New luminescent terbium complex for the determination of DNA

Article abstract of DOI:10.1016/j.saa.2004.03.019

New terbium complexes of derivatives of 2-oxo-4-hydroxy-quinoline-3- carboxylic acid are reported, which are highly luminescent, water soluble and do not require luminescence enhancers. The triplet-state energy levels of the ligands, the relative quantum yields (QYs) and the excitation maxima of the respective terbium chelates were determined. The large luminescence enhancement of one of these complexes by nucleic acids was investigated and a mechanism of its interaction with DNA is proposed. The optimal conditions for determination of DNA are equal concentrations of Tb³⁺ and ligand R₁ (C = 1 × 10^-6 M), pH 9.0. Under optimal conditions the luminescence intensity (RI) is proportional to the concentration of fish sperm DNA (fsDNA) or calf thymus DNA (ctDNA), respectively, within the range of 0.05-1.5 μg ml^-1. The detection limits were 10ng ml^-1 for fsDNA and 12 ng mr^-1 for ctDNA.

Full text of DOI:10.1016/j.saa.2004.03.019

Spectrochimica Acta Part A 61 (2005) 109–116
New luminescent terbium complex for the determination of DNA
Alla Yegorovaa, Alexander Karasyovb, Axel Duerkopb,∗
Igor Ukrainetsc, Valery Antonovicha
,
a
AV Bogatsky Physico-Chemical Institute of the National Academy of Sciences of Ukraine, 86, Lustdorfskaya doroga, Odessa 65080, Ukraine
b
Institute of Analytical Chemistry, Chemo- and Biosensors, University of Regensburg, Universitätsstraße 31, 93040, Regensburg, Germany
c
National University of Pharmacy, Blukher Street 4, Kharkov 61002, Ukraine
Received 19 December 2003; accepted 25 March 2004
Abstract
New terbium complexes of derivatives of 2-oxo-4-hydroxy-quinoline-3-carboxylic acid are reported, which are highly luminescent, water
soluble and do not require luminescence enhancers. The triplet-state energy levels of the ligands, the relative quantum yields (QYs) and
the excitation maxima of the respective terbium chelates were determined. The large luminescence enhancement of one of these complexes
by nucleic acids was investigated and a mechanism of its interaction with DNA is proposed. The optimal conditions for determination of
DNA are equal concentrations of Tb3+ and ligand R1 (C = 1 × 10−6 M), pH 9.0. Under optimal conditions the luminescence intensity (RI) is
proportional to the concentration of fish sperm DNA (fsDNA) or calf thymus DNA (ctDNA), respectively, within the range of 0.05–1.5 ␮g ml−1
The detection limits were 10 ng ml−1 for fsDNA and 12 ng ml−1 for ctDNA.
.
© 2004 Elsevier B.V. All rights reserved.
Keywords: Terbium complex; Nucleic acids; Luminescence; Intercalator
1. Introduction
bands, a large Stokes’ shift and long luminescence decay
times, which can be exploited to reduce interferences caused
by a biological background [18]. General requirements for
luminescent lanthanide chelates to be used as probes for
bioassays include a high luminescence quantum yield, high
kinetic stability and good water solubility.
We present here nine new terbium complexes of deriva-
tives of 2-oxo-4-hydroxy-quinoline-3-carboxylic acid which
are highly luminescent, have high stability and solubility
in water and do not require the addition of luminescence
enhancers such as micelles. These are the first terbium
complexes based on quinolones for the luminescent de-
termination of DNA. The complex showing the strongest
luminescence enhancement upon interaction with nucleic
acid was applied for a sensitive luminescent assay for the
determination of DNA.
Fluorescent probes including organic dyes [1–4], metal
ions [5–7] and metal complexes [8,9] are frequently em-
ployed to investigate nucleic acids. In recent years, the
use of coordination complexes of lanthanide ions as a
probe to study nucleic acids has attracted much attention.
