbdv.200800120

948
CHEMISTRY & BIODIVERSITY – Vol. 6 (2009)
Anti-Inflammatory, Antiproliferative, and Radical-Scavenging Activities of
Tolfenamic Acid and Its Metal Complexes
by Dimitra Kovala-Demertzi*a), Dimitra Hadjipavlou-Litina*b), Alexandra Primikiria),
Malgorzata Staninskaa), Chronis Kotogloua), and Mavroudis A. Demertzisa)
a) Inorganic and Analytical Chemistry, Department of Chemistry, University of Ioannina,
GR-45100 Ioannina (phone: þ302651-98425; fax: þ302651-98786; e-mail: dkovala@cc.uoi.gr)
b) Department of Pharmaceutical Chemistry, School of Pharmacy, Aristotle University of Thessaloniki,
GR-54124 Thessaloniki (phone: þ302310997627; fax: þ302310997679;
e-mail: hadjipav@pharm.auth.gr)
Some new complexes of tolfenamic acid (¼2-[(2-methyl-3-chlorophenyl)amino]benzoic acid;
Htolf) with potentially interesting biological activities are described. The complexes [Mn(tolf)2(H2O)2],
[Co(tolf)2(H2O)2], [Ni(tolf2(H2O)2], [Cu(tolf)2(H2O)]2, and [Zn(tolf)2(H2O)] were prepared by the
reaction of tolfenamic acid, a potent anti-inflammatory drug, with metal salts. The radical-scavenging
activities of the complexes were evaluated using the 1,1-diphenyl-2-picrylhydrazyl (DPPH) free radical-
scavenging assay. Their ability to inhibit soybean lipoxygenase, b-glucuronidase, and trypsin-induced
proteolysis was studied. Their inhibitory effects on rat paw edema induced by carrageenin was studied
and compared with those of tolfenamic acid. The complex [Zn(tolf)2(H2O)] exhibited the strongest in
vivo inhibitory effect at 0.1 mm/kg Body Weight (BW; 93.0ꢀ0.9%), superior than the inhibition induced
by tolfenamic acid at the same molar dose (76.0ꢀ0.9%). Tolfenamic acid and its metal complexes have
been evaluated for antiproliferative activity in vitro against the cells of three human cancer cell lines,
MCF-7 (breast cancer cell line), T24 (bladder cancer cell line), and A-549 (non-small cell lung
carcinoma), and a mouse fibroblast L-929 cell line. The complexes [Mn(tolf)2(H2O)2] and [Cu-
(tolf)2(H2O)]2 have shown selectivity against T24 cell line. The IC50 values of these two complexes
against T24 cancer cell lines are in a micromolar range similar or better to that of the antitumor drug
cisplatin.
Introduction. – Tolfenamic acid (¼2-[(2-methyl-3-chlorophenyl)amino]benzoic
acid; Htolf) is a potent, well-tolerated non-steroidal anti-inflammatory drug (NSAID)
with a low gastroulcerogenicity [1], a low overall toxicity, and high therapeutic indices.
It possesses analgesic and antipyretic properties as demonstrated in several animal
models and has shown good results in long-term treatment of rheumatoid arthritis and
osteoarthritis [2]. Chemically, it resembles mefenamic and flufenamic acids, other
fenamates in clinical use. The anti-inflammatory activity of NSAIDs and most of its
other pharmacological effects are related to the inhibition of the conversion of
arachidonic acid (AA) to prostaglandins, which are mediators of the inflammatory
process. The enzymes cyclooxygenase (COX) and lipoxygenase (LOX), which catalyze
the oxidative metabolism of amino acids, are useful targets for the design and the
development of new drugs that substantially inhibit the generation of the final
inflammatory products and the propagation of inflammation. Recent studies revealed
ꢀ 2009 Verlag Helvetica Chimica Acta AG, Zꢁrich

CHEMISTRY & BIODIVERSITY – Vol. 6 (2009)
949
that, in addition to arthritis and pain, cancer and neurodegenerative diseases like
Alzheimerꢂs disease could potentially be treated with Cox-2 inhibitors [3][4].
Synthesis and study of metal complexes with active drugs as ligands, as an approach
to new drug development [5][6], is a research area of increasing interest in inorganic,
pharmaceutical, and medicinal chemistry. X-Ray crystal-structure investigations of
Htolf revealed the existence of two polymorphic forms containing the molecule as
different conformers [7] (Fig. 1). The crystal structure of dimeric tetraorganodistan-
noxane adducts of Htolf has been reported by our group [8]. We also recently reported
synthesis, crystal structure, IR- and UV/VIS-spectroscopic data, and electrochemical
studies of CuII complexes of Htolf (Fig. 2) [9]. The superoxide dismutase activity was
measured, and IC50 value was determined by the Fridovich test. According to the usual
criteria, the Cu complex of tolfenamic is considered as potent SOD mimic [9]. A review
of the synthesis and crystal structures of a number of anti-inflammatory compounds as
ligands in organotin complexes, and a review of Cu complexes with NSAIDs have
recently been published [10][11].
Fig. 1. The structure of tolfenamic acid [7]
We have synthesized and characterized novel complexes of MnII, CoII, NiII, CuII, and
ZnII with Htolf, i.e., [Mn(tolf)2(H2O)2], [Co(tolf)2(H2O)2], [Ni(tolf)2(H2O)2], [Cu-
(tolf)2(H2O)]2, and [Zn(tolf)2(H2O], and the complexes were tested for their
antioxidant and anti-inflammatory activities in the carrageenin-induced rat paw edema
assay. Also, their inhibition of LOX, of b-glucuronidase, and of trypsin-induced
proteolysis was evaluated. The cytotoxic activity of Htolf and its metal complexes have
been evaluated for antiproliferative activity in vitro against the cells of three human
cancer cell lines, MCF-7 (human breast cancer cell line), T24 (bladder cancer cell line),
and A-549 (non-small cell lung carcinoma), and a mouse cell line, L-929 (a fibroblast-
like cell line cloned from strain L).
