Fluorescence emission and enhanced photochemical stability of ZnII-5-triethyl ammonium methyl salicylidene ortho-phenylendiiminate interacting with native DNA

Article abstract of DOI:10.1016/j.jinorgbio.2010.03.012

The photophysical and photochemical properties of the cationic Zn^II complex of 5-triethyl ammonium methyl salicylidene ortho-phenylendiimine (ZnL²⁺) interacting with native DNA were investigated by steady state and time-resolved fluo

Full text of DOI:10.1016/j.jinorgbio.2010.03.012

Journal of Inorganic Biochemistry 104 (2010) 765–773
Contents lists available at ScienceDirect
Journal of Inorganic Biochemistry
journal homepage: www.elsevier.com/locate/jinorgbio
Fluorescence emission and enhanced photochemical stability of ZnII-5-triethyl
ammonium methyl salicylidene ortho-phenylendiiminate interacting with
native DNA
b
Giampaolo Barone a, , Angela Ruggirello , Arturo Silvestri a,1, Alessio Terenzi a, Vincenzo Turco Liveri b
⁎
a
Dipartimento di Chimica Inorganica e Analitica “S. Cannizzaro”, Università di Palermo, Viale delle Scienze, Parco d'Orleans II, Edificio 17, 90128, Palermo, Italy
Dipartimento di Chimica Fisica “F. Accascina”, Università di Palermo, Viale delle Scienze, Parco d'Orleans II, Edificio 17, 90128 Palermo, Italy
b
a r t i c l e i n f o
a b s t r a c t
The photophysical and photochemical properties of the cationic ZnII complex of 5-triethyl ammonium
methyl salicylidene ortho-phenylendiimine (ZnL2+) interacting with native DNA were investigated by
steady state and time-resolved fluorescence spectroscopies. Experimental results indicate that, in the
presence of DNA, ZnL2+ is efficiently protected from a photochemical process, which occurs when it is in the
free state dispersed in aqueous solution. The analysis of the absorption and emission spectra of ZnL2+, both
stored in the dark and after exposure to tungsten lamp light for 24 h, corroborated by quantum chemical
calculations, allowed us to point out that ZnL2+ undergoes a photoinduced two-electron oxidation process.
According to this picture, the protective action of DNA toward the intercalated ZnL2+ was attributed to an
effective inhibition of the ZnL2+ photooxidation. In this context, it can be considered that DNA-intercalated
ZnL2+ is located in a region more hydrophobic than that sensed in the bulk water solvent. Moreover, by a
thorough analysis of steady state and time-resolved fluorescence spectra, the interaction process can be
consistently explained in terms of a complete intercalation of the complex molecules and that the polarity of
the environment sensed by intercalated ZnL2+ is comprised between that of methanol and ethanol.
© 2010 Elsevier Inc. All rights reserved.
Article history:
Received 8 January 2010
Received in revised form 23 March 2010
Accepted 23 March 2010
Available online 31 March 2010
Keywords:
DNA
Fluorescence spectroscopy
Intercalation
Photooxidation
Schiff bases
Zinc
1. Introduction
and water-soluble. Usually, water solubility has been increased by
functionalizing the ligand by polar or charged groups [1–13].
Salen and Salphen, the di-anions of the Schiff bases N,N′-
ethylenebis (salicylideneimine) and N,N′-phenylenebis (salicylide-
neimine), respectively, and ligands obtained by their chemical
modifications, are of current interest for their potential and actual
biomedical applications [1–13]. This is due to their peculiar molecular
structure characterized by a nearly planar area with extended π
system. Moreover, a wide class of complexes has been synthesized
from these ligands that are able to coordinate metal ions via a square
planar N2O2 system.
Among the various mechanisms through which they carry on their
action in bioenvironments, of utmost importance is their direct
interaction with DNA. Many DNA binding studies of metal complexes
of Salen-type ligands have been reported in the literature, for example
of Al [1], V [2], Mn [3–7], Fe [8,9], Co [6,10] Ni [2,5,6,11] Cu [2,6,12,13]
and Zn [2]. The capability to interact with DNA is determined by
several factors such as the nature of the ligand and the coordination
geometry. Further necessary requisites that such complexes should
obviously possess are to be stable and inert in biological environment
Schiff base ligands and their metal complexes show interesting
photophysical properties and display comparable fluorescence with
excitation maxima at about 300 nm and emission maxima at about
420 nm [14]. For the ligand, fluorescence occurs through the proton
transfer between the hydroxy and azomethine groups in the
chromophore. Consequently, the polarity of the solvent has a marked
effect [15]. Fluorescence emission from intraligand excited states of
metal–Salen complexes has also been detected [16]. In particular, Zn
(Salen) and related complexes are highly fluorescent [17] and their
spectra are consistent with singlet emission in competition with
efficient singlet-to-triplet intersystem crossing [18]. Concerning the
photochemical properties of ZnII complexes of Salen-type Schiff bases,
these undergo a ZnII mediated two-electron oxidation, due to
simultaneous one-electron oxidation of the two keto-imine moieties
of the bridging ligand. In fact, two redox processes at positive
potential have been detected: the first oxidation process is likely
localized on the phenolate ring of the Schiff base ligand while the
second oxidation process may be a subsequent oxidation to form
dicationic species [19]. Furthermore, in the presence of water, able to
coordinate the Zn metal in the apical position, Zn(Salphen) complexes
dissolved in relatively acidic organic solvents, such as CHCl3, are
known to lead to demetallated structures [20].
⁎
Corresponding author. Tel.: +39 091427584; fax: +39 091590015.
E-mail addresses: gbarone@unipa.it (G. Barone), turco@unipa.it (V.T. Liveri).
Professor Arturo Silvestri deceased on July 26th 2009.
