Multidentate europium chelates as luminoionophores for anion recognition: Impact of ligand design on sensitivity and selectivity, and applicability to enzymatic assays

Article abstract of DOI:10.1002/chem.201304942

The design of photoluminescent molecular probes for the selective recognition of anions is a major challenge for the development of optical chemical sensors. The reversible binding of anions to lanthanide centers is one promising option for the realization of anion sensors, because it leads in some cases to a strong luminescence increase by the replacement of quenching water molecules. Yet, it is an open problem to gain control of the sensitivity and selectivity of the luminescence response. Primarily, the selective detection of (poly)phosphate species such as nucleotides has emerged as a demanding task, because they are involved in many biological processes and enzymatic reactions. We designed a series of pyridyl-based multidentate europium complexes (seven-, six-, and five-dentate) including sensitizing chromophores and studied their luminescence intensity and lifetime responses to different (poly)phosphates (adenosine triphosphate (ATP), adenosine diphosphate (ADP), adenosine monophosphate (AMP), cyclic adenosine monophosphate (cAMP), pyrophosphate, and phosphate anions), and carboxyanions (citrate, malate, oxalacetate, succinate, α-ketoglutarate, pyruvate, oxalate, carbonate). The results reveal that the number of free coordination sites has a significant impact on the sensitivity and selectivity of the response. Because of its reversibility, the lanthanide probes can be applied to monitor the activity of ATP-consuming enzymes such ATPases and apyrases, which is demonstrated by means of the five-dentate complex. Variation of binding sites: The number of binding sites of multidentate sensitizing ligands for Eu^III can be easily varied. The resulting complexes show luminescence responses to certain anions such as adenosine triphosphate (ATP) or adenosine diphosphate with different sensitivity and selectivity. Hence, a five-dentate complex was synthesized that undergoes a high and reversible response to ATP (see figure).

Full text of DOI:10.1002/chem.201304942

DOI: 10.1002/chem.201304942
Full Paper
&
Luminescent Probes
Multidentate Europium Chelates as Luminoionophores for Anion
Recognition: Impact of Ligand Design on Sensitivity and
Selectivity, and Applicability to Enzymatic Assays
Michael Schꢀferling,* Timo ꢁꢀritalo, and Tero Soukka[a]
Abstract: The design of photoluminescent molecular probes
for the selective recognition of anions is a major challenge
for the development of optical chemical sensors. The reversi-
ble binding of anions to lanthanide centers is one promising
option for the realization of anion sensors, because it leads
in some cases to a strong luminescence increase by the re-
placement of quenching water molecules. Yet, it is an open
problem to gain control of the sensitivity and selectivity of
the luminescence response. Primarily, the selective detection
of (poly)phosphate species such as nucleotides has emerged
as a demanding task, because they are involved in many
biological processes and enzymatic reactions. We designed
a series of pyridyl-based multidentate europium complexes
(seven-, six-, and five-dentate) including sensitizing chromo-
phores and studied their luminescence intensity and lifetime
responses to different (poly)phosphates (adenosine triphos-
phate (ATP), adenosine diphosphate (ADP), adenosine mono-
phosphate (AMP), cyclic adenosine monophosphate (cAMP),
pyrophosphate, and phosphate anions), and carboxyanions
(citrate, malate, oxalacetate, succinate, a-ketoglutarate, pyru-
vate, oxalate, carbonate). The results reveal that the number
of free coordination sites has a significant impact on the
sensitivity and selectivity of the response. Because of its re-
versibility, the lanthanide probes can be applied to monitor
the activity of ATP-consuming enzymes such ATPases and
apyrases, which is demonstrated by means of the five-den-
tate complex.
Introduction
strongly on the pH and high ionic strength can disturb the rec-
ognition. Therefore, the design of anion-sensitive probes with
sufficient sensitivity and selectivity is a demanding task in ana-
lytical sciences and chemical-sensor technology. In particular,
probes for the detection of biologically relevant anions such as
phosphates, acetate, carbonate, or halides have been the focal
point over the past few years.[2] The difficulties that appeared
in the design of probes for selective determination of nucleo-
tides are an outstanding example. Numerous approaches have
been pursued to achieve probes that can recognize adenosine
triphosphate (ATP) or pyrophosphate with sufficient selectivity
over other polyphosphate anions.[3] These include fluorophores
functionalized with mono-[4,5] or binuclear zinc dipicolylamine
(DPA) complexes[6–12] as a recognition element for polyphos-
phates. Other attempts utilize functionalized polythio-
phenes,[13,14] pyrenes,[15] or the indicator displacement strategy
based on charge-transfer complexes.[16–18] Recently, a lumines-
cent nanoprobe for ATP based on cysteamine-capped quan-
tum dots was designed.[19] Guanosine nucleotides can be de-
tected by fluorescence quenching of naphthyridine-coated
nanoparticles.[20]
Fluorescent probes (fluoroionophores) for the detection of
metal ions are indispensable tools in chemical and environ-
mental analysis or in medical research and diagnosis. The iden-
tification of heavy-metal ions in environmental samples, the
imaging of Ca2+ in living cells, or the determination of Na+
and K+ concentrations in blood are just a few prominent ex-
amples. The design of probes that are sensitive to metal ions is
relatively simple and straightforward, because these are spheri-
cal and differ only in size and charge. A broad variety of selec-
tive receptor systems is available that can be combined with
appropriate fluorescent or phosphorescent reporter systems.
Chelators such as ethylenediaminetetraacetic acid (EDTA) deriv-
atives,
1,2-bis(o-aminophenoxy)ethane-N,N,N’,N’-tetraacetic
acid (BAPTA), crown ethers, aza crown ethers, or cryptands are
frequently used as recognition elements for cation-sensitive in-
dicators.[1]
In contrast, the development of fluoroionophores for anion
recognition is much more complex. Anions exhibit a huge
structural variety and flexibility; their overall charge depends
It has also been shown that sensitized luminescent lantha-
nide ions[21] such as europium and terbium show a response to
different phosphate species.[22,23] In principle, ligand- or metal-
centered interactions of anions with lanthanide complexes can
be used as a sensing mechanism.[24–26] The latter is based on
the exchange of water molecules that are directly coordinated
to the metal center. These quench the lanthanide lumines-
[a] Dr. M. Schꢀferling, T. ꢁꢀritalo, Prof. T. Soukka
Department of Biochemistry/Biotechnology
University of Turku, Tykistçkatu 6A
BioCity 6th floor, 20520 Turku (Finland)
E-mail: michael.schaeferling@utu.fi
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201304942.