Results and Discussion. – 1. Chemistry. Tolfenamic acid (Htolf) was synthesized
according to a published procedure, the Ullmann–Goldberg condensation [12]
(Scheme). The complexes [Mn(tolf)2(H2O)2], [Co(tolf)2(H2O)2], [Ni(tolf)2(H2O)2],
[Cu(tolf)2(H2O)]2, and [Zn(tolf)2(H2O)] were synthesized according to the Reactions
1–5, in MeOH and aqueous solutions.

950
CHEMISTRY & BIODIVERSITY – Vol. 6 (2009)
Scheme. Synthesis of Tolfenamic Acid and Its Metal Complexes
MnCl2 þ2 Htolf![Mn(tolf)2(H2O)2]þ2 HCl
Co(AcO)2 ·4 H2Oþ2 Htolf![Co(tolf)2(H2O)2]þ2 AcOHþ2 H2O
NiCl2 ·6 H2Oþ2 Htolf![Ni(tolf)2(H2O)2]þ2 HCIþ4 H2O
Cu(AcO)2 ·H2Oþ2 Htolf![Cu(tolf)2(H2O)]2 þ2 AcOH
ZnCl2 þ2 Htolf![Zn(tolf)2(H2O)]þ2 HCl
(1)
(2)
(3)
(4)
(5)
The complexes are microcrystalline or powder-like, and stable under atmospheric
conditions. All the complexes are insoluble in H2O and MeCN, and are soluble in non-
polar solvents as benzene, toluene, CHCl3, CH2Cl2, acetone, and THF, and polar
solvents as DMF and DMSO. The elemental analyses confirm their stoichiometry. The
meff values of the complexes show that these are all high-spin; the large and small orbital
contribution in the [Co(tolf)2(H2O)2] and [Ni(tolf)2(H2O)2], i.e., t2g5eg2 CoII and t2g6eg2
NiII, respectively, indicate six-coordinate structures.
2. Spectroscopy. 2.1. IR Spectroscopy. As the carboxy H-atom is more acidic than
the amino H-atom the deprotonation occurs in the COOH group. This is confirmed by
the IR spectra of the complexes, showing the characteristic bands for the secondary
amino groups and for the coordinated carboxylato group. A broad absorption at
3400 cmꢁ1 in the spectra of the complexes was attributed to the presence of
coordinated H2O. The absence of large systematic shifts of the ˜n(NH) and d(NH)
bands in the spectra of the complexes compared with those of the ligand indicates that
there is no interaction between the NH group and the metal ions. The n˜as(COO) and
n˜sym(COO) bands of the prepared complexes are at 1615–1580 and at 1500–1450 cmꢁ1
respectively; The difference D(n˜as(COO)ꢁn˜sym(COO)) for the Mn, Co, Ni, and Zn
complexes is less than the ionic value (for sodium tolfenamate, the D value is 160 cmꢁ1),
as expected for the bidentate chelating mode of carboxylate ligation [8][9][13][14].
The difference, D(n˜as(COO)ꢁn˜sym(COO)) for the Cu complex is close and less
(161 cmꢁ1) than for sodium tolfenamate, as expected for bridging bidentate carboxy-
lato groups, supporting a structure analogous to [Cu(tolf)2(DMF)]2 (Fig. 2) [9] and
[Cu(diclof)2(DMF)]2 [13]. The four carboxylate groups from four ligands are in a
bidentate bridging mode. The square-pyramidal geometry with a DMF O-atom or H2O
occupying both apical positions as was established by single-crystal X-ray study [9][12].
The medium bands at ca. 400 cmꢁ1 are attributed to the n˜(MꢁOH O ) stretching mode,
2
while the bands at 250–210 cmꢁ1 to the n˜(MꢁOoco) stretching mode [8][9][12][13].

CHEMISTRY & BIODIVERSITY – Vol. 6 (2009)
951
Fig. 2. The structure of the complex [Cu(tolf)2(DMF)]2. The H-atoms are omitted for clarity [9].
2.2. Electronic Spectroscopy. The electronic spectra were recorded in DMF and
CHCl3 solution, and absorption maxima in the UV/VIS region are listed in Table 1
along with suggested assignments [9][13–15]. The electronic spectra of the complex
[Mn(tolf)2(H2O)2] can be assigned to six-coordinate configuration [14][15]. The
absorption of the organic ligand tailing into the visible region obscures the very weak
d-d absorption bands of the MnII complexes. The d-d spectra of the solvated Mn, Co,
and Ni complexes can be assigned to transitions in pseudo-octahedral structures or six-
coordinated tetragonally distorted configurations. The band frequencies and the 10 Dq
value in [Ni(tolf)2(H2O)2] are characteristic of a NiIIO6 chromophore, which do not
cause a pronounced nephelauxetic effect [14][15]. The intense bands of complex of
CoII in the visible region, indicates a distorted octahedral configuration [14]. The
solution spectral data indicate that the Mn, Co, and Ni complexes are all solvated with
two solvent molecules coordinating in a pseudo-octahedral arrangement (Reaction 6).
[M(tolf)2(H2O)2]þ2 solvate![M(tolf)2(solvate)2]þ2 H2O
(6)
where M is MnII, NiII, or CoII, and solvate is DMF or CHCl3.
3. Biological Studies. The metal complexes and the parent drug were studied with
regard to their radical-scavenging ability as well as to the inhibition of a) soybean
lipoxygenase, b) b-glucuronidase, and c) trypsin-induced proteolysis.