1
0162-0134/$ – see front matter © 2010 Elsevier Inc. All rights reserved.
doi:10.1016/j.jinorgbio.2010.03.012

766
G. Barone et al. / Journal of Inorganic Biochemistry 104 (2010) 765–773
The photophysical, photochemical and solution properties of ZnII
complexes recently reported [14–20] induced us to revisit the issue of
stability of the ZnII complex of 5-triethyl ammoniummethyl salicylidene
ortho-phenylendiimine (H2L2+, see Fig. 1) and further investigate the
effect of light on ZnL2+ water solutions, also in the presence of native
DNA. In fact, we have recently reported on the interaction of native DNA
with NiL2+, CuL2+ and ZnL2+ complexes [21,22], CuL2+ also confined in
reverse micelles to simulate the intracellular solution environment [23].
The results obtained by UV spectrophotometric titrations, circular
dichroism, DNA thermal denaturation measurements, and fluorescence
quenching of ethidium bromide–DNA solutions, collectively show that
ML2+ (M=NiII, CuII, and ZnII) strongly interact with native DNA, by a
combined electrostatic and intercalative mechanism [21,22]. Consider-
ing the photophysical properties of ZnII–Schiff base complexes
describedabove, fluorescence spectroscopymeasurements were mainly
exploited, as supported by quantum chemical calculations. In this way,
we have extended our previous investigations of the DNA binding
properties of ZnL2+ in aqueous solution, to achieve a detailed picture of
the interaction mechanism and underlying phenomenology.
Fig. 2. Absorption spectra of buffered aqueous solutions of DNA, [DNAphosphate]=
5×10−5 M, (- - -) and of ZnL2+, [ZnL2+]=5×10−5 M, ( ).
2. Materials and methods
mixing the aqueous stock solutions of ZnL2+ ([ZnL2+]=1×10−4 M) and
DNA ([DNAphosphate]=1.31×10−2 M).
The complex ZnL2+ was synthesized as recently reported [22] by the
reaction of 5-(triethylammoniummethyl) salicylaldehyde chloride, 1,2
phenylendiamine and zinc(II) acetate in a 2:1.1:1 molar ratio [13], in
doubly distilled water, and isolated as perchlorate salts. The 5-
(triethylammoniummethyl) salicylaldehyde chloride ligand was pre-
pared from 5-chloromethyl salicylaldehyde [24] and triethylamine in
THF. The products were characterized by elemental analysis, UV and
NMR spectroscopy, as reported [22]. 5-(triethylammoniummethyl)
salicylaldeide chloride (H2LCl2) 1H-NMR (300.13 MHz, D2O, t, triplet; q,
quartet; dd, double doublet, m, multiplet) δ (ppm): 1.33–1.38 (t, 9H,
J=6.8 Hz, CH3); 3.11–3.23 (q, 6H, J=7.0 Hz, CH2); 4.41 (s, 2H, CH2);
7.01–7.04 (d, 1H, J=9.5 Hz, Ar); 7.59–7.64 (dd, 1H, J=2.5 Hz,
J=10.0 Hz, Ar); 7.81–7.82 (d, 1H, J=2.5 Hz, Ar); and 10.08 (s, 1H,
CHO); the 1H-NMR spectrum of H2LCl2 exposed to tungsten lamp light
for 24 h is practically identical to that of H2LCl2 stored in the dark. ZnL
(ClO4)2 1H-NMR (300.13 MHz, D2O, sample stored in the dark) δ (ppm):
1.37 (t, 18H, CH3, J=7.0 Hz); 3.20 (d, 12H, CH2, J=7.1 Hz); 4.26 (s, 4H,
CH2); 6.90 (d, 2H, Ar, J=8.8 Hz); 7.22–7.55 (m, 6H, Ar); 7.55–7.84 (m,
2H, Ar); and 8.88 (s, 2H, CH); the 1H-NMR spectrum of ZnL(ClO4)2
exposed to tungsten lamp light for 24 h is practically identical to that of
ZnL(ClO4)2 stored in the dark. The NMR spectra were recorded on a
Bruker Avance 300 spectrometer.
UV-visible (UV-vis) spectra were recorded at 25 °C in the wave-
length range 200–800 nm on a Perkin–Elmer (Lambda-900) spectrom-
eter using Suprasil quartz cells with 10 mm optical path length. Steady-
state fluorescence measurements were recorded at 25 °C in the
wavelength range 360–700 nm on a Horiba Jobin Yvon spectrofluorim-
eter (Fluoromax-4) with excitation at 370 nm and arranged in T-shaped
geometry. Time-correlated single photon counting (TCSPC) measure-
ments were carried out using the Fluoromax-4 apparatus equipped with
a Single Photon Counting Controller (FluoroHub, HoribaJobin Yvon) and
a pulsed diode light source (Nanoled, repetition rate 1 MHz, pulse
duration 1.27 ns, 368 nm); data fitting was accomplished using least-
squares methods with DAS6 Fluorescence Decay Analysis Software. All
spectra were made using a 10 mm path length cuvette and were
corrected for buffer fluorescence.
3. Theoretical calculations
The geometry of the H2L2+ Schiff base ligand and of ZnL2+ ground
states, both the square planar and the pentacoordinated aquo-
complexes [27] (Fig. 1), as well as those of their mono and dioxidized
forms, was fully optimized by the spin unrestricted Density Functional
Theory (DFT) B3LYP method [28], using the all-electron double-zeta
split-valence plus polarization (DZVP) basis set [29,30]. The spin state
of the considered species was singlet (S0) for the lowest oxidation
state, doublet (D0) for the monooxidized form and both singlet (S0)
and triplet (T0) spin states for the dioxidized forms, respectively.