Phenanthroline [10], diethylenetriaminepentaacetic acid
and p-aminosalicylic acid [11], 1,6-bi(1ꢀ-phenyl-3ꢀ-methyl-
5ꢀ-pyrazolone-4ꢀ-hexane-dione(BPMPHD)-cetyltrimethyl-
ammonium bromide (CTMAB) [12], tetracycline [13],
oxytetracycline [14] and tiron [15] are widely used as
chelators.
Lanthanide complexes are used as probes and labels for
direct determination of organic analytes, nucleic acids and
in immunodiagnostic assays [16]. Their luminescence is the
result of an efficient intramolecular energy transfer from the
excited triplet-state of the ligand to the emitting level of
the lanthanide [17]. These complexes show narrow emission
2. Experimental
2.1. Materials
∗
Corresponding author. Tel.: +49-941-943-4053;
fax: +49-941-943-4064.
Stock solutions of nucleic acids (0.1 mg ml−1) were pre-
pared by dissolving commercial fish sperm DNA (fsDNA)
E-mail address: axel.duerkop@chemie.uni-regensburg.de
(A. Duerkop).
1386-1425/$ – see front matter © 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.saa.2004.03.019

110
A. Yegorova et al. / Spectrochimica Acta Part A 61 (2005) 109–116
and calf thymus DNA (ctDNA) (Sigma, Steinheim, Ger-
many) in 0.05 mM sodium chloride. Stock standard solu-
tions (0.01 M) of terbium and aluminium chlorides were
prepared by dissolving the corresponding terbium oxide
or aluminium oxide (Sigma, 99.99%), respectively, in hy-
drochloric acid. Excess of HCl was evaporated to a wet
residue and diluted with water. The metal concentration
was determined by complexometric titration. Derivatives of
2-oxo-4-hydroxy-quinoline-3-carboxylic acid (R1–R9, see
Table 1) were synthesized as described earlier [19].
Stein [20] with quinine sulphate (Φ = 0.546 in 1 M H2SO4
[21]) as quantum yield standard.
3. Results and discussion
3.1. Choice of ligands
It is common knowledge that ligands containing
␤-diketo groups efficiently coordinate to lanthanide ions
since NMR-shift reagents like Eu(di-pivaloyl-methanate)3
were introduced [22]. Quinolone based antibiotics like
ciprofloxacine, norfloxacine, nalididxic acid and many oth-
ers were often determined as luminescent ␤-diketo com-
plexes with lanthanides. These quinolone derivatives were
used alternatively for the determination of lanthanide ions
[23]. To our best knowledge, we use quinolones for the de-
termination of DNA, for the first time. For the luminescent
detection of DNA we introduced a side chain containing a
quaternary ammonium group to each ligand. This is needed
for two reasons. One is to enable solubility of the complexes
in aqueous buffers which are generally used in bioanalytics.
The second is, that the luminescence enhancement of or-
ganic dyes like TOTO or TO-PRO and others was attributed
to intercalation of the quaternary ammonium group into the
grooves of DNA strands. We intended to use this group to
attach the complex via the quinolone ligand to the DNA.
The ligand structure was varied using different ring
structures 4-hydroxy-2-oxo-1,2-dihydro-3-quinolinecarbox-
amide, 6-hydroxy-4-oxo-1,2-dihydro-4H-pyrrolo[3,2,1-ij]
quinoline-5-carboxamide and 7-hydroxy-5-oxo-2,3-dihydro-
1H,5H-pyrido[3,2,1-ij]quinoline-6-carboxamide. Further-
more, the side chain attached to the 4-hydroxy-2-oxo-1,2-
dihydro-3-quinolinecarboxamide part of the ligand was
modified using ethyl, propyl and allyl chains. These ring
and side chain parts of the ligand bind to the outer side of
DNA whereas the ammonium group intercalates into the
groove of the DNA strand. We intended to find a ligand
ring structure which should be rigidized when the complex
was bound to DNA. This improves the energy transfer from
the ligand to the lanthanide ion and is accompanied by an
increased quantum yield. Finally, the structure of the qua-
ternary ammonium group was modified. This can determine
how deep the ligand intercalates into the DNA groove. This
influences the rigidization of the whole complex in bound
state, again, and is therefore another means to control the
quantum yield.