The reducing abilities of the tolf complexes were evaluated using the 1,1-diphenyl-
1-picrylhydrazyl (DPPH) free radical-scavenging assay [15] in an iron-free system. This

952
CHEMISTRY & BIODIVERSITY – Vol. 6 (2009)
Table 1. Spectral Data (UV/VIS) and meff Values of the Prepared Complexes and the Parent Drug
meff
UV (CHCl3)
UV (DMF)
Assignments
l [nm]
log e
l [nm]
log e
Tolfenamic acid
352 (sh)
288
547 (sh)
351
288
580
531
351
290
770 (sh)
683
3.73
3.81
347
295
551 (sh)
337 (sh)
299
560
538
331 (sh)
300
750 (sh)
677
338 (sh)
297
723
330 (sh)
299
3.82
4.16
n!p*
p!p*
4
[Mn(tolf)2(H2O)2]
5.96
4.88
6A1g ! T1g (G)
4.30
4.42
1.97
1.91
3.94
4.05
0.91
1.00
4.33
4.42
2.20
3.98
4.09
4.14
4.20
1.69
1.70
4.01
4.46
0.81
0.89
4.44
4.76
2.22
4.04
4.77
d-d , n!p*
p!p*
4
[Co(tolf)2(H2O)2]
[Ni(tolf)2(H2O)2]
[Cu(tolf)2(H2O)]2
4T1g ! A2g
4
4T1g ! T1g (P)
n!p*
p!p*
3
3.47
1.58
3A2g ! T1g
3A2g
n!p*
p!p*
d-d
!
1Eg
343
291
693
351
n!p*
288
p!p*
interaction expresses the reducing ability of the compounds. The scavenging activities
of the complexes were measured, and compared with those of the free drug and of
vitamin C. The scavenging ability of the stable DPPH radical is extensively used as a
model. DPPH is a stable free radical that can accept an electron or hydrogen radical,
and thus be converted into a stable, diamagnetic molecule. DPPH has an odd electron
and so has a strong absorption band at 517 nm. When this electron becomes paired off,
the absorption decreases stoichiometrically with respect to the number of electrons
taken up. Such a change in the absorbance produced in this reaction has been widely
applied to test the capacity of numerous molecules to act as free-radical scavengers. Not
all of the tested compounds were found to interact with the stable free radical DPPH
(Table 2). Tolfenamic acid did not interact with the stable free radical. The CoII, CuII,
and ZnII complexes present low interactions (3.9–27.2%). It seems that interaction
increases with the time and the concentration. The NiII complex exhibits mild
interaction (27.4–44.1%), and the complex [Mn(tolf)2(H2O)2] presents the highest
interaction with DPPH, which is time- and concentration-dependent. Antioxidants that
exhibit DPPH radical-scavenging activity is increasingly receiving attention. They have
been reported to have interesting anticancer, anti-aging, and anti-inflammatory
activities, and could offer protection in rheumatoid arthritis and inflammation, and
lead to potentially effective drugs. In fact, many NSAIDs have been reported to act
either as inhibitors of free-radical production or as radical scavengers.
During the inflammatory process, phagocytes generate the superoxide anion radical
.
at the inflamed site, and this is connected to other oxidizing species such as HO .
Hydroxyl radicals are produced by reactions which are dependent on transition metals,
particularly iron [16]. The competition of the compounds (0.1 mm) with DMSO for
.
HO generated by the Fe3þ/ascorbic acid system, expressed as the inhibition of
.
formaldehyde production, was used for the evaluation of their HO scavenging activity.

CHEMISTRY & BIODIVERSITY – Vol. 6 (2009)
953
Table 2. Interaction with DPPH (scavenging activity (SA) [%])a)
0.1 mm
0.5 mm
20 min
60 min
20 min
60 min
[Mn(tolf)2(H2O)2]
[Co(tolf)2(H2O)2]
[Ni(tolf)2(H2O)2]
[Cu(tolf)2(H2O)]2
[Zn(tolf)2(H2O)]
MnCl2
38.6
13.9
27.4
7.9
3.9
4.2
48.1
17.5
36.9
13.8
11.5
4.1
45.3
18.5
27.4
16.4
3.4
56.8
27.2
44.1
22.9
20.4
3.5
3.2
CoCl2
3.1
2.7
2.7
2.9
NiCl2 ·6 H2O
CuCl2
3.1
4.6
6.1
5.6
5.1
4.9
4.9
5.8
ZnCl2
3.5
5.2
3.8
5.3
Ascorbic acid
Tolfenamic acid
99.5
2.6
99.9
2.8
94.5
4.4
93.9
8.9
a) Each experiment was performed at least in triplicate, and the standard deviation of absorbance was
less than 10% of the mean.
In these experiments (Table 3), Htolf and its MnII and CoII complexes show low
inhibition, while complexes [Ni(tolf)2(H2O)2], Cu(tolf)2(H2O)]2, and [Zn(tolf)2-
(H2O)] exhibit very high competition (77–85%). Trolox was used as a reference
compound.
.
Table 3. Competition with DMSO (0.1 mm) for HO , Inhibition of LOX, and Antiproteolytic Activity of
Tolfenamic Acid and Its Metal Complexesa)
Competition with
Inhibition of
DMSO (0.1 mm) [%]
LOX at 5 min
glucuronidase
trypsin
(0.033 mm) [%]
(0.1mm) [%]
(0.3 mm) [%]
Tolfenamic acid
[Mn(tolf)2(H2O)2]
[Co(tolf)2(H2O)2]
[Ni(tolf)2(H2O)2]
[Cu(tolf)2(H2O)]2
[Zn(tolf)2(H2O)]
MnCl2
16.9
34.9
16.5
77.8
77
no
55.1
31.1
58.7
no
80.8
–
86.7
88.8
no
25.1
71.3
34.2
0.2
no
91.8
b
)
85.1
no
84
–
48
26
CoCl2
no
–
–
NiCl2 ·6 H2O
CuCl2
ZnCl2
Salicylic acid
Trolox
no
no
no
–
–
–
–
–
–
56.4
27
75
24
c
)
88.2
a) Each experiment was performed at least in triplicate, and the standard deviation of absorbance was
less than 10% of the mean. b) Not measured under the reported experimental conditions; no result.
c) IC50 ¼0.1 mm.

954
CHEMISTRY & BIODIVERSITY – Vol. 6 (2009)
Most of the recognized LOX inhibitors are antioxidants or free-radical scavengers
since lipoxygenation occurs via a C-centered radical. The activity of Htolf and its metal
complexes to inhibit the enzyme soybean LOX has been studied. Htolf and its CuII
complex did not exhibit inhibition of the enzyme LOX, while the CoII complex shows
low inhibition (31%). The MnII, NiII, and ZnII complexes are potent LOX inhibitors
(55–81% inhibition). The complex [Zn(tolf)2(H2O)] presents the highest activity
(Table 3).