Their absorption spectra were calculated by the Time Dependent DFT
(TD-DFT) method [31–34], by using the same functional and basis set
described above. To evaluate the emission spectra of the ZnL2+ aquo-
complexes, in the reduced and dioxidized forms, TD-DFT calculations
were performed on the first singlet excited states (S1) of the ZnII
(Salphen) simplified model complex, ZnL′ (Fig. 1, H2L′=N,N′-
phenylene-bis(salicylideneimine)), obtained by substitution of the
two triethyl ammonium methyl cationic groups by two hydrogen
atoms, whose geometry was optimized by the Configuration Interac-
tion Singles (CIS) method [35,36] and using the DZVP basis set.
Although it is known that the accuracy of the CIS method is
comparable to that of the Hartree–Fock method for the ground state
[37], nevertheless the main structural differences between the ground
and first excited states are consistently reproduced by the CIS
approach [38]. Implicit solvent effects in the electronic transitions
were considered by single point calculations within the conductor-
like polarized continuum model [39]. All calculations were performed
using the Gaussian03 program package [40]. The calculated
Lyophilized calf thymus DNA (Fluka, BioChemika) was resuspended in
1.0×10−3 M tris–hydroxymethyl-aminomethane (Tris–HCl) pH 7.5 and
dialyzed as described in the literature [25]. DNA concentration, expressed
in monomers units ([DNAphosphate]), was determined by UV spectropho-
tometry using the molar absorption coefficient 7000 M−1 cm−1 at
258 nm [26]. All experiments were carried in Tris–HCl aqueous buffer
at pH 7.5. Samples at fixed ZnL2+ concentration ([ZnL2+]=5×10−5 M)
and various R values (R=[DNAphosphate]/[ZnL2+]) were prepared by
Fig. 1. Structure of the pentacoordinated aquo-complexes ZnL·H2O2+ (H2L2+ =5-
triethyl ammonium methyl salicylidene ortho-phenylendiimine, R=CH2NEt+3 ) and ZnL′·
H2O (H2L′=N,N′-phenylene-bis(salicylideneimine), R=H).

G. Barone et al. / Journal of Inorganic Biochemistry 104 (2010) 765–773
767
absorption and emission electronic spectra were reproduced by the
help of the GaussSum-2.1.6 program package [41].
allowing us to collect the fluorescence spectra without significant
sample photodegradation.
It was found that, using the excitation wavelength of 370 nm, an
entrance slit width of 0.24 mm and a scan rate of 6 nm/s nearly identical
replicate spectra were collected. Furthermore, in order to evaluate the
photochemical stability of ZnL2+, the fluorescence spectra of some
selected samples at various R values (R=[DNAphosphate]/[ZnL2+]) were
collected at t=0 h (freshly prepared samples) and after 24 h by
maintaining the samples at room temperature i) in the dark or ii) under
the illumination of a tungsten lamp (100 watt) placed at a fixed distance
of 30 cm from the quartz cuvette (10 mm path length) containing the
sample. The comparison among these spectra is shown in Fig. 3. It can be
noted that, after 24 h, all the samples maintained in the dark are only
marginally affected. On the other hand, while the sample at R=0 after
illumination displays a marked enhancement of its fluorescence
intensity, by increasing the amount of DNA this effect progressively
disappears being negligible at R≥5. Taking into account that ZnL2+ is
intercalated within the DNA structure [22], quite surprisingly this
4. Results and discussion
4.1. Photochemical behavior of ZnL2+ and protective effect of DNA
Preliminary experiments were carried out to ascertain the feasibility
of the spectrofluorimetric investigation of the photophysical and
photochemical properties of ZnL2+ intercalated in native calf thymus
DNA. First of all, the absorption spectra of a buffered aqueous solution of
DNA ([DNAphosphate]=5×10−5 M) and of ZnL2+ ([ZnL2+]=5×10−5 M)
were collected to select the best suited excitation wavelength. Such
spectra are shown in Fig. 2. Since the complex shows an absorption band
centered at about 370 nm in a region where DNA does not absorb, we
selected this excitation wavelength for our fluorescence experiments.
After detecting a fluorescence band in the 360–700 nm region,
suitable experimental conditions and apparatus set up were searched,
Fig. 3. Fluorescence spectra of samples at fixed ZnL2+ concentration, [ZnL2+]=5×10−5 M, and various R values: a, R=0; b, R=0.2; c, R=0.8; d, R=2.0; e, R=5.0; f, R=10.0, for t=0 h (- - -),
stored in the dark for t=24 h ( ) and exposed to tungsten lamp light for t=24 h ( ).

768
G. Barone et al. / Journal of Inorganic Biochemistry 104 (2010) 765–773
finding indicates that the insertion of the ZnII complex in the DNA
double helix involves an efficient protection against the observed
photochemical process. Moreover, it emphasizes a novel and interesting
functionality of the DNA molecules which could be exploited for
biomedical applications. In fact, being ZnL2+ representative of poten-
tially interesting DNA-intercalating drugs, our finding could suggest the
formulation of photounstable drugs that become photostable after DNA
intercalation. Therefore, it should be of utmost interest to speculate i) on
the nature of this photochemical process and ii) on theability of the DNA
double helix to hinder it.
Measurements performed at different light exposition times show
that the fluorescence intensity of the ZnL2+ solutions exposed to light
has a drastic enhancement within the first 24 h. After this period,
further exposition to light negligibly changes both the fluorescence
intensity and the band shape. Moreover, the fluorescence spectra of
the samples stored in the dark for several days, after prolonged light
exposition, are essentially coincident with those registered immedi-
ately after light exposition. These results allow us to conclude that: 1)
the photoproduct is stable and 2) the process is irreversible.
Interestingly, the same dramatic intensity enhancement was ob-
served, after exposure to tungsten light for 24 h, for deoxygenated
solutions of the ZnL2+ complex in Tris–HCl 1 mM. The latter were
obtained by extensively insufflating nitrogen gas in the solvent and
keeping it under nitrogen atmosphere.