Stock solutions (1 × 10−3 M) of these ligands were pre-
pared by dissolving of accurately weighed preparations in
water. Working solutions were prepared by appropriate dilu-
tion with water. A 0.1 M Tris–HCl buffer solution was pre-
pared by dissolving 1.211 g of Tris base (USB, Cleveland,
Ohio, USA) in 90 ml of water, adjusting the pH to 9.0 with
HCl and making up the volume to 100 ml with water.
All chemicals used were of analytical grade. Doubly dis-
tilled water was used throughout.
2.2. Apparatus
Luminescence excitation and emission spectra were
recorded on an Aminco-Bowman Series 2 luminescence
spectrometer from SLM (Rochester, NY, USA). The spec-
trometer was equipped with a 150 W CW xenon lamp as
excitation light source. The luminescence measurements
described in Section 3.5. were obtained on an SDL-2 spec-
trofluorimeter (Leningrad Optomechanical Association,
St. Petersburg, Russia). UV Spectra were recorded on a
Cary 50 Bio spectrophotometer from Varian (Australia)
and a Lambda-9 from Perkin–Elmer. The pH values were
measured using an OP-211/1 laboratory digital pH-meter
(Radelkis, Budapest, Hungary).
2.3. Procedures
To a 10 ml test tube, solutions of appropriate concentra-
tions were added in the following order (in the absence and
presence of nucleic acids): Tb3+ solution, Rx (ligand) solu-
tion, DNA solution and Tris–HCl buffer solution (pH 9.0).
The mixture was diluted to 10 ml with water, stirred and al-
lowed to stand for 10 min. The luminescence intensity (RI)
was measured in a 1 cm quartz cell with an excitation wave-
length of λex = 340 nm and an emission wavelength of
λem = 545 nm. All measurements were performed at room
temperature. The enhanced relative luminescence intensity
of Tb3+–R1 was calculated as RI = RIꢀ–RI0, where RIꢀ and
RI0 are the relative luminescence intensities of the system
with and without nucleic acid, respectively.
3.2. Spectral characteristics
The triplet levels of the ligands were calculated from
phosphorescence spectra of yttrium complexes of the lig-
ands at 77 K. Quinine sulphate (Fluka, Buchs, Switzerland)
was dissolved in 1 M H2SO4 to obtain a solution for quan-
tum yield measurements. The luminescence quantum yields
(QYs) were obtained by the method described by Haas and
The absorption spectra of the ligands in aqueous solutions
are characterized by the presence of two bands in the UV
(Table 2). The high molar extinction coefficients of these
bands enable effective absorption of excitation light energy.
The triplet energy levels of the ligands are between 20800
and 22250 cm−1. The ligands can transfer energy to the

A. Yegorova et al. / Spectrochimica Acta Part A 61 (2005) 109–116
111
Table 1
Structural formulae and names of the ligands investigated
N
Structure and name
R1
N-[2-(diethylamino)ethyl]-4-hydroxy-2-oxo-1-propyl-1,2-dihydro-3-quinolinecarboxamide
R2
1-allyl-N-[2-(diethylamino)-ethyl]-4-hydroxy-2-oxo-1,2-dihydro-3-quinolinecarboxamide
R3
N-[2-(diethylamino)ethyl]-1-ethyl-6,7-difluoro-4-hydroxy-2-oxo-1,2-dihydro-3-
quinolinecarboxamide
R4
N-[2-(diethylamino)ethyl]-6-hydroxy-4-oxo-1,2-dihydro-4H-pyrrolo[3,2,1-ij]quinoline-
5-carboxamide
R5
N-[2-(diethylamino)ethyl]-7-hydroxy-5-oxo-2,3-dihydro-1H,5H-pyrido[3,2,1-ij]
quinoline-6-carboxamide
R6
N-[2-(ethylamino)ethyl]-4-hydroxy-2-oxo-1-propyl-1,2-dihydro-3-
quinolinecarboxamide
R7
7-hydroxy-N-[2-(hydroxyamino)ethyl]-5-oxo-2,3-dihydro-1H,5H-pyrido[3,2,1-ij]
quinoline-6-carboxamide
R8
4-hydroxy-2-oxo-N-[2-(1-piperazinyl)ethyl]-1-propyl-1,2-dihydro-3-
quinolinecarboxamide

112
A. Yegorova et al. / Spectrochimica Acta Part A 61 (2005) 109–116
Table 1 (Continued )
N
Structure and name
9-fluoro-7-hydroxy-3-methyl-5-oxo-N-[2-(1-piperazinyl)ethyl]-2,3-dihydro-1H,
5H-pyrido[3,2,1-ij]quinoline-6-carboxamide
R9
centre ion only, if their triplet energy level is higher than
that of the lanthanide ion. All ligands fulfil this requirement
because the triplet energy level of Tb3+ (5D4) is located at
ronment of the respective complex. Therefore, the changes
of the luminescence intensity of this band are most often
used for analytical applications with Tb3+ complexes.