Chem. Eur. J. 2014, 20, 5298 – 5308
5298
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
cence by means of nonradiative
energy transfer to high-frequen-
cy OH oscillations.[27,28] The dis-
placement of water ligands by
chelating polyphosphate anions
is reversible, and the response in
luminescence intensity (or life-
time) can be used to monitor
the conversion of ATP in enzy-
matic reactions in real-time.[23,29]
However, in these cases, the sen-
sitizing antenna chromophores
such as tetracycline or norfloxa-
cin only form weakly bound
complexes with the lanthanide
ion. Thus, an excess amount of
sensitizer usually has to be
added, which leads to com-
Scheme 1. Chemical structures of the 7-dentate, 6-dentate, and 5-dentate europium complexes.
pounds with undefined structure and a luminescence response
that is prone to many interferences, because oxyanions gener-
ally have a high affinity toward europium ions. For example,
europium tetracycline responds to citrate and related com-
pounds that occur in the citric acid cycle such as malate and
oxaloacetate.[30] At least, the results revealed that multidentate
polyphosphates are required to induce a significant lumines-
cence response, which can finally even displace the sensitizer,
thus leading to a decrease in luminescence. This was also ob-
served with b-diketonate ligands.[31] The group of Parker
et al.[32,33] designed a series of seven-dentate cyclen complexes
with europium that bear additional antenna chromophores
that can recognize phosphate species and other oxyanions
such as hydrogen carbonate[34] or citrate[35] in water. They were
also modified for intracellular measurements.[36] A similar ap-
proach is based on a carboxy–bipyridyl ligand, but it only
shows poor selectivity between different polyphosphates.[37]
The aim of our study is to demonstrate that the systematic
variation of binding sites of a multidentate ligand can influ-
ence the luminescence response of europium complexes to
different anions. The change in the number of coordination
sites provides a mode of access to control the sensitivity and
selectivity of the response to additional anionic ligands, be-
cause defined numbers of free binding sites are now available
for polydentate anions. Thus, we synthesized a series of non-
macrocyclic 7- to 5-dentate sensitizing ligands, which form
stable complexes with europium in aqueous solution and have
a limited accessibility for larger anions to the metal center. As
Eu3+ ions possess up to nine binding sites,[38] two to four open
binding sides are available, respectively. We used a phenyl–eth-
ynyl–pyridine chromophore as sensitizer[39,40] and evaluated
our concept by studying the luminescence changes in the
presence of polyphosphates and biologically relevant oxidic
anions such as citrate and related compounds of the citric acid
cycle. As the ligand-exchange mechanism occurs fast and re-
versibly, the 5-dentate complex can be used to monitor the
ATP conversion by ATPases or apyrase in real time.
Results and Discussion
Scheme 1 shows the chemical structures of the studied 7-, 6-,
and 5-dentate chelate complexes. Generally, we used the hy-
5
7
persensitive D0! F2 electronic transition of Eu3+, which dis-
plays a maximum at 615 nm, to monitor the luminescence re-
sponses. We evaluated the luminescence intensity response to
a series of different phosphate species (ATP), adenosine di-
phosphate (ADP), adenosine monophosphate (AMP), cyclic ad-
enosine monophosphate (cAMP), pyrophosphate (PPi), and
phosphate anions (Pi)) and carboxyanions (citrate, malate, oxal-
acetate, succinate, a-ketoglutarate, pyruvate, oxalate, carbon-
ate) at pH 7.4 both with a fluorescence spectrometer and a mi-
crowell plate reader. As expected, the replacement of water
molecules in the first coordination sphere of Eu3+ leads to a lu-
minescence increase. The standard deviations of four replicate
measurements were generally around 5%.
Synthesis of the 6- and 5-dentate chelate complexes
The synthesis of the europium-7-dentate (7-D) complex has
been reported previously.[41] The 5-dentate (5-D) complex was
prepared in five steps starting from 4-bromopyridine-2,6-dicar-
boxylic acid diethyl ester.[41] This was converted to 4-bromo-6-
(bromomethyl)picolinic acid ethyl ester 1 with phosphorus tri-
bromide after selective reduction of one carboxy group with
NaBH4. The pyridine triacetate unit 2 was obtained by reaction
with diethyl iminodiacetate. Then, the chromophore 3 was
formed by coupling the triacetate with 4-ethynyl-N,N-dimethyl-
aniline. Finally, the carboxylic ester groups were cleaved to
yield the 5-dentate ligand 4, and EuCl3 was added (see
Scheme S1 in the Supporting Information).
The synthesis of the 6-dentate (6-D) started from 4-bromo-
2,6-bis(bromomethyl)pyridine, which was converted with dieth-
yl iminodiacetate to the corresponding diacetate 5. Then, the
second methylbromide unit was reacted with N-methylglycine
methyl ester to yield the triester 6. The chromophore 7 was
again assembled by coupling with 4-ethynyl-N,N-dimethylani-
line. Scission of all ester groups and transformation to the eu-
Chem. Eur. J. 2014, 20, 5298 – 5308
www.chemeurj.org
5299
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
ropium complex was performed in one step (Scheme S2 in the
Supporting Information).
sured after the addition of excess amounts of EDTA. The result-
ing plot for 7-D is shown by way of example in the Supporting
Information (Figure S1). The stability constants KL have been
calculated in analogy to previously described studies[45,46] ac-
cording to the Equation (1):
General properties of the europium chelates
The luminescence properties of the europium chelates are
summarized in Table 1. The emission spectra are dominated by
5
7
5
7
½EuðEDTAÞꢁ½Lꢁ
the D0! F1 and D0! F2 transitions, which peak at 593 and
ð1Þ
KL ¼ KEDTA
5
½EDTAꢁ½EuLꢁ
615 nm, respectively. The luminescence lifetime t of the D0!
7F2 transition of 7-D in aqueous solution is around 380 ms.[40]
The number of water molecules q(H2O) that are directly coordi-
nated to europium is calculated on the basis of the difference
of the luminescence lifetimes (t) in H2O and D2O, according to
the equation derived by Supkowski and Horrocks (q(H2O)=
1.11[tꢀ1(H2O)ꢀtꢀ1(D2O)ꢀ0.31]).[42] Correspondingly, it can be
deduced that the free binding sites at the europium center in
7-D, 6-D, and 5-D are occupied by two to four water mole-
cules, respectively. Alternative equations for q(H2O) have been
with:
EuL+EDTA$Eu(EDTA)+L
in which L=sum of different protonated free ligands, and
K
EDTA =conditional stability constant for Eu(EDTA). [EuL] was de-
termined from the fluorescence decrease after the addition of
EDTA, and with the assumption that [EDTA]=[EDTA]0 (initial
EDTA concentration) and [Eu(EDTA)]=[L]=[EuL]0ꢀ[EuL], and
logKEDTA =14.53,[44] we calculated the conditional sta-
bility constants of the europium chelates to logKL =
19.8ꢂ0.4 (7-D), 16.6ꢂ0.3 (6-D), and 15.7ꢂ0.2 (5-D),
respectively. For comparison, for the macrocyclic 7-
5-D
Table 1. Spectral properties of the europium chelates.
Europium chelate
7-D
6-D
355
0.0019
260ꢂ50
1040ꢂ250
2.9ꢂ0.6
dentate [Eu(do3a)] (H3do3a=1,4,7,10-tetraazacyclo-
Absorbance maximum [nm]
Relative luminescence intensity[a,b] (H2O)
Luminescence lifetime[a] (H2O) [ms]
Luminescence lifetime[a] (D2O) [ms]
q(H2O)[c]
330
1
355
0.0006
246ꢂ45
2045ꢂ310
3.7ꢂ0.7
dodecane-1,4,7-triacetic acid) complex logK values of
around 21 have been reported.[45]
378ꢂ0.5[40]
2017ꢂ4[40]
2.0
All spectral and luminescence intensity response
measurements were carried out in a time-resolved
(gated) mode with delay times and signal integration
times of 400 ms, respectively. Accordingly, it is possi-
ble to further optimize the performance of each che-
late by tuning the time gates.
[a] 615 nm emission, 345 nm excitation. [b] Referenced to 7-D. [c] According to Hor-
rocks’ equation.[42]
derived from experimental data in the literature by considering
Lifetimes of 7-D were determined by frequency domain
the contributions of different oscillators or the quenching
effect of unbound water molecules,[42,43] although we reference
this simple form of the Horrocks equation. By applying the
equation presented by Beeby et al.,[43] the number of coordi-
nated water molecules would be 2.2 for 7-D,[44] 3.3 for 6-D,
and 4.2 for 5-D.
measurements.