The striking increase in the fluorescence quantum yield of the
illuminated ZnL2+ in the free state strongly suggests the occurrence
of a photochemical process resulting in an increase of the structural
rigidity of the zinc(II) complex that remarkably reduces the
radiationless decay rate from the excited to the ground state [42].
An additional support to this hypothesis comes from the light
sensitivity observed in a CoII–Salen-type complex, that is irrevers-
ibly photooxidized to a cationic species, and in which the electron is
not removed from the metal but from the ligand [43]. Moreover, it
has been reported that a MnIII complex of a Salen-type ligand in
CHCl3 is photooxidized to a MnIV complex although, in addition,
secondary processes can lead to various photoproducts [44].
Concerning the latter remark, it is known that visible light promotes
the oxidation of the ligand in a MnIII Salen-type complex, followed
by the hydrolysis and rearrangement of the coordinated Schiff base
ligand [45].
According to these considerations and recalling the results by
Germain et al. [19], proving that Salen-type Schiff bases may undergo
a ZnII mediated two-electron oxidation, we attempted to rationalize
our experimental findings in terms of the photooxidation of the ZnL2+
complex, by comparing experimental spectra with those obtained
through TD-DFT calculations.
It is well known that, in aqueous solutions at neutral pH, any kind
of native DNA, independently from its chain length distribution and
base pair composition is characterized by the same surface charge
density [46]. It is then reasonable to expect that also native DNA from
any other source than from calf thymus should play the same
protective effect against ZnL2+ photooxidation.
It should also be noted that the fluorescence spectrum of aqueous
H2L2+ Schiff base ligand is also somewhat influenced by light
exposure but, remarkably, it is strongly different from that of the
illuminated ZnL2+ sample at R=0 (see Fig. 4).
This leads us to exclude that the observed photochemical process
consists in the complex demetallation. An explanation of the latter
evidence may be given by considering that bulk water is known to be a
base less strong than water traces dissolved in apolar solvents (or in
relatively acidic solvents such as CHCl3). Moreover, the presence of the
Tris–HCl buffer at neutral pH may enhance the stability of the ZnL2+
complex towards demetallation [20].
Fig. 4. Fluorescence spectra of aqueous solutions of ZnL2+ and H2L2+, [ZnL2+]=[H2L2+]=
5×10−5 M, stored in the dark (ZnL2+- - -, H2L2+ ) and exposed to tungsten lamp light for
24 h (ZnL2+ , H2L2+- . -).
Remarkably, the 1H-NMR spectrum of ZnL2+ exposed to light shows
the same peaks of the sample stored in the dark. Considering that
the photooxidation of the ZnL2+ complex does not modify
significantly its structure, these findings are consistent with the
hypothesis that light exposure causes partial photooxidation of the
ZnL2+ complex.
4.2. Comparison between experimental and calculated absorption and
emission spectra of ZnL2+ and ZnL′
The absorption spectra of the H2L2+ ligand, of the ZnL2+
tetracoordinated complex (Fig. 1) and of the ZnL2+·H2O pentacoor-
dinated complex (Fig. 1) were calculated by the TD-DFT method, in
the spin states S0 for the lowest oxidation state, D0 for the
monooxidized form and S0 and T0 for the dioxidized forms.
The comparison between calculated and experimental absorption
spectra shown in Fig. 6, induced us to consider that in water solution
ZnL2+ should be coordinated by an axial water molecule and the
presence of square planar ZnL2+ and/or of the H2L2+ ligand can be
safely excluded. In fact, there is a better matching between the
experimental absorption spectrum of ZnL2+, stored in the dark
(Fig. 6d), and that calculated for ZnL2+·H2O in the singlet spin ground
state (Fig. 6b, see also Fig. 7 dashed lines).
The latter conclusion is better illustrated by Fig. 7, where the
absorption spectrum of ZnL2+·H2O S0 and the average of the two
spectra corresponding to the reduced ZnL2+·H2O S0 and dioxidized
ZnL2+·H2O T0 state are reported in Fig. 7b. In particular, the similarity
between the calculated absorption spectrum obtained by a 50%
mixture of the S0 ZnL2+·H2O and T0 ZnL4+·H2O with experimental
absorption spectrum of illuminated ZnL2+ in water solution, supports
the hypothesis that the effect of the tungsten lamp light for 24 h is to
partially oxidize ZnL2+ to ZnL4+, up to an approximately equimolar
ratio between the two forms. To understand the role of light exposure
on the fluorescence emission of ZnL2+ water solutions, we have
calculated the structure of the first excited state of simplified model
complexes ZnL′·H2O and ZnL′2+·H2O (see Fig. 1) by the CIS method,
in the S1 spin states for both the reduced and the dioxidized states
(see Theoretical calculations section). Their emission spectra calcu-
lated by the TD-DFT method are shown in Fig. 8. Despite the
remarkable difference between the experimental fluorescence inten-
sities of ZnL2+ solutions stored in the dark and illuminated, due to
their different nonradiative decay rates, the shape of the calculated
emission spectra in Fig. 8 nicely reproduces the spectral changes
accompanying the exposure to light. In fact, in the experimental
spectrum (Fig. 8a) the peak at 420 nm is red shifted in the illuminated
To further investigate photochemical stability in water solutions
of ZnL2+ 1H-NMR spectra were recorded in D2O for samples stored
,
in the dark and exposed to tungsten lamp light for 24 h (see Fig. 5).

G. Barone et al. / Journal of Inorganic Biochemistry 104 (2010) 765–773
769
Fig. 5. 1H-NMR spectra of ZnL(ClO4)2 in D2O, stored in the dark (a) and exposed to tungsten lamp light for 24 h (b).
ZnL2+ solutions and the intensity of the peak at 480 nm decreases
compared to that at 420 nm. Moreover, the analysis of Figs. 6 and
8 shows that the ground state of the dioxidized ZnL4+ aquo-complex
is a triplet spin state, while the emission from the first excited state of
the dioxidized species occurs from the S1 singlet spin state,
presumably following a triplet to singlet intersystem crossing.