The quantum yields of complexes Tb3+ with R1–R9
were measured in Tris–HCl buffer (pH 9.0) relative
20500 cm−1
.
It is obvious from Table 2, that the differences of
triplet-state energies of the ligands are too small to be at-
tributed to changes of their structure. This is also true for the
variations of the molar absorbances and absorption maxima.
The excitation and emission spectra of Tb3+ complexes
R1–R9 were recorded. Fig. 1 (left) shows the excitation spec-
trum of the Tb3+–R1 complex monitored at 545 nm. The
excitation maximum is at 309 nm which is similar to the
other complexes (see Table 2). The emission bands of the
Tb3+–R1 complex (Fig. 1, right) are located at 490, 545, 585
to quinine sulphate at λex
=
340 nm. The quantum
yields (QYs) of the Tb3+ chelates with R1–R9 are
given in Table 2 and vary from 0.23 to 0.40. From
the structures displayed in Table 1, it is obvious that
the 7-hydroxy-5-oxo-2,3-dihydro-1H,5H-pyrido[3,2,1-ij]
quinoline-6-carboxamides R5, R7 and R9 display high
QYs. This is independent from the different groups at-
tached to the ammonium moiety of these ligands. The
4-hydroxy-2-oxo-1,2-dihydro-3-quinolinecarboxamide str-
uctures R1 and R8 can also show high quantum yields
upon complexation with Tb3+. But here, a propyl side
chain seems to be a prerequisite for high QYs, because the
complexes without this chain (R2 and R3) display signif-
icantly lower quantum yields. On comparing the QYs of
R1 and R8 it is obvious that the structure of the different
groups attached to the ammonium moiety has only minor
influence on the energy transfer from ligand to the centre
ion. The 6-hydroxy-4-oxo-1,2-dihydro-4H-pyrrolo[3,2,1-ij]
quinoline-5-carboxamide structure has the lowest QY and
seems not very suitable for luminescent determination of
DNA. As can be seen from Table 2, Tb3+–R1 has the highest
quantum yield. Therefore, this complex was the preferred
candidate for the luminescent determination of DNA.
5
7
5
7
and 620 nm generated from the D4 → F6, D4 → F5,
5D4 → F4 and D4 → F3 transitions of Tb3+, respec-
tively. The maxima of the emission bands of all nine com-
plexes do not vary in wavelength but in intensity. Especially
the 545 nm bands are hypersensitive to the molecular envi-
7
5
7
Table 2
Triplet-state energy levels (E), absorption maxima (λabs) with molar ab-
sorbances (ꢀ) of ligands R1–R9; quantum yields (Φ) and excitation max-
ima (λex), of the Tb complexes with these ligands
Ligand
1
E (cm−1
22100
)
λabs (nm)
ε (l M−1 cm−1
)
Φ
λex (nm)
290
238
25600
73300
0.40
309
2
3
4
5
6
7
8
9
22250
22200
21050
21740
22200
21500
21740
20830
290
238
25800
69100
0.32
0.25
0.23
0.38
0.24
0.37
0.36
0.39
322
310
312
309
316
317
315
319
296
231
20500
64800
292
246
23400
53800
292
237
26000
64000
290
238
29300
87800
292
237
31800
79300
291
238
18100
52400
Fig. 1. Excitation and emission spectra of Tb3+–R1 complex
(CTb3+ = 1 × 10−6 M; CR1 = 2 × 10−6 M; pH 9.0; λex = 340 nm and
λem = 545 nm).