The 7-dentate complex
Although 7-D offers two free binding sites that are saturated
by water molecules, no response of its luminescence could be
observed in the presence of different anions at pH 7.4 over
a broad concentration range, neither to phosphate anions
(ATP, ADP, PPi, and Pi) nor to citrate, which is known to be an
excellent chelating agent for lanthanides. Thus, the 7-dentate
chelate ligand already efficiently shields the europium center
from interactions with anions and no replacement of water
molecules takes place. It has to be kept in mind that there is
an excessive negative charge at the coordination center, which
could also prevent the binding of anions.
The quantum yield of 7-D chelate was reported to be
0.06.[44] The additional ligation of water molecules causes
a more than 500-fold quenching of the luminescence of 6-D
and a more than 1600-fold quenching of 5-D relative to 7-D.
However, the smaller extinction coefficients of 5-D and 6-D
can also contribute to the luminescence decrease (see the Sup-
porting Information). The supplementary water ligands cause
a significant decrease in the luminescence lifetime in water.
The measurements in D2O indicate that the 5-dentate ligand
with the carboxylic acid directly attached to the pyridine ring
might cause a closer coordination of the europium to the an-
tenna chromophore, thus leading to a more efficient energy
transfer. The bathochromic shifts of the absorption maxima of
5-D and 6-D are due to the introduction of the amino substitu-
ent. The concentration of the europium chelates was
50 mmolLꢀ1 in all measurements.
The 6-dentate complex
The 6-D complex has three free binding sites that are occupied
by water molecules. After the addition of polyphosphate
anions, a significant increase in its luminescence intensity can
be observed (Figure 1a). For better comparability, the refer-
enced luminescence intensity change (I/I0) is displayed as ob-
tained by means of a microwell plate reader. The response is
rather unselective; there is no distinction between ATP, ADP,
The conditional stability constants of the complexes were
calculated according to a method introduced by Wu and Hor-
rocks.[45] The decrease of luminescence emission was thus mea-
Chem. Eur. J. 2014, 20, 5298 – 5308
www.chemeurj.org
5300
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
for ATP. All luminescence changes were recorded one minute
after the addition of the anions.
We also assessed the influence on several carboxyanions
such as citrate, malate, oxaloacetate, succinate, a-ketogluta-
rate, pyruvate, oxalate, and carbonate. Among these, citrate
generates a high response with a signal increase of more than
40-fold in the presence of 500 mmolLꢀ1 citrate (Figure 2a),
Figure 1. a) Referenced luminescence responses (I/I0) of 6-D (c=50 mmolLꢀ1
)
!
&
*
^
to ATP ( ), ADP ( ), PPi ( ), and Pi ( ) in TRIS buffer (pH 7.4). I0 is the fluo-
rescence intensity in the absence of phosphate; lexc =340 nm, lem =615 nm;
time-resolved measurement, delay time: 400 ms, signal integration time:
400 ms. b) Linear range of the ATP response with linear fit (intercept: 145,
slope: 8.50, R2 =0.996) and error bars (SD<5%).
and Pi in a concentration range from 10 to 250 mmolLꢀ1. At
higher concentrations, the response to ATP is saturated, where-
as in the case of ADP and PPi, the increase continues up to
concentrations of 1 mmolLꢀ1. In this case, a nearly 30-fold am-
plification of the luminescence can be achieved. It is clear that
the restricted space for binding of voluminous anions provided
by the antenna ligand enables no differentiation between
anions with two or three phosphate units. Only at a higher
excess amount of ADP and PPi to the chelate (10:1), a differ-
ence can be observed relative to the response to ATP. It is
probable that a second phosphate molecule can coordinate to
europium in the case of the diphosphates if they are present
in a high excess amount. The nucleoside rest has no influence
on the binding, as there is no difference in the response to
ADP and PPi. Phosphate ions (Pi) have no effect on the lumi-
nescence of the chelate. The same is the case for other tested
monophosphates such as AMP or cAMP (results not shown).
Hence, multidentate polyphosphate anions are required to dis-
place the water molecules from the first coordination sphere,
whereas monophosphates do not have this ability. Figure 1b
representatively shows the linear range of the response to ATP
from 10 to 100 mmolLꢀ1. The limit of detection (LOD) can be
calculated from the slope of the linear fit and the standard de-
viation (3ꢃ) of the blank sample and amounts to 20 mmolLꢀ1
Figure 2. a) Referenced luminescence responses (I/I0) of 6-D (c=50 mmolLꢀ1
)
&
^
to citrate ( ) and oxalate ( ) in TRIS buffer (pH 7.4). I0 is the fluorescence in-
!
tensity in the absence of anions. For comparison, the response to PPi ( ) is
also displayed; lexc =340 nm, lem =615 nm; time-resolved measurement,
delay time: 400 ms, signal integration time: 400 ms. b) Linear range of the cit-
rate response with linear fit (intercept: 140, slope: 26, R2 =0.992) and error
bars (SD<5%).
whereas oxalate induces a weak response (six-fold increase).
All other tested anions (malate, oxaloacetate, succinate, a-keto-
glutarate, pyruvate, carbonate) did not significantly affect the
luminescence in this concentration range. The linear range of
the citrate detection spans from 10 to 200 mmolLꢀ1 with a LOD
around 7 mmolLꢀ1 (Figure 2b). This range can be shifted with
the concentration of 6-D. The results cannot really be com-
pared to other probes for citrate on the basis of europium
complexes owing to different emphases and evaluation meth-
ods. However, the response seems to be more sensitive than
in the case of a cyclen-derived probe.[47] Europium tetracycline
provides a lower LOD for citrate determination,[30] but it has
only very poor selectivity.
Chem. Eur. J. 2014, 20, 5298 – 5308
www.chemeurj.org
5301
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
The 5-dentate complex
This can be confirmed by the comparison of the purine nu-
cleosides ATP/GTP and ADP/GDP, respectively. The guanosines
that bear one additional carbonyl group induce a lower lumi-
nescence increase than the corresponding adenosine tri- and
diphosphates (Figure 4).
The additional free binding site at the 5-D complex results in
a dramatic amplification of the response to ATP. A nearly 60-
fold increase in the luminescence intensity can be achieved
after the addition of 1 mmolLꢀ1 of ATP (Figure 3a). This leads
Figure 4. Comparison of the referenced luminescence responses (I/I0) of 5-D
(c=50 mmolLꢀ1) to ATP( ), ADP( ), and GTP( ), GDP( ), respectively. TRIS
buffer (pH 7.4); lexc =340 nm, lem =615 nm; time-resolved measurement,
delay time: 400 ms, signal integration time: 400 ms.
!
!
&
&
The signal increase in the presence of citrate is weaker than
in the case of 6-D, which leads to an inversed sensitivity to-
wards ATP (or PPi) and citrate (Figure 5). A minor reaction to
oxalate can also be observed, whereas other carboxyanions
(malate, oxaloacetate, succinate, a-ketoglutarate, pyruvate, car-
bonate) again did not affect the luminescence.
Figure 3. a) Referenced luminescence responses (I/I0) of 5-D (c=50 mmolLꢀ1
)
!