Fig. 6. Calculated absorption spectra (εcalc) and oscillator strengths (f) of ZnL2+ (a), ZnL2+·H2O (b) and H2L2+ (c) in the reduced S0 state (—), in its monooxidized D0 state ( ) and
in the dioxidized S0 ( ) and T0 ( ) spin states, within the implicit water solvent; (d) experimental absorption spectra of ZnL2+ ([ZnL2+]=5×10−5 M), stored in the dark (—) and
exposed to the tungsten lamp light for 24 h ( ).

770
G. Barone et al. / Journal of Inorganic Biochemistry 104 (2010) 765–773
Fig. 8. (a) Experimental fluorescence spectra of ZnL2+, [ZnL2+]=5×10−5 M, stored in
the dark (- - -) and exposed to the tungsten lamp light for 24 h ( ); (b) calculated
Fig. 7. Experimental absorption spectra of ZnL2+ (a) and calculated spectra of ZnL2+·H2O
within the implicit water solvent (b): (a) experimental ([ZnL2+]=5×10−5 M), stored in
the dark (- - -) and exposed to the tungsten lamp light for 24 hours ( ); (b) reduced S0
state (- - -) and 50% mixture of the reduced S0 state and dioxidized T0 state ( ).
emission spectra (εcalc
) and oscillator strengths (f) for the simplified ZnL′·H2O
complex, within the implicit water solvent, reduced S1 state (- - -) and dioxidized S1
state species ( ).
Noteworthy, the increase in the positive charge accompanying the
two-electron oxidation process, may well explain the increase of the
photoinduced structural rigidity of the cationic zinc(II) complex, as
deduced above from the analysis of the experimental spectra (see
Fig. 3a). The most intense electronic transitions of the absorption
spectra calculated for ZnL2+·H2O S0 and for ZnL4+·H2O T0 (see
Fig. 6b) and of the emission spectra calculated for ZnL′·H2O S1 and for
ZnL′2+·H2O S1 (see Fig. 8b), can be attributed to electronic transitions
among the Molecular Orbitals (MOs) depicted in Fig. 9, as detailed in
Table 1.
Interestingly, it can be noticed that the contribution of the orbitals
centered on the metal ion is negligible in all the MOs represented
Fig. 9. This result shows that the calculated electronic transitions, of
both the absorption and emission spectra, involve intraligand states,
in agreement with the attributions reported in the literature [16].
Another interesting finding concerns the similarity between the
MOs involved in the absorption and emission electronic transitions.
For example, the shape of the Highest Occupied Molecular Orbital
(HOMO), HOMO-4, Lowest Unoccupied Molecular Orbital (LUMO)
and LUMO+1 orbitals calculated for ZnL2+·H2O essentially coincide
with that of the HOMO, HOMO-1, LUMO and LUMO+1 orbitals
calculated for ZnL′·H2O. On the other hand, being ZnL4+·H2O in the
triplet spin state and ZnL′2+·H2O in the singlet spin state, the shape of
the HOMO and LUMO of ZnL4+·H2O is comparable to that of the
LUMO and LUMO+1 of ZnL′2+·H2O. Finally, it is worth considering
that the triethyl ammonium methyl groups are essentially not
involved in the MOs of both ZnL2+·H2O and ZnL4+·H2O shown in
Fig. 9. This suggests that ZnL′·H2O is a reliable model for describing
the excited state of ZnL2+·H2O. Table 2 shows the relevant bond
lengths and angles involving the oxygen and nitrogen atoms
coordinated to the Zn atom in the four metal complexes ZnL2+·H2O,
ZnL4+·H2O, ZnL′·H2O and ZnL′2+·H2O. The geometry of the first
singlet excited state of the reduced and oxidized forms of the ZnII
complex has been obtained on a simplified system (see Fig. 1) and
using the CIS method. The latter is less accurate than the B3LYP
method used for the description of the reduced and oxidized forms in
the ground state. Nevertheless the analysis of Table 2 suggests that
only minor structural modifications occur on the coordination site of
the title complex when it is photooxidized and/or when it is
photoexcited. However, interestingly, such small distortions largely
take into account of the remarkable spectral differences, experimen-
tally detected both in the absorption and in the emission electronic
transitions (see Figs. 6 and 8).

G. Barone et al. / Journal of Inorganic Biochemistry 104 (2010) 765–773
771
Fig. 9. Molecular orbitals involved in the electronic transitions calculated for the absorption spectra of the ZnL2+·H2O S0 and of the dioxidized ZnL4+·H2O T0 (top rows), and for the
emission spectra of the corresponding simplified models, ZnL′·H2O S1 and ZnL′2+·H2O S1 excited states (bottom rows).
According to this picture, the protective action of DNA toward the
intercalated ZnL2+ can be attributed to the inhibition of the ZnL2+
photooxidation through an effective stabilization of its ground S0
state (see Fig. 3). The occurrence of ZnL2+ photooxidation in bulk
water solution induces to hypothesize that an electron transfer takes
place from each of the salicylideneimine groups to a water molecule. It
is worth in fact recalling that the same photooxidation process also
occurs in deoxygenated solutions (see above). Hence, a possible
explanation of the way by which DNA hinders this process and
protects ZnL2+ from the photooxidation, can be given considering
that the DNA-intercalated ZnL2+ compound is in a more hydrophobic
region, less accessible to water molecules (see below). Then, in
agreement with the above reported findings, to investigate the
photophysical properties of intercalated ZnL2+, all the subsequent
fluorescence experiments were carried out using samples prepared
and stored in the dark before each measurement.