The role played by proteases in the early stage of inflammatory process is well-
documented. Some anti-inflammatory agents have been reported to exhibit antipro-
teolytic activity [16]. The antiproteolytic activity of the metal complexes and the parent
drug was studied. Htolf and its MnII, NiII, and ZnII complexes were found to inhibit
significantly b-glucuronidase, while the CoII complex did not show any effect against
this enzyme under the reported experimental conditions. Complexes of NiII and CuII
were found inactive against the enzyme trypsin. Htolf and its CoII and ZnII complexes
present low effect (25–48% inhibition). The complex [Mn(tolf)2(H2O)2] was found to
inhibit significantly the enzyme trypsin 71.3% (Table 3).
The in vivo anti-inflammatory effects of Htolf and its metal complexes were
assessed by using the functional model of carrageenin-induced rat paw edema, and they
are presented in Table 4 as percentage of weight increase at the right hind paw in
comparison to the un-injected left hind paw. Carrageenin-induced edema is a
nonspecific inflammation resulting from a complex of diverse mediators [17]. Since
edemas of this type are highly sensitive to NSAIDs, carrageenin has been accepted as a
useful agent for studying new anti-inflammatory drugs [18]. This model reliably
predicts the anti-inflammatory efficacy of the NSAIDs, and, during the second phase, it
detects compounds that are anti-inflammatory agents, as a result of inhibition of
prostaglandin amplification [19]. In Table 4, the protection, observed against carra-
geenin-induced paw edema, by the investigated complexes is shown. The reference
drug, Htolf induced 76.0% protection at an equivalent concentration (tests were
performed at 0.1 mmol/kg Body Weight (BW)). The anti-inflammatory activity of
Htolf, and of the complexes [Co(tolf)2(H2O)2] and [Zn(tolf)2(H2O)] on carrageenin-
induced edema revealed statistically significant anti-inflammatory activity. The
complex [Zn(tolf)2(H2O)] exhibited a strong inhibitory effect at 0.1 mmol/kg BW
(93.0ꢀ0.9% inhibition), superior to the inhibition induced by Htolf at the same dose
(76.0ꢀ0.8% inhibition) on carrageenin-induced edemas. Zn has been reported to give
Table 4. Inhibition of Carrageenin-Induced Rat Paw Edema (CPE)a)
CPEꢀSD [%]b)
Tolfenamic acid
[Co(tolf)2(H2O)2]
[Zn(tolf)2(H2O)]
76*ꢀ0.8
82.6**ꢀ0.8
93** ꢀ 0.9
a) The complexes have been tested at 0.1 mmol/kg. b) The effect on edema is expressed as percent of
weight increase of hind paw (and as percent of inhibition of edema) in comparison to controls. Each value
represents the mean obtained from six to eight animals in two independent experiments. In all cases,
significant difference from control; *: p<0.1; **: p<0.01 (Studentꢂs t-test).

CHEMISTRY & BIODIVERSITY – Vol. 6 (2009)
955
potent anti-inflammatory complexes with other NSAIDs (e.g., Naproxen). The
observed differences in activity seem to depend on the nature of the metal.
The results of cytotoxic activity in vitro are expressed as IC50, i.e., the concentration
of compound (in mm) that inhibits a proliferation rate of the tumor cells by 50% as
compared to control untreated cells (Table 5). Htolf and its metal complexes were
tested for their antiproliferative activity in vitro against the cells of three human cancer
cell lines, MCF-7 (breast cancer cell line), T24 (bladder cancer cell line), and A-549
(non-small-cell lung carcinoma), and a mouse fibroblast L-929 cell line, and the results
are compared with those of the known chemotherapeutic cisplatin.
Table 5. The Antiproliferative Activity in vitro of Tolfenamic Acid and Its Metal Complexes (expressed as
IC50 [mm]) against MCF-7, T-24, A-549, and L-929 Cancer Cell Lines
MCF-7
T-24
62.4ꢀ5.1
A-549
L-929
Tolfenamic acid
[Mn(tolf)2(H2O)2]
[Co(tolf)2(H2O)2]
[Ni(tolf)2(H2O)2]
[Cu(tolf)2(H2O)]2
[Zn(tolf)2(H2O)]
MnCl2
CoCl2
NiCl2
CuCl2
ZnCl2
87.9ꢀ6.3
41.6ꢀ2.9
44.9ꢀ3.0
54.5ꢀ3.3
26.1ꢀ2.1
41.3ꢀ3.5
161ꢀ11
n.a.a)
145ꢀ12
65.3ꢀ6.1
76.8ꢀ7.1
54.1ꢀ4.9
31.4 ꢀ2.2
57.9ꢀ4.2
n.a.
n.a.
n.a.
493ꢀ3
281ꢀ20
1.5ꢀ0.1
214ꢀ18
149ꢀ6
33.4ꢀ2.8
100.1ꢀ9.1
5.3ꢀ0.5
123ꢀ6
n.a.
n.a.
n.a.
n.a.
3.9ꢀ0.3
38.3ꢀ1.9
35.30ꢀ2.1
13.9ꢀ1.1
41.0ꢀ2.5
93.5ꢀ5.9
91.4ꢀ7.3
108ꢀ9
n.a.
337ꢀ22
96.7ꢀ8.7
n.a.
n.a.
35.0ꢀ3.1
Cisplatin
8.0ꢀ0.8
41.7ꢀ4.5
0.7ꢀ0.1
a) n.a.: Not available.
The IC50 value for Htolf against the T24 cell line is 62.4 mm, and the IC50 values for
the complexes [Cu(tolf)2(H2O)]2 and [Mn(tolf)2(H2O)2] are 13.89 and 3.91 mm,
respectively. In the case of T-24 cell line, [Mn(tolf)2(H2O)2] and [Cu(tolf)2(H2O)]2
compared to cisplatin are 10.7 and 3.0 times, respectively, more cytotoxic, while Htolf is
1.5 time less cytotoxic than cisplatin. The complex [Cu(tolf)2(H2O)]2, against L-929
cancer cell lines, is 7.5 times less cytotoxic than cisplatin. The complexes
[Mn(tolf)2(H2O)2] and [Cu(tolf)2(H2O)]2 have shown selectivities against T24 cell
line, and the complex [Cu(tolf)2(H2O)]2 against L-929 cell line. The in vitro results
revealed that Htolf and the metal chlorides are not active or less active than the
corresponding metal complexes of Htolf, indicating that the mechanism of action
should be different.