&
*
^
to ATP ( ), ADP ( ), AMP (+), PPi ( ), and Pi ( ) in TRIS buffer (pH 7.4). I0 is
the fluorescence intensity in absence of phosphate; lexc =340 nm,
lem =615 nm; time-resolved measurement, delay time: 400 ms, signal inte-
gration time: 400 ms. b) Linear range of the ATP response with linear fit (in-
tercept: 51225, slope: 19370, R2 =0.980) and error bars (SD ꢃ5%).
to an improved sensitivity; the dynamic range of the response
to ATP is extended and spans from 0.5 to 500 mmolLꢀ1. Fig-
ure 3b shows the corresponding linear range from 0.5 to
25 mmolLꢀ1. The calculated LOD is 0.5 mmolLꢀ1 for ATP. Regret-
tably, the increase in the presence of PPi is similar. The re-
sponse to ADP is nearly identical to 6-D. In both cases a 27-
fold rise of the luminescence can be achieved with 1 mmolLꢀ1
of ADP. Phosphate ions induce a slight increase (six fold with
1 mmolLꢀ1 of Pi), whereas AMP or cAMP (not displayed) did
not affect the luminescence. It is evident that the number of
phosphate units determines the coordination of the anions to
the metal center in 5-D, and therefore the luminescence re-
sponse is in the order ATP>ADP>AMP.
Figure 5. Referenced luminescence responses (I/I0) of 5-D (c=50 mmolLꢀ1
)
&
*
to citrate ( ) and oxalate ( ) in TRIS buffer (pH 7.4). I0 is the fluorescence in-
!
tensity in absence of anions. For comparison, the response to PPi ( ) is also
displayed; lexc =340 nm, lem =615 nm; time-resolved measurement, delay
time: 400 ms, signal integration time: 400 ms.
Luminescence spectra and lifetimes
5
7
5
7
However, the bulky nucleoside units have an interfering in-
fluence, which can be deduced from the lower increase in the
presence of ADP and AMP relative to PPi and Pi, respectively.
The spectral changes of the D0! F1 (lmax =593 and D0! F2
(lmax =613/617 nm) transitions in the presence of ATP are dis-
5
7
played in Figure 6. The emission spectra show intense D0! F2
Chem. Eur. J. 2014, 20, 5298 – 5308
www.chemeurj.org
5302 ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 7. Referenced change of the luminescence lifetime (t/t0) of 6-D
(c=50 mmolLꢀ1) to ATP ( ) and citrate ( ) in TRIS buffer (pH 7.4); t0 is the
luminescence lifetime (260 ms) in the absence of anions; lexc =340 nm,
lem =615 nm.
!
&
sponse of the luminescence lifetime (t/t0) of 6-D to ATP and
citrate is shown in Figure 7. In the presence of ATP, the lifetime
increases over a dynamic range from 0 to 250 mmolLꢀ1 of ATP
from 260 to 675 ms. The results for citrate are similar. In the
case of 5-D, the lifetime increases from 246 to 790 ms in the
same concentration range. The lifetimes are determined by
means of a time-gated detection of the luminescence intensity
after a short pulse of light. Figure S2 in the Supporting Infor-
mation shows by way of example the resulting decay curves
for 5-D in the absence and presence of 250 mmolLꢀ1 of ATP.
The single exponential fits were used to calculate t. The life-
times of both complexes exceed the lifetime of the 7-dentate
after the addition of ATP. This indicates that only one water
molecule remains at the metal center. However, from an ana-
lytical perspective, the evaluation of the lifetime responses is
less attractive in this case, because the changes are rather
small relative to the huge intensity increases, and the measure-
ments were accompanied by high standard deviations up to
20%.
Applicability to enzyme assays
Figure 6. a) Emission spectra of 7-D (c=50 mmolLꢀ1). b) Emission spectra of
6-D (c=50 mmolLꢀ1) in the absence and presence of 10 and 50 mmolLꢀ1 of
ATP. c) Emission spectra of 5-D (c=250 mmolLꢀ1) in the absence and pres-
ence of 150 and 500 mmolLꢀ1 of ATP; lexc =345 nm; time-resolved measure-
ment, delay time: 400 ms, signal integration time: 400 ms.
We evaluated the reversibility of the response and its applica-
bility to monitor the enzymatic conversion of ATP online by
means of two enzymatic assays. Therefore, we used Na+/K+
-ATPase, which cleaves ATP to ADP and Pi in the presence of
Na+, K+, and Mg2+ ions, and apyrase, which decomposes
either ATP or ADP to AMP. First, we studied the influence of
ionic strength on the sensitivity of the probe 5-D. The interfer-
ence of high salt concentrations on the sensing of anions by
luminoionophores turned out to be a fundamental problem.
Therefore, we added increasing amounts of NaCl, KCl, and
MgCl2 to the 2-amino-2-hydroxymethylpropane-1,3-diol (TRIS)
buffer and measured the fluorescence increase after the addi-
tion of different concentrations of ATP and ADP. Generally, the
ascending slope of the response curve decreases with higher
salt concentrations. Figure 8 summarizes the results after the
addition of 500 mmolLꢀ1 of the nucleotides. A tremendous at-
tenuation of the luminescence response can be observed in
transitions with different fine structures after the addition of
ATP relative to the reference 7-D, which reflects the low sym-
metry of the complexes with ATP. The spectrum of 7-D shows
7
7
the typical splitting of the 5D0! F2 and 5D0! F1 transi-
tions.[44,48,49] However, the three components of the latter
could not be completely resolved with the available maximum
instrumental resolution (emission slit=1.5 nm).
The displacement of water molecules from the ligand
sphere of europium is accompanied by an increase in the lumi-
5
7
nescence lifetime. The lifetimes of the D0! F2 transition of 6-
D and 5-D in H2O are indicated in Table 1. The referenced re-
Chem. Eur. J. 2014, 20, 5298 – 5308
www.chemeurj.org
5303
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 8. Referenced luminescence increase (I/I0) of 5-D (c=100 mmolLꢀ1) in
the presence of ATP and ADP (c=500 mmolLꢀ1) at different ionic strength.
1) TRIS buffer (pH 7.4) without additional salts; 2) Tris+Mg2+ (0.5 mmolLꢀ1),
K+ (2.5 mmolLꢀ1), and Na+ (5 mmolLꢀ1); 3) Tris+Mg2+ (2.5 mmolLꢀ1), K+
(12.5 mmolLꢀ1), and Na+ (25 mmolLꢀ1); 4) Tris+Mg2+ (5 mmolLꢀ1), K+
(25 mmolLꢀ1), and Na+ (50 mmolLꢀ1).
Figure 9. Referenced luminescence intensities (I/I0) and monoexponential fit
of 5-D (c=100 mmolLꢀ1) in TRIS buffer (2) in the presence of different mole
fractions of ATP, with x(ATP)=[ATP]/([ATP]+[ADP]) and [ADP]=[Pi]. [ATP]+
[ADP]=500 mmolLꢀ1; I0: intensity for x(ATP)=1.
the presence of MgCl2 (5 mmolLꢀ1), KCl (25 mmolLꢀ1), and
NaCl (50 mmolLꢀ1), which represents optimal salt concentra-
tions for high ATPase activity. The response is predominantly
attenuated by the presence of Mg2+ ions, which have a high
affinity for ATP and shield it from binding to the receptor. This
also causes an almost complete loss of the selectivity between
ATP and ADP. Nevertheless, after the addition of MgCl2
(0.5 mmolLꢀ1), KCl (2.5 mmolLꢀ1), and NaCl (5 mmolLꢀ1), a suffi-
ciently high response and selectivity between ATP and ADP
can be achieved. Thus, we selected these buffer conditions to
perform the ATPase assay.
data is plotted from that time when the signal maximum is
reached (approximately 4–5 min after addition of ATP).
It is necessary to add the enzyme first and then ATP, because
the enzyme also induces a certain increase in the luminescence
of the europium chelate, which occurs on a slow timescale. We
found out that an incubation time of 120 min is required to
ensure a stable initial signal for the measurement after addi-
tion of the enzyme before ATP can be added. Figure 10 shows
ATPase assay
In the next step, we carried out a calibration of the ATPase re-
action [Eq. (2)]:
ATPase
ATP
ADP þ Pi
ð2Þ
ƒƒƒ!