292
236
26600
63300

A. Yegorova et al. / Spectrochimica Acta Part A 61 (2005) 109–116
113
that the maximum luminescence intensity of the system is
reached at pH 9.0. This is in accordance with the fact that
the ligand will coordinate with Tb3+ more efficiently in its
enolic form. Therefore, we choose pH 9.0 (0.1 M Tris–HCl
buffer) for the further experiments.
Tb3+ forms complexes with R1–R9 in a 1:2 ratio. This
was determined by the molar ratio method. The influence
of the Tb3+ concentration on the luminescence intensity
was investigated in the concentration range of 5 × 10−7 M
to 5 × 10−6 M of Tb3+ with 0.50 ␮g ml−1 of fsDNA and
1 × 10−6 M of R1 in Tris–HCl buffer (pH 9.0). The influ-
ence of R1 concentration on the luminescence intensity was
investigated at constant Tb3+concentration of 1 × 10−6 M.
The optimal conditions were equal concentrations of Tb3+
and R1 (C = 1 × 10−6 M) which were chosen for further
experiments.
It is well known that the addition of solvents, vari-
ous surfactants and synergetic agents (trioctylphosphinox-
ide/TOPO) are the prerequisites for lanthanide luminescence
to be enhanced [12]. The influence of various solvents was
investigated. The maximum luminescence intensity was
observed in Tris-buffered solution. Luminescence quench-
ing was observed in 50% (v/v) mixtures of water with
various solvents in the following order: methanol, ethanol,
dimethylformamide, dimethylsulfoxide (20–25% lumines-
cence decrease for each mixture); acetone, iso-propanol
(50% luminescence decrease for each mixture). It was found
as well that the effect of cationic, anionic and non-ionic
surfactants, respectively, and a synergetic agent (TOPO)
on the luminescence intensity is insignificant. This is a
major advantage compared to all lanthanide based methods
mentioned in Table 3 because the addition of the enhancer
always elongates the time for the determination of the
sample and causes higher costs for the additional reagent.
All effects mentioned above are typical for Tb3+–R1 and
Tb3+–R1–DNA complexes.
Fig. 2. Excitation spectrum of Tb3+–R1 in absence and presence of DNA.
CTb3+ = 1 × 10−6 M; CR1 = 1 × 10−6 M; CfsDNA = 0.5 ␮g ml−1; pH 9.0;
λem = 545 nm (1: Tb3+–R1; 2: Tb3+–R1–fsDNA).
The luminescence of all Tb3+-complexes in absence and
in presence of 0.5 ␮g ml−1 fsDNA was investigated. This
was done to determine the dynamic range of the lumi-
nescence enhancement upon binding to DNA. Here, the
Tb3+–R1 complex displays the most pronounced lumines-
cence enhancement compared to the other complexes. There-
fore Tb3+–R1 complex was used for further experiments
with DNA. The excitation maximum of the Tb3+–R1–DNA
complex is increased and red shifted by 10 nm compared
to Tb3+–R1 complex (Fig. 2). This indicates the interac-
tion between Tb3+–R1 complex and nucleic acid. The lumi-
nescence spectra of Tb3+–R1–DNA complex are similar to
those of Tb3+–R1, but the luminescence intensity is greatly
enhanced upon addition of nucleic acid (see Section 3.4).