4.3. Steady-state and time-resolved photophysical properties of ZnL2+
intercalated in native DNA
Under excitation at 370 nm, ZnL2+ aqueous solutions show an
intense structured emission band centered at 483 nm, whose intensity
increases in the presence of DNA. The increase of the fluorescence
intensity is accompanied by a red-shift of the fluorescence maximum
(see Fig. 10). These findings are typical clues of the occurrence of an
intercalation process and in particular indicates that in the intercalated
state the complex probes either (i) an environment less polar than water
or (ii) a restriction of molecular motions or (iii) solvent relaxation that
deactivates the excited state without emission of a photon [47]. In
particular, by adding increasing amounts of DNA at fixed complex
Table 1
Attribution of the most intense calculated absorption and emission transitions of the
considered ZnII complexes (see Figs. 6b and 8b) to the frontier molecular orbitals
essentially involved (see Fig. 9).
Table 2
Table 2. Relevant geometrical parameters (Å and °) of the compounds ZnL2+·H2O S0 and
ZnL4+·H2O T0, optimized by the B3LYP/DZVP method, and of the compounds ZnL′·H2O S1
and ZnL′2+·H2O S1, optimized by the CIS/DZVP method (see Fig. 1 for atom labels).
Compound
λ (nm)
Attribution
ZnL2+·H2O-S0
297
371
415
388
393
480
398
HOMO-4→LUMO 61%
HOMO→LUMO+1 69%
HOMO→LUMO 89%
HOMO-6→LUMO+1 35%
HOMO-6→LUMO+1 31%
HOMO→LUMO 39%
HOMO-1 → LUMO 44%
HOMO→LUMO+1 36%
HOMO→LUMO 87%
HOMO-5→LUMO 89%
HOMO→LUMO+1 57%
ZnL2+·H2O-S0
ZnL4+·H2O-T0
ZnL′·H2O-S1
ZnL′2+·H2O-S1
Zn–O1
Zn–O2
Zn–N1
Zn–N2
Zn–OH2
O1–Zn–O2
N1–Zn–N2
O1–Zn–N1
N1–Zn–O2
1.994
2.001
2.117
2.122
2.178
100.2
78.0
2.043
2.042
2.113
2.112
2.103
94.5
79.3
88.2
156.2
1.985
1.988
2.124
2.113
2.230
105.2
78.2
1.985
1.989
2.153
2.155
2.143
102.6
75.7
ZnL4+·H2O-T0
ZnL′·H2O-S1
489
410
490
ZnL′2+·H2O-S1
89.0
161.2
88.2
166.0
87.0
156.8

772
G. Barone et al. / Journal of Inorganic Biochemistry 104 (2010) 765–773
Fig. 10. Fluorescence spectra of solutions of the ZnL2+ complex, stored in the dark, in the
presence of increasing amounts of DNA at fixed ZnL2+ concentration ([ZnL2+]=5×10−5 M),
R=0.0 ( ), 0.10 ( ), 0.20 ( ), 0.30 ( ), 0.40 ( ), 0.50 ( ), 0.80 ( ), 1.0 ( ), 2.0
Fig. 11. Wavelength at the band maximum as a function of R (horizontal lines indicate
the λmax value of ZnL2+ in the specified solvent medium).
(
), 3.0 ( ), 5.0 ( ), 8.0 ( ); 10.0 ( ) from the lowest to highest absorbance at
500 nm. In the inset the intensity (Imax) at the emission band maximum is reported as a
function of R (line is only a guide for the eye).
found that these spectra can be consistently described in terms of two
exponential decay functions where, according to literature, the more
fast decay characterized by a quite constant relaxation time of about
0.2 ns can be attributed to the sample scattering [48–50].
concentration, the emission intensity at the band maximum is
progressively enhanced reaching a plateau at R∼3 corresponding to a
binding size of about 0.7 per base pairs which is in fair agreement with
the literature value [22]. By plotting the fluorescence intensity at the
band maximum as a function of R (inset in Fig. 10), it can be noted that a
short induction regime (1) in the region 0bRb0.8 was characterized by a
nearly linear increase followed by a rapid increase (2) and finally a trend
toward a plateau (3). This peculiar behavior reveals the existence of
different binding regimes of ZnL2+ to the DNA double helix. It is worth to
note that during phase (1), although the fraction of intercalated ZnL2+ is
small, the DNA is completely saturated by ZnL2+ while the major part of
the complex is in the aqueous medium or surrounding the DNA
molecule. Accordingly, regime (1) could be attributed to the depletion
of ZnL2+-layers nearest the DNA surface [18]. Then, further increasing of
the DNA concentration in the region 0.8bRb3, together with an increase
of the fraction of DNA-intercalated zinc(II) complex, the concentration of
ZnL2+ outside tends to vanish. Finally, when the DNA concentration is
very high all the ZnL2+ molecules are totally secluded within the DNA
double helix. On the other hand, by considering the behavior of the
wavelength at the maximum of the emission band (see Fig. 11), we
observe a bathochromic shift in the 0bRb2 range followed by a
hypsochromic shift at RN2. This finding indicates that this parameter
monitors a different aspect of the intercalation process. Specifically, it is a
fine probe of the environment experienced by ZnL2+. The region where
the bathochromic shift occurs corresponds to an increase of the fraction
of the complex intercalated. On the other hand, the region of the
hypsochromic shift reveals that, during the dilution process of ZnL2+
among the DNA binding sites, some changes of the DNA interior,
involving an increase of its polarity, occur.
On the other hand, the second decay, attributable to the excited
ZnL2+ complex, is characterized by a relaxation time which changes
little going from the bulk aqueous phase (1.1 ns) to the DNA-
intercalated state (0.9 ns). This piece of experimental result can be
taken as an indication that the complex intercalation does not involve
significant restriction of its molecular motions which should lead to a
decrease of the nonradiative decay rate [51]. Moreover, the existence
of a single decay time for the intercalated complex points toward the
existence of nearly homogeneous binding sites.