Therefore, we determined the luminescence response in the
presence of different molar ratios of ATP and ADP+Pi, assum-
ing an initial concentration of 500 mmolLꢀ1 of ATP (I0). Figure 9
shows the stepwise decrease in the luminescence of 5-D with
decreasing concentrations of ATP and increasing concentra-
tions of ADP/Pi, until the virtual end point is reached, at which
point all ATP is consumed and the concentration of ADP and Pi
equals 500 mmolLꢀ1, respectively. In this case, the lumines-
cence drops to approximately 50% of its initial value. The data
is fitted monoexponentially according to y=A1 exp(ꢀx/k1)+y0,
with A1 =0.01438, k1 =ꢀ0.27738, and y0 =0.4675 (R2 =0.998).
Figure 10. Time traces of the luminescence decrease owing to the conver-
sion of ATP (initial concentration=1 mmolLꢀ1) by different concentrations of
ATPase from 0.15 to 1.50 mU indicated by 5-D (c=100 mmolLꢀ1). I0 is the
signal maximum after addition of ATP.
the time traces of the conversion of ATP by different amounts
of ATPase indicated by the decreasing luminescence of 5-D.
Thus, the turnover of the enzymatic reaction is directly tran-
scribed into a change in the luminescence signal, and the ki-
netics can be monitored more or less in real time. This is possi-
ble because the response to ATP concentrations occurs very
fast (within approximately 60 s; Figure S3 in the Supporting In-
formation). The intensities were measured approximately every
30 s. For better comparability, the referenced intensities I/I0 are
displayed, with I0 as the value of maximal signal after the addi-
tion of ATP.
The enzymatic assay was performed in the following order.
1) Addition of 5-D and Na+/K+-ATPase to the TRIS buffer that
contained Mg2+ (0.5 mmolLꢀ1), K+ (2.5 mmolLꢀ1), and Na+
(5 mmolLꢀ1). 2) Incubation of the sample mixture for 120 min
at 258C. 3) Addition of ATP and start of the measurement after
1 min at 258C. The final concentration of 5-D was
100 mmolLꢀ1, and of ATP it was 1 mmolLꢀ1. For evaluation, the
Chem. Eur. J. 2014, 20, 5298 – 5308
www.chemeurj.org
5304
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
The denoted ATPase activities were determined in the pres-
ence of 100 mmolLꢀ1 of 5-D at 258C in the selected TRIS buffer
by means of a commercially available ATPase assay standard
protocol[48] based on the colorimetric Taussky–Shorr reagent
(ammonium molybdate/iron(II) sulfate), which indicates the
amount of released Pi.[49] It was found that under the assay
conditions used here the amount of enzyme required to ach-
ieve an activity of 0.06 mU corresponds to an activity of
0.4 mU under optimal conditions (378C, Mg2+ (5 mmolLꢀ1), K+
(5 mmolLꢀ1), Na+ (100 mmolLꢀ1)). Thus, the activity is approxi-
mately seven times lower than specified by the manufacturer
for optimized conditions.
buffer as luminescent indicator according to the apyrase reac-
tion [Eq. (3)]:
apyrase
ATP
AMPþ2 Pi
ð3Þ
ƒƒƒ!
Figure S7 in the Supporting Information shows the time traces
of the consumption of ATP by different amounts of apyrase.
The enzyme does not interfere with the luminescence of 5-D
in this case. Thus, the order of the addition of the reagents
was changed. First, ATP was added and we waited for 30 min
until constant signal levels were obtained, then the reaction
was started by the addition of the enzyme.
For the first 12 min of the reaction, the data points can be
fitted linearly (Figure S4 in the Supporting Information). It can
be deduced that the slope of the fit and therefore the initial
reaction rate depends linearly on the activity of the added
enzyme (Figure S5 in the Supporting Information). The stan-
dard errors of the slopes ranged from 3 to 10%. By compari-
son of the signal drop after 12 min with the calibration func-
tion in Figure 9, the amount of consumed ATP can be deter-
mined and the resulting enzymatic activity calculated. Table 2
In a second run, we used ADP as substrate, following the re-
action [Eq. (4)]:
apyrase
ADP
AMPþPi
ð4Þ
ƒƒƒ!
We have noticed that high signal differences between ADP
and AMP can be achieved in a succinate buffer (results not
shown). Again, the ADP was added first, and then, after an in-
cubation time of 30 min, the apyrase. Figure 11 shows the re-
sulting time traces after the addition of different amounts of
enzyme. Again, the specified apyrase activities have been de-
termined under the respective assay conditions by using a col-
orimetric standard procedure.[53]
Table 2. Results of the ATPase assay and comparison with reference
method.[48]
Activity of added AT- Measured activity with Measured initial reaction
Pase[a] [mU]
5-D[b] [mU]
rate[c] (DI) [min]
0.15
0.75
1.13
1.50
0.66
0.018ꢂ0.0018
0.88
1.25
1.66
0.023ꢂ0.0016
0.030ꢂ0.0028
0.035ꢂ0.0014
[a] Measured by ATPase assay standard protocol (Taussky–Shorr reagent)
under reaction conditions used in this work. [b] According to kinetic anal-
ysis. [c] From slopes of linear fits of Figure S1 in the Supporting Informa-
tion.
shows the results of the kinetic evaluation using 5-D and the
comparison with the commercially available ATPase assay. Low
enzyme activities yielded a comparatively high noise of the re-
corded time traces, thereby resulting in large deviations in the
determined activities relative to the reference method. Howev-
er, if higher amounts of enzymes are added (ꢄ5 mU), the de-
termined values are in good conformity with the effective ac-
tivities measured by the Taussky–Shorr reference method.
For a further assessment of the method, the known inhibitor
Ouabain (g-strophanthin) was added to the reaction mixture.
Figure S6 in the Supporting Information shows the resulting
time traces with 0.75 mU ATPase. The results demonstrate that
the probe 5-D can also indicate the inhibition of ATPases. The
enzyme is mostly inhibited in the presence of 1 mmolLꢀ1 of
Ouabain, whereas 100 mmolLꢀ1 completely deactivates the
enzyme, which is in accordance with the literature.[50]
Figure 11. Time traces of the luminescence decrease owing to the conver-
sion of ADP (initial concentration=0.5 mmolLꢀ1) by different concentrations
~
&
*
of apyrase representing 0.036 U ( ), 0.072 U ( ), and 0.108 U ( ) indicated
by 5-D (c=50 mmolLꢀ1). I0 is the signal directly after addition of apyrase
(succinate buffer pH 6.6 containing 4 mmolLꢀ1 CaCl2).
Conclusion
With this study we have demonstrated that the selectivity and
sensitivity of the luminescence response of sensitized europi-
um complexes to anions can be influenced by a variation of
the binding sites of the ligand. The number of free coordina-
tion sites at the metal center plays an important role in the
ability to bind polyphosphates or multidentate oxyanions and
to displace water molecules from the ligand sphere of the lan-
thanide ion.
Apyrase assay
Apyrases can implement either ATP or ADP as substrate. First
we detected the turnover of ATP again by using 5-D in TRIS
Chem. Eur. J. 2014, 20, 5298 – 5308
www.chemeurj.org
5305
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
It has to be noted that the luminescence increase occurs
very fast, for 6-D as well as for 5-D, and can be monitored
after just 1 min reaction time with a saturation after 90 s. The
dynamic range of the response can be adjusted with the con-
centration of the europium complex. The results revealed that
7-D has no sensitivity to any of the tested anions. Multidentate
anions such as polyphosphates or citrate are required for the
replacement of water molecules in the cases of 6-D and 5-D.