3.3. Effect of pH, stoichiometry and buffer composition
The luminescence intensity of Tb3+–R1–fsDNA is
strongly dependent of pH (Fig. 3). It can be seen from Fig. 3
Table 3
Common luminescence probes for nucleic acid determination
Luminescence probe
Nucleic acid
LOD
(ng ml−1
Reference
)
Ethidium bromide
Hoechst 33258
Methylene blue
Vitamin K3
nDNA
nDNA
nDNA
nDNA, RNA
RNA
dDNA, RNA
nDNA
ctDNA, fsRNA
ctDNA, fsRNA
nDNA
nDNA
dsDNA
ssDNA
RNA
ctDNA
fsDNA
10
10
28
10, 26
100
100
10
76, 68
24, 13
9
11
0,25
0,1
1
10
[1]
[2]
[3]
[4]
[5,6]
[10]
[13]
[8]
Tb3+
Tb-1,10-phenanthroline
Eu-tetracycline
La-8-hydroxyquinoline
Al-8-hydroxyquinoline
Tb-BPMPHD-CTMAB
Eu-oxytetracycline
PicoGreen
[9]
[12]
[14]
[27]
[27]
[27]
OliGreen
RiboGreen
Tb3+–R1
Fig. 3. Dependence of RI of Tb3+–R1–fsDNA complex of the pH
(CTb3+ = 1 × 10−6 M; CR1 = 1 × 10−6 M; CfsDNA = 0.5 ␮g ml−1).
Tb3+–R1
12

114
A. Yegorova et al. / Spectrochimica Acta Part A 61 (2005) 109–116
Fig. 5. Calibration graph for the determination of fsDNA
(CTb3+ = 1 × 10−6 M; CR1 = 1 × 10−6 M; pH 9.0; λem = 545 nm).
Fig. 4. Luminescence enhancement of Tb3+–R1 complex upon addition of
various concentrations of fsDNA (CTb3+ = 1×10−6 M; CR1 = 1×10−6 M;
pH 9.0; λex = 340 nm and λem = 545 nm); CfsDNA, ␮g ml−1 (curve
number): 0 (1); 0.05 (2); 0.1 (3); 0.4 (4); 0.5 (5); 0.7 (6); 0.8 (7); 0.9
(8); 1.0 (9); 1.5 (10).
not disturbed by fluorescence of a short-lived (up to 20 ns)
biological background. The longer lifetime of terbium com-
plexes compared to europium complexes enables sensitive
detection even with very simple equipment for measurement
of luminescence decay times. This will be the subject of
further research to improve the LODs.
We use an excitation wavelength of 340 nm which is not
in the UV compared to the 267 and 265 nm of the two com-
plexes employing the luminescence of 8-hydroxyquinoline
[8,9]. The detection wavelength of 545 nm of our method is
also more longwave than the 485 nm of others [8,9]. This
improves the signal-to-noise ratio of the detection because
in bioanalysis background fluorescence is significantly de-
creased at longwave excitation and detection wavelengths.
3.4. Luminescence determination of DNA
The luminescence enhancement of the Tb3+–R1
complex was studied within a concentration range of
0.05–1.5 ␮g ml−1 of both, fsDNA and ctDNA. The emis-
sion spectra of the corresponding luminescence titration of
Tb3+–R1 with fsDNA are shown in Fig. 4. A strong, up
to eight-fold enhancement of the luminescence intensity is
observed.
Linear calibration plots were established from the spec-
tra for both, fsDNA and ctDNA. The regression data and
LODs (3σ/slope) are given in Table 4. A good agreement
between data and calibration plot can be seen from Fig. 5
for Tb3+–R1 with fsDNA. The calibration plots are almost
identical for fsDNA and ctDNA. Therefore, only the data for
the calibration with fsDNA is shown in Fig. 5. The limit of
detection was 0.010 ␮g ml−1 for fsDNA and 0.012 ␮g ml−1
for ctDNA.
3.5. The interaction mechanism of Tb3+–R1 with nucleic
acid
From the luminescence data presented in Fig. 5, binding
constants were calculated [24] to be (log k = 6.9 0.5).
This value is typical for intercalating organic dyes. To con-
firm the assumption of intercalation, we used the ethid-
ium bromide displacement assay. However, the overlap of
A comparison of the sensitivities of Tb3+–R1 and other
common probes for nucleic acid detection is given in Table 3.