The cytotoxic activities shown by [Mn(tolf)2(H2O)2] and [Cu(tolf)2(H2O)]2 against
T24 cancer cell line indicate that coupling of Htolf to CuII and MnII centers results in
complexes with important biological properties, since they display IC50 values in a
molar range similar or better to that of the antitumor drug cisplatin.
The reported therapeutic action of these metal complexes could be either due to a
small proportion of metal complexes that can bind to biomolecules and can be
transported to the site of action, or to a ternary metal–tolf–biomolecule complex
which is also the transporting agent [20].

956
CHEMISTRY & BIODIVERSITY – Vol. 6 (2009)
Conclusions. – The complexes of Htolf, i.e., [Mn(tolf)2(H2O)2], [Co(tolf)2(H2O)2],
[Ni(tolf)2(H2O)2], [Cu(tolf)2(H2O)]2, and [Zn(tolf)2(H2O)], were synthesized and
characterized. Optical and IR-spectral data of these new complexes are reported.
Monomeric six-coordinated species were isolated in the solid state for MnII, NiII, and
CoII, dimeric five-coordinated species for CuII, and monomeric five-coordinated species
for ZnII. In DMF or CHCl3 solution, the coordination number is retained, and the
coordinated molecules of H2O are replaced by solvent molecules.
Htolf and its metal complexes have been studied with regard to their antioxidant
ability, as well as their inhibition of LOX, of b-glucuronidase, and of trypsin-induced
proteolysis. The complex [Mn(tolf)2(H2O)2] exhibits the highest antioxidant activity
and a modest inhibitory effect against the soybean LOX, properties that are not
demonstrated by Htolf. The same complex also exhibits a high inhibitory effect on b-
glucuronidase and of trypsin-induced proteolysis. The anti-inflammatory effects of the
complexes on rat paw edema induced by carrageenin were studied and compared with
those of Htolf. The complex [Zn(tolf)2(H2O)] did not interact with DPPH, whereas it
.
exhibited a very high competition with DMSO for HO radical (85.1%), the highest
inhibition against LOX enzyme (80.8%), and a high inhibition against the enzyme b-
glucuronidase (84%). The complex [Zn(tolf)2(H2O)] exhibited the strongest in vivo
inhibitory effect at 0.1 mm/kg BW (93.0ꢀ0.9%), superior to the inhibition induced by
Htolf at the same molar dose (76.0ꢀ0.9%). The IC50 value shown by [Cu-
(tolf)2(H2O)]2 against the L-929 cancer cell lines are, in a molar range similar to
cisplatin, 7.5 times less cytotoxic. The complexes [Mn(tolf)2(H2O)2] and [Cu-
(tolf)2(H2O)]2 have shown selectivity against cell line T24. The IC50 values shown by
these two complexes against cancer cell line T24 are, in a mm range, similar or better to
that of cisplatin.
The aim of this research was to extend the pharmacological profile of Htolf, in order
to search for new properties such as antioxidant and anticancer activity, to prepare new
compounds, i.e., complexes of Htolf with essential metal ions, which probably would
exhibit improved or different biological properties compared to Htolf. The complex
[Zn(tolf)2(H2O)] may prove useful for treating inflammatory diseases, and the
complexes [Mn(tolf)2(H2O)2] and [Cu(tolf)2(H2O)]2 may lead to the development of
an antitumor drug.
Experimental Part
1. General. The reagents (Aldrich and Merck) were used as supplied, while the solvents were purified
according to standard procedures. Trypsin (pancreas protease) 200 Fip U/g, salicylic acid (SA), N-
tosylarginine methyl ester (TAME), 1,1-diphenyl-2-picrylhydrazyl (DPPH), and nordihydroguairetic
acid (NDGA) were purchased from Aldrich Chemical Co. Milwaukee, WI, (USA). Soybean LOX and
linoleic acid sodium salt were obtained from Sigma Chemical, Co. (St. Louis, MO, USA). Carrageenin,
type K, is commercially available. M.p.: in open capillaries; uncorrected. UV Spectra: Jasco V-570
spectrophotometer UV/VIS/NIR; l in nm. IR and far-IR spectra: Perkin-Elmer Spectrum JX Fourier
transform spectrophotometer; with KBr pellets (4000–400 cmꢁ1) and nujol mulls dispersed between
polyethylene disks (400–40 cmꢁ1); in cmꢁ1. Magnetic moments: at r.t. by the Evans method with
mercuric tetrathiocyanocobaltate(II) as the susceptibility standard with a Sherwood Scientific Magnetic
Susceptibility Balance; the effective magnetic moments (meff) values calculated by using the Curie
equation: meff ¼2.83(cMcorr
)
1/2T1/2. C, H, and N analyses were carried out by the microanalytical service of
the University of Ioannina.

CHEMISTRY & BIODIVERSITY – Vol. 6 (2009)
957
2. Preparations. Synthesis of Tolfenamic Acid (Htolf). 2-[(2-Methyl-3-chlorophenyl)amino]benzoic
acid (Htolf) was synthesized according to a published procedure, i.e., the Ullmann–Goldberg
condensation [12]. 3-Chloro-2-methylbenzenamine (0.074 mol), potassium 2-bromobenzoate
(0.076 mol), 4-ethylmorpholine (0.076 mol), and 0.8 g of anh. Cu(OAc)2 in 30 ml of distilled DMF
under N2 were refluxed at 1458 for 4 h. To the resulting soln., 20 ml of distilled DMF and 30 ml of 12%
HCl were added. The aq. layer was decanted, and MeOH was added. The solid was collected and
recrystallized three times from acetone. Yield: 40%. M.p. 2028. IR (KBr): 3341 (br.; ˜n(NH)), 1552
(n˜as(COO)), 1434 (n˜sym(COO)). Anal. calc. for C14H12ClNO2 (261.7): C 64.26, H 4.62, N 5.35; found: C
64.25, H 4.70, N 5.20.