5. Conclusions
The fluorescence intensity of the cationic ZnII-5-triethyl ammoni-
um methyl salicylidene ortho-phenylendiiminate complex, ZnL2+, in
water solution at neutral pH, dramatically increases in the range 360–
700 nm by exposure of the solution samples to tungsten light.
Surprisingly, experimental results indicate that, in the intercalated
state, ZnL2+ is effectively protected from this photochemical process.
Quantum chemical calculations allowed us to suggest that in the free
state ZnL2+ undergoes a photoinduced two-electron oxidation
process and, consequently, the protective action of DNA toward the
intercalated ZnL2+ can be attributed to an effective inhibition of the
In order to estimate the environment sensed by the intercalated
complex we have also collected its spectrum in methanol and ethanol.
The comparison with that in aqueous solution is shown in Fig. 12. It
can be noted that the wavelength at the band maximum in water
(482 nm, dielectric constant 78) shifts at about 488 nm in methanol
(dielectric constant 32.6) and at 510 nm in ethanol (dielectric
constant 24.3). Considering that the highest value is observed in the
presence of DNA (495 nm), it can be concluded that the intercalated
complex probes an environment whose polarity is comprised
between that of methanol and ethanol.
Further insights on the environment sensed by the intercalated
complex were searched by monitoring the time-resolved fluorescence
spectra of the complex in aqueous solution of native DNA. We have
Fig. 12. Fluorescence spectra of ZnL2+ in water (- - -), methanol ( ) and ethanol (
at fixed concentration ([ZnL2+]=5×10−5 M).
)

G. Barone et al. / Journal of Inorganic Biochemistry 104 (2010) 765–773
773
ZnL2+ photooxidation. The latter conclusion is in agreement with the
result that DNA-intercalated ZnL2+ is confined in a region less polar
than water and inaccessible to this solvent, possibly being water the
electron acceptor molecule of the ZnL2+ photooxidation.
[20] S.J. Wezenberg, E.C. Escudero-Adán, J. Benet-Buchholz, A.W. Kleij, Org. Lett. 10
(2008) 3311–3314.
[21] G. Barone, N. Gambino, A. Ruggirello, A. Silvestri, A. Terenzi, V. Turco Liveri, J.
Inorg. Biochem. 103 (2009) 731–737.
[22] A. Silvestri, G. Barone, G. Ruisi, D. Anselmo, S. Riela, V. Turco Liveri, J. Inorg.
Biochem. 101 (2007) 841–848.
Information achieved by steady state and time-resolved fluores-
cence spectroscopy indicates that the polarity of the environment
probed by ZnL2+, intercalated within DNA, is comprised between that
of methanol and ethanol and shows a continuous variation with the
DNA to complex molar ratio. Indirectly, this finding allows to
hypothesize that the structure of the DNA double helix progressively
changes by increasing the fraction of occupied binding sites. From a
more general prospect, the capability of DNA molecules to protect
intercalated species from photodegradation processes seems to be of
utmost importance from a biological point of view and could be
amenable to biomedical applications.
[23] G. Barone, A. Longo, A. Ruggirello, A. Silvestri, A. Terenzi, V. Turco Liveri, Dalton
Trans. (2008) 4172–4178.
[24] S.J. Angyal, P.J. Morris, J.R. Tetaz, J.G. Wilson, J. Chem. Soc. (1950) 2141–2145.
[25] P. McPhie, Methods Enzymol. 22 (1971) 23–32.
[26] S.D. Kennedy, R.G. Bryant, Biophys. J. 50 (1986) 669–676.
[27] W.-K. Dong, Y.-X. Sun, Y.-P. Zhang, L. Li, X.-N. He, X.-L. Tang, Inorg. Chim. Acta 362
(2009) 117–124.
[28] A.D. Becke, J. Chem. Phys. 98 (1993) 5648–5652.
[29] N. Godbout, D.R. Salahub, J. Andzelm, E. Wimmer, Can. J. Chem. 70 (1992)
560–571.
[30] C. Sosa, J. Andzelm, B.C. Elkin, E. Wimmer, K.D. Dobbs, D.A. Dixon, J. Phys. Chem. 96
(1992) 6630–6636.
[31] E. Runge, E.K.U. Gross, Phys. Rev. Lett. 52 (1984) 997–1000.
[32] R.E. Stratmann, G.E. Scuseria, M.J. Frisch, J. Chem. Phys. 109 (1998) 8218–8224.
[33] R. Bauernschmitt, R. Ahlrichs, Chem. Phys. Lett. 256 (1996) 454–464.
[34] M.E. Casida, C. Jamorski, K.C. Casida, D.R. Salahub, J. Chem. Phys. 108 (1998)
4439–4449.
Acknowledgements
[35] J.B. Foresman, M. Head-Gordon, J.A. Pople, M.J. Frisch, J. Phys. Chem. 96 (1992)
135–149.
[36] R.J. Cave, K. Burke, E.W. Castner Jr., J. Phys, Chem. A 106 (2002) 9294–9305.
[37] D. Jacquemin, E.A. Perpète, G. Scalmani, M.J. Frisch, I. Ciofini, C. Adamo, Chem.
Phys. Lett. 421 (2006) 272–276.
Financial support from the Università di Palermo, Cariplo, and MIUR
within the National Research Project “Dalle singole molecole a
complessi e nanostrutture: struttura, chiralità, reattività e teoria”
(PRIN 2006) is gratefully acknowledged. We thank also the “Fondazione
Banco di Sicilia” (Palermo, Italy) for the co-funding of the Fluoromax 4
(Jobin Yvon) spectrofluorometer (Convenzione PR 19.b/06).
[38] D. Jacquemin, E.A. Perpète, X. Assfeld, G. Scalmani, M.J. Frisch, C. Adamo, Chem.
Phys. Lett. 438 (2007) 208–212.