Tb3+–R1 displays comparable sensitivity to ethidium bro-
mide (EB), however it is not cancerogenic. Furthermore, the
large Stokes’ shift of Tb-complexes facilitates the separation
of the excitation and emission light which is particularly im-
portant in filter-based instruments like microtiterplate read-
ers. The long luminescence decay times of Tb3+ complexes
offer gated detection as a further detection method which is
7
the emission band of Eu(III) (5D0 → F1, λ = 590 nm),
which was present in terbium oxide in amounts of about
1 × 10−3%, with the emission band of the ethidium bro-
mide (λ = 595 nm) prevented to choose Tb3+–R1 for the
displacement assay. Fortunately, the intercalation character-
istics of aluminium complexes are close to those of terbium.
Therefore, it is possible to infer from data of the complex
of R1 with Al3+ to similar properties of Tb3+–R1. This ap-
proach was used in an experiment of the ethidium bromide
displacement with the non-luminescent Al3+–R1 complex.
In case the investigated complex is an intercalator, it is able
to displace ethidium bromide from its complex with DNA.
The luminescence intensity of the whole system will then
be proportional to the amount of ethidium bromide which
is bound to DNA. Therefore, the luminescence intensity is
Table 4
Analytical parameters for DNA determination
DNA
Linear range
(␮g ml−1
Correlation
coefficient
Detection limit
(␮g ml−1
)
)
fsDNA
ctDNA
0.05–1.50
0.05–1.50
0.991
0.989
0.010
0.012

A. Yegorova et al. / Spectrochimica Acta Part A 61 (2005) 109–116
115
ligand structure was optimized to obtain high quantum yields
and a large signal enhancement upon binding to DNA. The
complexes are water soluble, not cancerogenic and display
favourably long Stokes’ shifts and high quantum yields.
Unlike other methods employing lanthanide complexes for
DNA determination, the addition of surfactants or lumi-
nescence enhancers is not required. The long luminescence
decay times allow additional detection schemes like gated
measurements in microplate readers to enhance sensitivity
by off-gating of short-lived background luminescence. Com-
plex Tb3+–R1 was chosen for the luminescent determina-
tion of DNA. An up to 10-fold luminescence enhancement
of this terbium complex upon titration with DNA was ob-
served. The optimal conditions are equal concentrations of
Tb3+ and ligand R1 (C = 1×10−6 M) at pH 9.0. Under these
conditions, the luminescence intensity is linearly dependent
on the concentration of fsDNA and ctDNA within the range
of 0.05–1.5 ␮g ml−1. The detection limits are 10 ng ml−1 for
fsDNA and 12 ng ml−1 for ctDNA, respectively. The binding
constant of Tb3+–R1 to DNA was determined to be log k =
6.1 by both, luminescence titration and ethidium bromide
displacement assay.
Fig. 6. Ethidium bromide displacement (%) by Al3+–R1 complex in
semi-logarithmic representation of C(Al3+–R1).
reduced on addition of Al3+–R1 to the solution containing
EB and DNA.
The C50 value is defined as the concentration of the com-
plex reducing the luminescence of the DNA-bond ethidium
bromide by 50%. It was determined from the value of the
abscissa inflection of the sigmoid curve in log CAl
of
3+–R1
the EB displacement percent coordinates (Fig. 6). Binding
constants were calculated from the C50 values using the for-
mula [25]
References
[1] J.B. Lepecq, C. Paoletti, Anal. Biochem. 17 (1966) 100–107.
[2] C. Labara, K. Paigen, Anal. Biochem. 102 (1980) 344–352.
[3] W.Y. Li, J.G. Xu, X.Q. Guo, Q.Z. Zhu, Y.B. Zhao, Anal. Lett. 30
(1997) 527–536.
[4] W.Y. Li, J.G. Xu, X.Q. Guo, Q.Z. Zhu, Y.B. Zhao, Anal. Lett. 30
(1997) 245–257.
[5] S. Sueda, T. Ihara, B. Juskowiak, M. Takagi, Anal. Chim. Acta 365
(1998) 27–31.
[6] J.H. Yang, C.L. Tong, N.Q. Jie, H.Y. Gong, Chin. Biochem. J. 12
(1996) 143–146.
[7] M.D. Topal, J.R. Fresco, Biochemistry 19 (1980) 5531–5537.