Synthesis of [Mn(tolf)2(H2O)2]. A soln. of MnCl2 (0.0504 g, 0.4 mmol) in MeOH (5.0 ml) was added
to a soln. of Htolf (0.1976 g, 0.8 mmol) in MeOH (2.0 ml). Drops of a methanolic soln. of 1n NaOH were
added until the apparent pH value was ca. 7. The mixture was stirred at r.t. for 1 h and cooled to 58 in a
refrigerator for 4 h. The brown precipitate was collected by filtration and recrystallized from hot MeOH.
The solid was washed with cold MeOH/H2O 5 :1 and dried in vacuo to afford [Mn(tolf)2(H2O)2]. Yield:
14%. M.p. 2808 (dec.). IR: 3414m (n˜(NH)), 1614s (n˜asym(COO)), 1503s (n˜sym(COO)), 1582s
(n˜asym(COO)), 1459s (n˜sym(COO)), 444m–s, 413m–w (n˜(MꢁOH O )), 382ms, 314ms (n˜(MꢁOoco)).
2
meff ¼5.96 MB. Anal. calc. for C28H26Cl2MnN2O6: C 54.92, H 4.28, N 4.58; found: C 54.45, H 4.19, N 4.62.
Synthesis of [Co(tolf)2(H2O)2]. A soln. of Co(OAc)2 ·4 H2O (0.0519 g, 0.4 mmol) in MeOH (5.0 ml)
was added to a soln. of Htolf (0.1976 g, 0.8 mmol) in MeOH (5.0 ml). Drops of a methanolic soln. of 1n
NaOH were added until the apparent pH value was ca. 7. The mixture was stirred at r.t. for 1 h and cooled
to 58 in a refrigerator for 4 h. After the addition of a few drops of dist. H2O, a pale pink powder
precipitated. The pink precipitate was collected by filtration and recrystallized from hot MeOH. The
powder was washed with cold MeOH/H2O 5 :1 and dried in vacuo to afford [Co(tolf)2(H2O)2]. Yield:
34%. M.p. 1458 (dec.). IR: 3335s (n˜(NH)), 1616s (n˜asym(COO)), 1499s (n˜sym(COO)), 1581s
(n˜asym(COO)), 1453s (n˜sym(COO)), 420m–w, 397m–w (n˜(MꢁOH O )), 305m, 282m (˜n(MꢁOoco)). meff
¼
2
4.88 MB. Anal. calc. for C28H26Cl2CoN2O6: C 54.57, H 4.25, N 4.55; found: C 54.09, H 4.52, N 4.31.
Synthesis of [Ni(tolf)2(H2O)2]. A soln. of NiCl2 ·6 H2O (0.0951 g, 0.4 mmol) in MeOH (5.0 ml) was
added to a soln. of Htolf (0.197 g, 0.26 mmol) in MeOH (5.0 ml). Drops of a methanolic soln. of 1n
NaOH, were added until the apparent pH value was ca. 7. The mixture was stirred at r.t. for 1 h and
cooled to 58 in a refrigerator for 4 h. After the addition of a few drops of dist. H2O, a pale green powder
precipitated. The precipitate was collected by filtration and recrystallized from hot MeOH. The powder
was washed with cold MeOH/H2O 5 :1 and dried in vacuo to afford [Ni(tolf)2(H2O)2]. Yield: 47%. M.p.
2458 (dec.). IR: 3451 (br.; ˜n(NH)), 1613s (n˜asym(COO)), 1498 (n˜sym(COO)), 1582s (n˜asym(COO)), 1458
(n˜sym(COO)), 467m–w, 444m–w (v(MꢁOH O )), 324m–w, 280m–w (˜n(MꢁOoco)). meff ¼3.47 MB. Anal.
2
calc. for C26H22Cl2N2NiO6: C 52.81, H 4.26, N 4.19; found: C 53.07, H3.77, N 4.76. Anal. calc. for
C28H26Cl2N2NiO6: C 54.59, H 4.25, N 4.55; found: C 54.27, H 4.07, N 4.26.
Synthesis of [Cu(tolf)2(H2O)]2 ·2 H2O. The complex [Cu(tolf)2(H2O)]2 was prepared according to
the procedure reported in [9]. Yield: 50%. M.p. 1458 (dec.). IR: 3480s (n˜(H2O)), 3414 (n˜(NH)), 1614s
(n˜asym(COO)), 1459m (n˜sym(COO)), 434m, 406m (n˜(MꢁOH O )), 307m–w, 288m–w (n˜(MꢁOoco)). meff
¼
2
1.59 MB. Anal. calc. for C56H60Cl4Cu2N4O12; C 53.81, H 4.83; Cu, 10.17, N 4.48; found: C 53.90, H 4.70, N
4.75.
Synthesis of [Zn(tolf)2(H2O)]. A soln. of ZnCl2 (0.0545 g, 0.4 mmol) in MeOH (5.0 ml) was added to
a soln. of Htolf (0.1976 g, 0.8 mmol) in MeOH (5.0 ml). Drops of a methanolic soln. 1n NaOH were
added until the apparent pH value was ca. 7. The mixture was stirred at r.t. for 1 h. After the addition of a
few drops of dist. H2O, a light yellow powder precipitated. The soln. was cooled to 58 in a refrigerator for
4 h. The precipitate was collected by filtration and recrystallized from hot MeOH. The powder was
washed with cold MeOH/H2O 5 :1 and dried in vacuo to afford [Zn(tolf)2(H2O)]. Yield: 35%. M.p. 2508
(dec.). IR: 3337s (˜n(NH)), 1664s (n˜asym(COO)), 1523s (n˜asym(COO)), 1583s (n˜asym(COO)), 1452s
(n˜sym(COO)), 420m, 397m–w (n˜(MꢁOH O )), 280m–s (n˜(MꢁOoco)). Anal. calc. for C28H24Cl2N2O5Zn: C
2
55.61, H 4.00, N 4.63; found: C 55.15, H 3.89, N 4.37.