The limited space for coordination of bulky anions at the 6-D
center results in a narrow dynamic range of the response to
anions and overall a poor selectivity. In contrast, 5-D exhibits
broad dynamic ranges with high sensitivity, particularly for ATP,
and a better distinction between different anion species. To
summarize, 5-D appears to be a promising candidate for ATP
detection in the absence of PPi, whereas 6-D is more suitable
as a probe for citrate. These luminoionophores have the typical
merits of lanthanide complexes, such as a large wavelength
shift between excitation and emission, a sharp emission band,
and decay times higher than 100 ms, which are perfectly suited
for time-resolved luminescence determination. A displacement
of the sensitizing chelate ligand can take place at a high
excess amount of anions relative to the europium complex
(>25:1), which is indicated by a decrease in luminescence in-
tensities and lifetimes.
Experimental Section
Materials
All reagents were used in American Chemical Society grade. Ade-
nosine 5’-triphosphatase (ATPase) from porcine cerebral cortex,
apyrase from potato, and Ouabain octahydrate was obtained from
Sigma. Tetrasodium pyrophosphate decahydrate (PPi) was obtained
from Alfa Aesar, sodium hydrogenphosphate dehydrate (Pi) from
J.T. Baker, sodium citrate tribasic dehydrate from Riedel-de-Haꢄn;
and adenosine-5’-triphosphate disodium salt (ATP), adenosine-5’-di-
phosphate sodium salt (ADP), adenosine 5’-monophosphate
sodium salt (AMP), adenosine-3’,5’-cyclic monophosphate sodium
salt (cAMP), adenosine 3’,5’-cyclic monophosphate (cAMP), oxalo-
acetic acid, sodium pyruvate, a-ketoglutaric acid disodium salt de-
hydrate, sodium oxalate, l-malic acid, sodium carbonate monohy-
drate, and succinic acid were obtained from Sigma–Aldrich.
Synthesis of 5-D complexes
4-Bromo-6-(bromomethyl)picolinic acid ethyl ester (1): 4-Bromo-
pyridine-2,6-dicarboxylic acid diethyl ester (7.1 g, 23 mmol) was dis-
solved in ethanol (70 mL), and granular NaBH4 (770 mg) was
added. The mixture was heated at reflux for one hour, and a white
precipitate was formed. The solvent was evaporated with a rotary
evaporator, and water (70 mL) was added. After stirring at room
temperature for 10 min, the product was extracted with CH2Cl2 (ca.
100 mL). Purification with column chromatography (10% MeOH/
CH2Cl2) gave the intermediate (1.2 g). This was dissolved in DMF
(6 mL) and the solution was added dropwise to PBr3 (0.516 mL) in
DMF (3 mL). The mixture was stirred at room temperature over-
night. After this, 5% NaHCO3 (ca. 10 mL) was added and the solu-
tion was extracted with petroleum ether. After evaporation, pure
product 1 (1.2 g, 4 mmol) was obtained with short filtration
through a silica layer using CH2Cl2 as eluent.
As the probe 5-D shows a fast, reversible, and selective re-
sponse to ATP (relative to ADP), it can be applied to monitor
the conversion of ATP in enzymatic reactions. This has been
demonstrated by means of an online ATPase assay. Enzyme ac-
tivities can be calculated from the recorded luminescence ki-
netics and the impact of inhibitors can be screened. It is also
possible to monitor the conversion of ATP or ADP by apyrases.
Although the selectivity between ATP and PPi is rather low, the
differences between the signal responses to ATP, ADP, and
AMP are large enough to indicate these enzymatic reactions.
Thus, this probe represents a valuable alternative to com-
mercially available assays for ATPase or apyrase activity, which
are all based on end-point methods and require laborious
sample-preparation steps. The response characteristic suggests
that it can be also used to monitor the activity of kinases or
GTPases. The assay protocols can be further optimized by the
adaption of temperature, reagents, and salt concentrations,
and the type of buffer.
Diethyl-2,2’-({[bromo-(6-ethoxycarbonyl)pyridine-2-yl]methyl}-
azenediyl)diacetate (2): 4-Bromo-6-(bromomethyl)picolinic acid
ethyl ester (1.2 g), diethyliminodiacetate (0.845 mL), and K2CO3
(800 mg) were added to acetonitrile (50 mL), and the mixture was
heated to reflux. The reaction was followed with TLC (30% petrole-
um ether/ethyl acetate) and after 4 h, only one compound was
visible. The mixture was filtered and the filtrate was evaporated to
dryness. After column chromatography using 25% ethyl acetate/
petroleum ether (EtOAc/PE) as eluent, pure product 2 (1.5 g,
3.5 mmol) was obtained.
In our future work we aim to further reduce the number of
binding sites of the ligand and to evaluate the influence of
other factors on the luminescence response such as odd posi-
tive charges on the metal center and additional voluminous
substituents. The pyridine–diamino–tetraacetic acid structure
of the chelating ligand can be easily modified in different ways
to achieve a large variety of europium chelates with tailored
properties.[54] This could pave the way for the design of lumi-
nescent sensors with an improved sensitivity for specific
anions. The ligands can be functionalized with amino-reactive
groups (e.g., isothiocyanate) as shown in 7-D and coupled to
amino-modified polymer matrices, which can provide an im-
proved selectivity towards different anions and is the basis for
the fabrication of ready-to-use optical sensor devices.
Diethyl 2,2’-{[(4-{[4-(dimethylamino)phenyl]ethynyl}-6-(ethoxy-
carbonyl)pyridin-2-yl)methyl]azanediyl}diacetate (3): Compound
2 (310 mg, 0.73 mmol) and 4-ethynyl-N,N-dimethylaniline (120 mg)
were dissolved in Et3N/THF (5 mL/5 mL), and the solution was satu-
rated with argon for 5 min, then [Pd(PPh3)Cl2] (7 mg) and CuI
(4 mg) were added. The mixture was kept under argon flow for
10 min, then the bottle was capped, and it was stirred for 3 days
under an argon atmosphere at 548C. The solvents were evaporat-
ed, and CH2Cl2 (40 mL) was added. The organic phase was washed
twice with water (40 mL) and then evaporated to dryness. Purifica-
tion was carried out with silica gel chromatography by using an
EtOAc/PE gradient. The treatment with 15% EtOAc/PE eluted vari-
ous catalyst parts and the first visible fractions, whereas the last
yellow fraction was eluted with 50% EtOAc/PE and gave the pure
product 3 (260 mg, 0.52 mmol).
Chem. Eur. J. 2014, 20, 5298 – 5308
www.chemeurj.org
5306
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
2,2’-{[(6-Carboxy-4-{[4-(dimethylamino)phenyl]ethynyl}pyridin-2-
yl)methyl]azanediyl} diacetic acid (4): Compound (50 mg,
For spectroscopic analysis, all measurements were carried out in
quartz cuvettes. The concentration of the europium chelates was
always 50 mmolLꢀ1 in 25 mm TRIS buffer (pH 7.4) if not otherwise
stated. The corresponding amounts of anions were subsequently
added in the form of highly concentrated solutions in TRIS buffer.
Absorption spectra were recorded using a Varian Cary 300 Bio UV/
Vis instrument. Luminescence spectra and lifetimes were recorded
using a Varian Cary Eclipse fluorescence spectrophotometer. The
uncorrected spectra were obtained in the phosphorescence mode
with delay and gate times of 400 ms, respectively, at an excitation
wavelength of 345 nm and emission slits of 1.5 or 2.5 nm. The life-
time measurements were carried out by integrating 100 excitation
cycles with a delay time of 100 ms and a gate time of 40 ms.