[8] C.Z. Huang, K.A. Li, S.Y. Tong, Anal. Lett. 29 (1996) 1705–
1717.
[9] C.Z. Huang, K.A. Li, S.Y. Tong, Anal. Lett. 30 (1997) 1305–
1319.
[10] Y.X. Ci, Y.Z. Li, W.B. Chang, Anal. Chim. Acta 248 (1991) 589–
594.
[11] S.S. Saavedra, E.G. Picozza, Analyst 114 (1989) 835–
841.
[12] X. Wu, J. Yang, F. Huang, M. Wang, L. Sun, G. Xu, Anal. Lett. 32
(1999) 2417–2425.
[13] Y.X. Ci, Y.Z. Lu, X.J. Liu, Anal. Chem. 67 (1995) 1785–
1788.
[14] R. Liu, J. Yang, X. Wu, J. Lumin. 96 (2002) 201–209.
[15] Y.-B. Zhao, D.-Y. Wang, X.-Q. Guo, J.-G. Xu, Anal. Chim. Acta
353 (1997) 329–335.
[16] K. Blomberg, P. Hurskainen, I. Hemilla, Clin. Chem. 45 (1999)
855–861.
[17] F.S. Richardson, Chem. Rev. 82 (1982) 541–552.
[18] P.R. Selvin, in: W. Rettig, B. Strehmel, S. Schrader, H. Seifert
(Eds.), Applied Luminescence in Chemistry, Biology and Medicine,
Springer, Berlin, 1998, p. 457.
[19] I.V. Ukrainets, D.N. Amin, I.V. Gorlacheva, O.V. Gorokhova, L.V.
Sidorenko, A.V. Turov, Chem. Heterocycl. Compd. 2 (2000) 207–
211.
ꢀ
ꢁ
CEB
C50(Al
KAl
=
× KEB;
3+–R1
3+–R1)
(KEB = 1 × 107 M−1).
The luminescence intensity of the peaks of EB (relative con-
trol) at 595 nm (excitation at 365 nm) in a series of solutions
containing DNA (CDNA = 1 ␮M), sodium azide (CNaN
=
3
0.3 mM), EB (CEB = 0.5 ␮M) and Al3+–R1 in various con-
centrations within a range 1×10−7 M to −1×10−5 M were
measured. A solution free of the investigated complex was
used as control. The obtained value of the binding constant
(log k = 6.1 0.4) is typical for mono-intercalators into
DNA [26] and is close to the value of the constant obtained
by a luminescence titration.
We suppose that side chain fragments of R1 interact
with the grooves of DNA due to the formation of hydrogen
bonds between non-divided electron pairs of nitrogen atoms
of DNA base pairs and the –N+(C2H5)2H group of the
complex. This interaction leads to significant ordering and
reduction of non-radiative energy losses and, as a result,
the luminescence intensity is increased upon binding with
DNA.
4. Conclusions
Nine quinolones were used as ligands for new terbium
complexes for the luminescent determination of DNA. The
[20] Y. Haas, G. Stein, J. Phys. Chem. 75 (1971) 3668–3677.

116
A. Yegorova et al. / Spectrochimica Acta Part A 61 (2005) 109–116
[21] S.R. Meech, D. Phillips, J. Photochem. 23 (1983) 193–198.
[22] C.C. Hinckley, J. Am. Chem. Soc. 91 (1969) 5160–5162.
[23] A. Gómes-Hens, M.P. Aguilar-Caballos, Trends Anal. Chem. 21
(2002) 131–141.
[25] I. Antonini, P. Polucci, L.R. Kelland, J. Med. Chem. 42 (1999)
2535–2541.
[26] H. Lyakhova, S. Lyakhov, L. Litvinova, Z. Topilova, I. Velcheva,
Khim.-Pharm. Zhurn. 37 (2003) 16–21.
[24] C.V. Kumar, E.H. Asuncion, J. Am. Chem. Soc. 115 (1993) 8547–
8551.
[27] R.P. Haugland (Ed.), Handbook of Fluorescent Probes and Research
Products, 9th ed., Molecular Probes, 2003, pp. 301–304.