3. Experiments in vitro and in vivo [21]. In the in vitro assays, each experiment was performed at least
in triplicate, and the standard deviation of absorbance was less than 10% of the mean.

958
CHEMISTRY & BIODIVERSITY – Vol. 6 (2009)
Determination of the Radical-Scavenging Activity of the Tested Complexes Using Their Interaction
with the Stable Radical 1,1-Diphenyl-2-picrylhydrazyl (DPPH). To a soln. of DPPH (0.1 and 0.5 mm) in
abs. EtOH an equal volume of the compounds dissolved in EtOH was added. As control, EtOH soln. was
used. The concentrations of the solns. of the compounds were 0.1 and 0.5 mm. After 20 and 60 min at r.t.,
the absorbance was recorded at 517 nm.
.
.
Competition of the Tested Compounds with DMSO for HO Radicals. The HO radicals, generated by
the Fe3þ/ascorbic acid system, were detected, according to Nash, by the determination of HCHO
produced from the oxidation of DMSO. The reaction mixture contained EDTA (0.1 mm), Fe3þ (167 mm),
DMSO (0.1 and 6.6 mm) in phosphate buffer (50 mm, pH 7.4), the tested compounds (1 mm), and
ascorbic acid (10 mm). After 30 min of incubation (378), the reaction was stopped with CCl3COOH
(17% (w/v)).
Soybean LOX Inhibition Study in vitro. In vitro study was conducted as reported previously. The
tested compounds (0.033 and 0.333 mm) dissolved in EtOH were incubated at r.t. with sodium linoleate
(0.1 mm) and 0.2 ml of enzyme soln. (1/3ꢂ104 (w/v) in saline). The conversion of sodium linoleate to 13-
hydroperoxylinoleic acid at 234 nm was recorded during 5 min and compared with the appropriate
standard inhibitor (NDGA 0.1 mm, 83.7% inhibition; 1 mm 94.7% inhibition).
Inhibition of b-Glucuronidase and of Proteolysis. Compounds in acetate buffer (0.1 mm, pH 7.4) were
tested against b-glucuronidase (0.1 ml of 1 U/ml) with 2.5 mm p-nitrophenyl-b-d-glucopyranosiduronic
acid. After incubation at 378 for 30 min, 2 ml of 0.5n NaOH soln. was added to the mixture, and the
absorbance of the mixture was measured at 410 nm. TAME was used as substrate for trypsin. The
reaction mixture consisted of 0.75 ml buffer (0.1m Tris·HCl, pH 7.8, in 50% MeOH (v/v)) and 0.14 ml
TAME (0.01m in 50% (v/v) MeOH). Compounds dissolved in 50% MeOH were added (0.3 mm). The
reaction was started by addition of 0.1 ml trypsin (1 mg/ml 0.001n HCl). The increase in the absorbance
at 256 nm was determined in the next 4 min.
Inhibition of the Carrageenin-Induced Edema [21]. Edema was induced in the right hind paw of
Fisher 344 rats (150–200 g) by the intradermal injection of 0.1 ml of 2% carrageenin in H2O. Both sexes
were used. Females pregnant were excluded. Each group was composed of 6–15 animals. The animals,
which have been bred in our laboratory, were housed under standard conditions, and received a diet of
commercial food pellets and water ad libitum during the maintenance, but they were entirely fasted
during the experiment period. Our studies were in accordance with recognized guidelines on animal
experimentation (Guidelines for the care and use of laboratory animals published by the Greek
Government 160/1991, based on EU regulations 86/609). The tested compounds, 0.01 or 0.1 mmol/kg
BW, were suspended in H2O with a few drops of Tween 80 and ground in a mortar before use, or dissolved
in H2O and were given intraperitoneally simultaneously. The rats were euthanized 3.5 h after carrageenin
injection. The difference between the weight of the injected and uninjected paws was determined for
each animal. The change in paw weight was compared with that in control animals (treated with H2O)
and expressed as a percent inhibition of the edema CPE % values (Table 4). CPE % Values are the mean
from two different experiments with a standard error of the mean less than 10%.
Antiproliferative Assay in vitro [22–24]. Compounds. Test solns. of the compounds tested (1 mg/ml)
were prepared by dissolving the substance in 100 ml of DMSO completed with 900 ml of tissue culture
medium. Then, the tested compounds were diluted in culture medium to reach the final concentrations of
100, 50, 10, 1, and 0.1 mg/ml. The solvent (DMSO) in the highest concentration used in the test did not
reveal any cytotoxic activity.
Cells. The established in vitro human cancer cell lines MCF-7 (breast cancer cell line), T24 (bladder
cancer cell line), and A-549 (non-small-cell-lung carcinoma), and a mouse L-929 (a fibroblast-like cell
line cloned from strain L) were applied. The cell lines are maintained in the Cell Culture Collection of the
University of Ioannina. Twenty-four h before addition of the tested agents, the cells were plated in 96-
well plates at a density of 104 cells per well. The T-24 and MCF-7 cells were cultured in D-MEM
(Modified Eagleꢂs Medium) supplemented with 1% antibiotic and 10% fetal calf serum. L-929 Cells were
grown in Hepes-buffered RPMI 1640 medium supplemented with 10% fetal calf serum, penicillin (50 U/
ml), and streptomycin (50 mg/ml). A-549 Cells were grown in F-12K Hamꢂs medium supplemented with
1% glutamine, 1% antibiotic/antimycotic, 2% NaHCO3, and 10% fetal calf serum. The cell cultures were
maintained at 378 in a humid atmosphere saturated with 5% CO2. Cell number was counted by the

CHEMISTRY & BIODIVERSITY – Vol. 6 (2009)
959
Trypan Blue dye exclusion method. MCF-7, L-929, and A-549 cells were determined by the
sulforhodamine B assay [22], and T24 cells by the MTT (¼ 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-
2H-tetrazolium hydrobromide) assay.
SRB (Sulforhodamine B) Assay. The details of this technique were described by Skehan et al. [23].