3
100 mmol) was dissolved in methanol (5 mL) and moved to
a round bottom flask. Concentrated HCl (5 mL) was added, and the
methanol was evaporated. Afterwards, the solution was heated
under reflux conditions for 3 h and evaporated to dryness to
obtain product 4 (40 mg, 97 mmol).
Synthesis of 5-D: Saponified ligand 4 (25 mg) was dissolved in
water (0.5 mL), and the solution was neutralized with NaHCO3. Af-
terwards, EuCl3·6H2O (15 mg) dissolved in water (0.1 mL) was
added while the solution was kept at neutral pH.
Synthesis of 6-D complexes
Microwell plate assays
Diethyl 2,2’-({[4-bromo-6-(bromomethyl)pyridin-2-yl]methyl}aza-
nediyl)diacetate (5): 4-Bromo-2,6-bis(bromomethyl)pyridine (3 g,
9 mmol) was dissolved in acetonitrile (60 mL) at 508C. Potassium
carbonate (6 g) and diethyliminodiacetate (1 mL) were added, and
the mixture was stirred at 608C for 3 h. Then the mixture was fil-
tered, and the filtrate was evaporated to dryness. Purification with
column chromatography (20% EtOAc/PE) resulted in product 5
(1.5 g, 3.3 mmol) and unreacted starting material (1.07 g).
The luminescence responses to the anions were measured using
a Victor 1420 Multilabel counter from Wallac, Perkin–Elmer Life and
Analytical Sciences (Wellesley, MA). The excitation wavelength was
340 nm, and the detection wavelength 615 nm. A time-resolved
detection mode was used with a delay time and signal integration
time of 400 ms, respectively. All measurements were performed
1 min after addition of the anions in 96-microwell plates with
transparent flat bottom from Nunc in four replicates. The final
sample volume was 100 mL in each well, which contained the euro-
pium chelate (50 mmolLꢀ1) and the corresponding amount of
anions in 25 mm TRIS buffer (pH 7.4).
Diethyl
2,2’-{[(4-bromo-6-{[(2-methoxy-2-oxoethyl)(methyl)-
amino]methyl}pyridin-2-yl)methyl]azanediyl}diacetate (6): Com-
pound 5 (250 mg, 1.75 mmol), K2CO3 (100 mg), and N-methylgly-
cine methyl ester (400 mg) were dissolved in acetonitrile (10 mL),
and the mixture was heated under reflux conditions for 20 h. The
mixture was filtered, then the filtrate was evaporated to dryness
and purified with column chromatography by using PE as the
main solvent and applying an EtOAc gradient from 20 to 50%.
Compound 6 (280 mg, 0.60 mmol) was obtained with a 50% gradi-
ent.
The ATPase assays were carried out by dissolving the required
amount of 5-D and ATPase (+inhibitor) in TRIS buffer that con-
tained Mg2+ (0.5 mmolLꢀ1), Na+ (5 mmolLꢀ1), and K+ (2.5 mmolLꢀ1
)
to achieve the desired final concentrations or activities, respective-
ly. The sample mixture was incubated at 258C for 2 h. Finally, the
necessary amount of an ATP stock solution, which was thermostat-
ted at 258C, was added to give a final volume of 100 mL per well
and the measurement was started after 1 min. In this case, the
final concentrations were 100 mmolLꢀ1 of 5-D and 1 mmolLꢀ1 of
ATP per well. The mean intensities from three parallel runs are
always displayed.
Diethyl
2,2’-{[(4-{[4-(dimethylamino)phenyl]ethynyl}-6-{[(2-me-
thoxy-2-oxoethyl)(methyl)amino]methyl}pyridin-2-yl)methyl]aza-
nediyl}diacetate (7): Compound 6 (280 mg) and 4-ethynyl-N,N-di-
methylaniline (120 mg) were dissolved in Et3N/THF (5 mL/5 mL)
and the mixture was flushed with argon for 10 min. Then [Pd-
(PPh3)Cl2] (1 mg) and CuI (1 mg) were added, and the mixture was
kept under argon flow for another 10 min. After this, the flask was
capped and stirred at 548C overnight. The solvents were evaporat-
ed and the residue dissolved in a small volume of EtOAc. Purifica-
tion with column chromatography using an EtOAc/PE gradient (10
to 50%) resulted in product 7 (300 mg, 0.56 mmol).
The apyrase assays were performed in a similar way. The succes-
sion of reagent addition was in this case 5-D, ATP or ADP (incuba-
tion for 30 min at 258C), and finally apyrase. The reaction with ATP
as substrate was carried out in 25 mm TRIS buffer (pH 6.6) that
contained Ca2+ (1 mmolLꢀ1) at 258C. The reaction with ADP was
carried out in 40 mm succinate buffer (pH 6.6) that contained Ca2+
(4 mmolLꢀ1).
The conditional stability constants were determined by the addi-
tion of excess amounts of EDTA to the europium chelates in three
replicate measurements. The luminescence evaluation was carried
out by using 7-D (1 nmolLꢀ1) and different batches of EDTA be-
tween 100 and 500 mmolLꢀ1 after a reaction time of 48 h at 508C,
6-D (1 mmolLꢀ1) with EDTA batches between 10 and 100 mmolLꢀ1
(48 h, RT), and 5-D (5 mmolLꢀ1) with EDTA batches between 75 and
200 mmolLꢀ1 (48 h, RT).
Synthesis of 6-D: Compound 7 (240 mg) was dissolved in metha-
nol (5 mL). Then concentrated HCl (8 mL) was added, and the mix-
ture was heated under reflux conditions for 3 h, then evaporated
to dryness. The saponified ligand was dissolved in water (10 mL)
and neutralized with NaHCO3. EuCl3·6H2O (170 mg) was dissolved
in water (2 mL) and added. After 2 h at room temperature, the
formed precipitate was collected by centrifugation. After drying
under reduced pressure, 6-D (118 mg, 0.19 mmol) was obtained.
Acknowledgements
Spectroscopy
The authors thank Joonas Mꢀkelꢀ for providing the 7-D che-
late. M.S. acknowledges TEKES (Finnish Funding Agency for
Technology and Innovation) for financial support.
Spectroscopic characterization of the ligands and europium com-
plexes (UV/Vis, NMR spectroscopy, MS) can be found in the Sup-
porting Information.
Chem. Eur. J. 2014, 20, 5298 – 5308
www.chemeurj.org
5307
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[28] D. Parker, R. S. Dickins, H. Puschmann, C. Crossland, J. A. K. Howard,
Chem. Rev. 2002, 102, 1977–2010.
Keywords: anions · chelates · lanthanides · luminescence ·
[29] C. M. Spangler, C. Spangler, M. Gçttle, Y. Shen, W. J. Tang, R. Seifert, M.
Schꢀferling, Anal. Biochem. 2008, 381, 86–93.
sensors
[30] Z. Lin, M. Wu, M. Schꢀferling, O. S. Wolfbeis, Angew. Chem. 2004, 116,
1767–1770; Angew. Chem. Int. Ed. 2004, 43, 1735–1738.
[31] N. Shao, J. Jin, G. Wang, Y. Zhang, R. Yang, J. Yuan, Chem. Commun.
2008, 1127–1129.
[32] P. Atkinson, Y. Bretonniꢇre, D. Parker, Chem. Commun. 2004, 438–439.