[39] V. Barone, M. Cossi, J. Phys, Chem. A 102 (1998) 1995–2001.
[40] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman, J.A.
Montgomery Jr., T. Vreven, K.N. Kudin, J.C. Burant, J.M. Millam, S.S. Iyengar, J.
Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G.A. Petersson, H.
Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T.
Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J.E. Knox, H.P. Hratchian, J.B.
Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R.E. Stratmann, O. Yazyev, A.J.
Austin, R. Cammi, C. Pomelli, J.W. Ochterski, P.Y. Ayala, K. Morokuma, G.A. Voth, P.
Salvador, J.J. Dannenberg, V.G. Zakrzewski, S. Dapprich, A.D. Daniels, M.C. Strain, O.
Farkas, D.K. Malick, A.D. Rabuck, K. Raghavachari, J.B. Foresman, J.V. Ortiz, Q. Cui, A.G.
Baboul, S. Clifford, J. Cioslowski, B.B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I.
Komaromi, R.L. Martin, D.J. Fox, T. Keith, M.A. Al-Laham, C.Y. Peng, A. Nanayakkara,
M. Challacombe, P.M.W. Gill, B. Johnson, W. Chen, M.W. Wong, C. Gonzalez, J.A.
Pople, Gaussian 03, Revision D.02, Gaussian, Inc., Wallingford CT, 2005.
[41] N.M. O'Boyle, A.L. Tenderholt, K.M. Langner, J. Comput. Chem. 29 (2008) 839–845,
http://gausssum.sourceforge.net.
References
[1] A. Rajendran, C.J. Magesh, P.T. Perumal, Biochim. Biophys. Acta 1780 (2008)
282–288.
[2] A. Arola-Arnal, J. Benet-Buchholz, S. Neidle, R. Vilar, Inorg. Chem. 47 (2008) (1919)
11910–11911.
[3] B. Peng, W.-H. Zhou, L. Yan, H.-W. Liu, L. Zhu, Trans. Met. Chem. 34 (2009) 231–237.
[4] K. Takahashi, K. Fukiura, H. Arii, M. Chikira, Nucl. Acids Symp. Ser. 51(2007)189–190.
[5] G.-D. Liu, J.-P. Liao, S.-S. Huang, G.-L. Shen, R.-Q. Yu, Anal. Sci. 17 (2001) 1031–1036.
[6] G.-D. Liu, X. Yang, Z.-P. Chen, G.-L. Shen, R.-Q. Yu, Anal. Sci. 16 (2000) 1255–1259.
[7] D.J. Gravert, J.H. Griffin, Inorg. Chem. 35 (1996) 4837–4847.
[8] K.I. Ansari, J.D. Grant, G.A. Woldemariam, S. Kasiri, S.S. Mandal, Org. Biomol. Chem.
7 (2009) 926–932.
[9] G.A. Woldemariam, S.S. Mandal, J. Inorg. Biochem. 102 (2008) 740–747.
[10] S. Bhattacharya, S.S. Mandal, J. Chem. Soc. Chem. Comm. (1995) 2489–2490.
[11] J.G. Muller, L.A. Kayser, S.J. Paikoff, V. Duarte, N. Tang, R.J. Perez, S.E. Rokita, C.J.
Burrows, Coord. Chem. Rev. 185–186 (1999) 761–774.
[42] N.J. Turro, Modern Molecular Photochemistry, Benjamin-Cummings, Menlo Park,
CA, 1978.
[43] H. Kunkely, A. Vogler, J. Photochem, Photobiol. A. Chem. 138 (2001) 51–54.
[44] H. Kunkely, A. Vogler, Inorg. Chem. Commun. 4 (2001) 692–694.
[45] T. Fukuda, F. Sakamoto, M. Sato, Y. Nakano, X.S. Tana, Y. Fujii, Chem. Commun.
(1998) 1391–1392.
[12] S. Routier, J.-L. Bernier, M.J. Waring, P. Colson, C. Houssier, C. Bailly, J. Org. Chem.
61 (1996) 2326–2331.
[46] W. Saenger, Principles of Nucleic Acid Structure, Springer Verlag, New York, 1984.
[47] M. Raguz, J. Brnjas-Kraljevic, J. Chem. Inf. Model. 45 (2005) 1636–1640.
[48] C.D. Byrne, A.J. de Mello, Biophys. Chem. 70 (1998) 173–184.
[49] E.M. Talavera, P. Guerrero, F. Ocana, J.M. Alvarez-Pez, Appl. Spectrosc. 56 (2002)
80–87.
[50] R.A. Hochstrasser, D.P. Millar, Proceedings of SPIE, Int. Soc. Opt. Eng. 1640 (1992)
599–605.
[51] F. Ito, T. Kakiuchi, T. Nagamura, J. Phys, Chem. C. 111 (2007) 6983–6988.
[13] K. Sato, M. Chikira, Y. Fujii, A. Komatsu, J. Chem. Soc. Chem. Comm. (1994) 625–626.
[14] G.B. Roy, Inorg. Chim. Acta 362 (2009) 1709–1714.
[15] T. Kawasaki, T. Kamata, H. Ushijima, M. Kanakubo, S. Murata, F. Mizukami, Y. Fujii,
Y. Usui, J. Chem. Soc. Perkin Trans. 2 (1999) 193–198.
[16] H. Kunkely, A. Vogler, Inorg. Chim. Acta 321 (2001) 171–174.
[17] M.E. Germain, M.J. Knapp, Inorg. Chem. 47 (2008) 9748–9750.
[18] K.E. Splan, A.M. Massari, G.A. Morris, S.-S. Sun, E. Reina, S.B.T. Nguyen, J.T. Hupp,
Eur. J. Inorg. Chem. (2003) 2348–2351.
[19] M.E. Germain, T.R. Vargo, P.G. Khalifah, M.J. Knapp, Inorg. Chem. 46 (2007)
4422–4429.