The cytotoxicity assay was performed after 72-h exposure of the cultured cells to varying concentrations
(from 0.1 to 100 mg/ml) of the tested agents. The cells attached to the plastic were fixed by gently layering
cold 50% Cl3CCOOH (TCA) on the top of the culture medium in each well. The plates were incubated at
48 for 1 h and then washed five times with tap water. The background optical density was measured in the
wells filled with culture medium, without the cells. The cellular material fixed with TCA was stained for
30 min with 0.4% sulforhodamine B dissolved in 1% AcOH. Unbound dye was removed by rinsing with
1% AcOH. The protein-bound dye was extracted with 10 mm unbuffered Tris base for determination of
optical density (at 540 nm) in a computer-interfaced, 96-well microtiter plate.
MTT (Sigma). This technique was applied for cytotoxicity screening against T-24 cells growing in
suspension culture. An assay was performed after 72-h exposure to varying concentrations (from 0.1 to
100 mg/ml) of the tested agents. For the last 3–4 h of incubation 20 ml of MTT soln. was added to each
well (stock soln.: 5 mg/ml). The mitochondria of viable cells reduce a pale yellow MTT to a navy blue
formazan, so, if more viable cells are present in the well, more MTT will be reduced to formazan. When
the incubation time was completed, 80 ml of the lysing mixture was added to each well (lysing mixture:
225 ml of DMF, 67.5 g of sodium dodecylsulfate (both from Sigma), and 275 ml of dist. H2O). After 24 h,
when formazan crystals had dissolved, the optical densities of the samples were read on an Eliza
SpectraMax 190 photometer at 570-nm wavelength. Each compound at a given concentration was tested
in triplicate in each experiment, which was repeated three times [24].
REFERENCES
[1] O. Eskerod, Pharmacol. Toxicol. 1994, 75 (Suppl. II), 44.
[2] P. A. Insel, in ꢃGoodman and Gilmanꢂs The Pharmacological Basis of Therapeuticsꢂ, McGraw-Hill,
New York, 1996, chapt. 27.
[3] T. Kawamori, C. V. Rao, K. Seibert, B. S. Reddy, Cancer Res. 1998, 58, 409.
[4] H. M. Tucker, R. E. Rydel, S. Wright, S. Estus, J. Neurochem. 1998, 71, 506.
[5] R. Cini, Comments Inorg. Chem. 2000, 22, 151, and refs. cit. therein.
[6] D. Kovala-Demertzi, J. Inorg. Biochem. 2000, 79, 153, and refs. cit. therein.
[7] K. L. Andersen, S. Larsen, B. Alhede, N. Gelting, O. Buchardt, J. Chem. Soc., Perkin Trans. II 1989,
1443.
[8] D. Kovala-Demertzi, N. Kourkoumelis, A. Koutsodimou, A. Moukarika, E. Horn, E. R. T. Tiekink,
J. Organomet. Chem. 2001, 620, 194.
[9] D. Kovala-Demertzi, A. Galani, M. A. Demertzis, S. Skoulika, C. Kotoglou, J. Inorg. Biochem. 2004,
98, 358, and refs. cit. therein.
[10] J. E. Weder, C. T. Dillon, T. W. Hambley, B. J. Kennedy, P. A. Lay, J. R. Biffin, H. L. Regtop, N. M.
Davies, Coord. Chem. Rev. 2002, 232, 95.
[11] D. Kovala-Demertzi, J. Organomet. Chem. 2006, 691, 1767.
[12] J. S. Kaltenbrom, R. A. Scherrer, F. W. Short, H. R. Beatty, M. M. Saka, C. V. Winder, J. Wax,
W. R. N. Williamson, Arzneim. Forsch. 1983, 33, 621.
[13] D. Kovala-Demertzi, A. Theodorou, M. A. Demertzis, C. Raptopoulou, A. Terzis, J. Inorg.
Biochem. 1997, 5, 151.
[14] D. Kovala-Demertzi, S. K. Hadjikakou, M. A. Demertzis, Y. Deligiannakis, J. Inorg. Biochem. 1998,
69, 223.
[15] A. B. P. Lever, ꢃInorganic Electronic Spectroscopyꢂ, 2nd edn., Elsevier, Amsterdam, 1984, p. 480–
505.
[16] M. Tandon, P. Tandon, J. P. Barthwal, T. N. Bhalla, K. P. Bhargava, Arzneim. Forsch. 1982, 32, 1233.
[17] T. Y. Shen, ꢃNon-steroidal anti-inflammatory agentsꢂ, in ꢃBurgerꢂs Medicinal Chemistryꢂ, Ed. M. E.
Wolf, John Wiley & Sons, New York, 1980, p. 1217–1219.

960
CHEMISTRY & BIODIVERSITY – Vol. 6 (2009)
[18] C. A. Winter, ꢃAnti-Inflammatory Testing Methods: Comparative Evaluation of Indomethacin and
Other Agentsꢂ, in ꢃNon-Steroidal Anti-Inflammatory Drugsꢂ, Eds. S. Garattini and M. N. G. Dukes,
Excerpta Medica Foundation, Amsterdam, 1965, pp. 190–202.
[19] T. Kuroda, F. Suzuki, T. Tamura, K. Ohmori, H. Hosoe, J. Med. Chem. 1992, 53, 1130.
[20] A. Theodorou, M. A. Demertzis, D. Kovala-Demertzi, E. E. Lioliou, A. A. Pantazaki, D. A.
Kyriakidis, BioMetals 1999, 12, 167.
[21] C. Kontogiorgis, D. Hadjipavlou-Litina, J. Enzyme Inhib. Med. Chem. 2003, 18, 63.
[22] M. Ohno, T. Abe, J. Immunol. Methods 1991, 145, 199.
[23] P. Skehan, R. Storeng, D. Scudiero, A. Monks, J. McMahon, D. Vistica, J. T. Warren, Heidi Bokesch,
S. Kenney, M. R. Boyd, J. Natl. Cancer Inst. 1990, 82, 1107.
[24] D. Kovala-Demertzi, A. Galani, N. Kourkoumelis, J. R. Miller, M. A. Demertzis, Polyhedron 2007,
26, 2871.
Received March 24, 2008