[33] P. Atkinson, B. S. Murray, D. Parker, Org. Biomol. Chem. 2006, 4, 3166–
3171.
[34] Y. Bretonniꢇre, M. J. Cann, D. Parker, R. Slater, Chem. Commun. 2002,
1930–1931.
[35] D. Parker, J. Yu, Chem. Commun. 2005, 3141–3143.
[36] E. J. New, D. Parker, D. G. Smith, J. W. Walton, Curr. Opin. Chem. Biol.
2010, 14, 238–246.
[37] L. J. Charbonniꢇre, R. Schurhammer, S. Mameri, G. Wipff, R. F. Ziessel,
Inorg. Chem. 2005, 44, 7151.
[38] G. R. Choppin, Pure Appl. Chem. 1971, 27, 23–41.
[39] H. Takalo, E. Hꢀnninen, J. Kankare, Helv. Chim. Acta 1993, 76, 877–883.
[40] I. Hyppꢀnen, T. Soukka, J. Kankare, J. Phys. Chem. A 2010, 114, 7856–
7867.
[41] H. Takalo, V.-M. Mukkala, H. Mikola, P. Liitti, I. Hemmilꢀ, Bioconjugate
Chem. 1994, 5, 278–282.
[42] R. M. Supkowski, W. DeW. Horrocks, Inorg. Chim. Acta 2002, 340, 44–
48.
[1] M. Schꢀferling, Angew. Chem. 2012, 124, 3590–3614; Angew. Chem. Int.
Ed. 2012, 51, 3532–3554.
[2] T. Gunnlaugsson, M. Glynn, G. M. Tocci, P. E. Kruger, F. M. Pfeffer, Coord.
Chem. Rev. 2006, 250, 3094–3117.
[3] C. Spangler, M. Schꢀferling, O. S. Wolfbeis, Microchim. Acta 2008, 161,
1–39.
[4] S. J. Lee, A. S. Rao, Y. H. Shin, H.-J. Chung, Y. Huh, K. H. Ahn, J. Jung, J.
Mol. Histol. 2013, 44, 241–247.
[5] M. Schꢀferling, T. Lang, A. Schnettelker, J. Fluoresc. 2013, DOI
10.1007s10895–013–1292–9.
[6] C. Park, J.-I. Hong, Tetrahedron Lett. 2010, 51, 1960–1962.
[7] Y. Kurishita, T. Kohira, A. Ojida, I. Hamachi, J. Am. Chem. Soc. 2010, 132,
13290–13299.
[8] A. Ojida, S. K. Park, Y. Mito-oka, I. Hamachi, Tetrahedron Lett. 2002, 43,
6193–6195.
[9] A. Ojida, Y. Miyahara, J. Wongkongkatep, S. Tamaru, K. Sada, I. Hamachi,
Chem. Asian J. 2006, 1, 555–563.
[10] Y. J. Jang, E. J. Jun, Y. J. Lee, Y. S. Kim, J. S. Kim, J. Yoon, J. Org. Chem.
2005, 70, 9603–9606.
[11] D. H. Lee, S. Y. Kim, J. I. Hong, Angew. Chem. 2004, 116, 4881–4884;
Angew. Chem. Int. Ed. 2004, 43, 4777–4780.
[12] A. J. Moro, J. Schmidt, T. Doussineau, A. Lapresta-Fernandꢅz, J. Wegener,
G. J. Mohr, Chem. Commun. 2011, 47, 6066–6068.
[13] C. Li, M. Numata, M. Takeuchi, S. Shinkai, Angew. Chem. 2005, 117,
6529–6532; Angew. Chem. Int. Ed. 2005, 44, 6371–6374.
[14] C. Spangler, T. Lang, M. Schꢀferling, Dyes Pigm. 2012, 95, 194–200.
[15] Z. Xu, N. J. Singh, J. Lim, J. Pan, H. N. Kim, S. Park, K. S. Kim, J. Yoon, J.
Am. Chem. Soc. 2009, 131, 15528–15533.
[16] S. L. Wiskur, H. Ait-Haddou, J. J. Lavigne, E. V. Anslyn, Acc. Chem. Res.
2001, 34, 963–972.
[17] S. Atilgan, E. U. Akkaya, Tetrahedron Lett. 2004, 45, 9269–9271.
[18] P. P. Neelakandan, M. Hariharan, D. Ramaiah, J. Am. Chem. Soc. 2006,
128, 11334–11335.
[43] A. Beeby, I. M. Clarkson, R. S. Dickins, S. Faulkner, D. Parker, L. Royle,
A. S. de Sousa, J. A. G. Williams, M. Woods, J. Chem. Soc. Perkin Trans. 2
1999, 493–593.
[44] Q. Wang, K. N. Nono, M. Syrjꢀnpꢀꢀ, L. J. Charbonniꢇre, J. Hovinen, H.
Hꢀrmꢀ, Inorg. Chem. 2013, 52, 8461–8466.
[45] S. L. Wu, W. DeW. Horrocks, Anal. Chem. 1996, 68, 394–401.
[46] N. N. Katia, A. Lecointre, M. Regueiro-Figueroa, C. Platas-Iglesias, L. J.
Charbonniꢇre, Inorg. Chem. 2011, 50, 1689–1697.
[47] R. Pal, D. Parker, L. C. Costello, Org. Biomol. Chem. 2009, 7, 1525–1528.
[48] J.-C. G. Bꢆnzli, C. Piguet, Chem. Soc. Rev. 2005, 34, 1048–1077.
[49] N. M. Shavaleev, F. Gumy, R. Scopelliti, J.-C. G. Bꢆnzli, Inorg. Chem. 2009,
48, 5611–5613.
[50] Sigma Quality control test procedure: Enzymatic assay of adenosine-5’-
triphosphatase (EC 3.6.1.3), Sigma Prod. No. A7510.
[51] H. H. Taussky, E. Shorr, J. Biol. Chem. 1953, 202, 675–685.
[52] F. Ebner, Br. J. Pharmacol. 1990, 101, 337–343.
[53] Sigma Quality control test procedure: Enzymatic assay of apyrase (EC
3.6.1.5), ATP or ADP as substrate.
[19] W. Tedsana, T. Tuntulani, W. Ngeontae, Anal. Chim. Acta 2013, 783, 65–
73.
[20] P. J. M. Cywinski, A. J. Moro, T. Ritschel, N. Hildebrandt, H.-G. Lçhmanns-
rçben, Anal. Bioanal. Chem. 2011, 399, 1215–1222.
[21] J.-C. G. Bꢆnzli, Chem. Rev. 2010, 110, 2729–2755.
[22] Y. Miao, J. Liu, F. Hou, C. Jiang, J. Lumin. 2006, 116, 67–72.
[23] M. Schꢀferling, O. S. Wolfbeis, Chem. Eur. J. 2007, 13, 4342–4349.
[24] S. J. Butler, D. Parker, Chem. Soc. Rev. 2013, 42, 1652–1666.
[25] M. L. Cable, J. P. Kirby, H. B. Gray, A. Ponce, Acc. Chem. Res. 2013, 46,
2576–2584.
[54] P. Kadjane, M. Starck, F. Camerel, D. Hill, N. Hildebrandt, R. Ziessel, L. J.
Charbonniꢇre, Inorg. Chem. 2009, 48, 4601–4603.
[26] L. M. P. Lima, R. Tripier, Curr. Inorg. Chem. 2011, 1, 36–60.
[27] A. Heller, J. Am. Chem. Soc. 1966, 88, 2058–2059.
Received: December 18, 2013
Published online on March 27, 2014
Chem. Eur. J. 2014, 20, 5298 – 5308
www.chemeurj.org
5308
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim