Non-cytotoxic, bifunctional EuIII and TbIII luminescent macrocyclic complexes for luminescence resonant energy-transfer experiments

Article abstract of DOI:10.1002/chem.200700819

A new macrocyclic ligand, L³, has been synthesised, based on the cyclen framework grafted with three phenacyl light-harvesting groups and a C5-alkyl chain bearing a carboxylic acid function as a potential linker for biological material. Acidity

Full text of DOI:10.1002/chem.200700819

DOI: 10.1002/chem.200700819
Non-Cytotoxic, Bifunctional EuIII and TbIII Luminescent Macrocyclic
Complexes for Luminescence Resonant Energy-Transfer Experiments
Anne-Claire Ferrand,[a] Daniel Imbert,[a, b] Anne-Sophie Chauvin,[a]
Caroline D. B. Vandevyver,[a] and Jean-Claude G. Bünzli*[a]
Abstract: A new macrocyclic ligand,
L3, has been synthesised, based on the
cyclen framework grafted with three
phenacyl light-harvesting groups and a
C5-alkyl chain bearing a carboxylic acid
function as a potential linker for bio-
logical material. Acidity constants are
determined by spectrophotometric ti-
trations, as well as conditional stability
constants for the resulting 1:1 com-
plexes with trivalent lanthanide ions.
The complexes have stabilities compa-
rable to 1,4,7,10-tetrakis(carbamoyl-
methyl)-1,4,7,10-tetraazacyclododecane
(dtma) complexes, with pLnꢁ12–13.
Photophysical properties of the ligand
and of the EuL3 and TbL3 complexes
have been determined for both micro-
crystalline samples and solutions in
water and acetonitrile. They point to
the metal ion being present in an envi-
ronment with axial symmetry derived
from the C4 point group. The hydration
number determined for TbL3 decreases
with increasing pH value and becomes
fractional at pH 7.5, which points to an
equilibrium between two differently
solvated species and probably to the
participation of the deprotonated car-
boxylic acid chain in the complexation.
The quantum yields in water (1.9% for
EuIII, 3.4% for TbIII) are smaller than
those for complexes with the symmetri-
cally substituted parent macrocycle, but
efficient luminescence resonant energy
transfer (LRET) was observed when
Cy5 dye was added to the solutions. Fi-
nally, the influence of the TbL3 com-
plex on cell viability is tested on both
malignant (5D10 mouse hybridoma,
Jurkat human T leukaemia, MCF-7
human breast carcinoma) and non-ma-
lignant (Hacat human keratinocyte)
cell lines. Cell viability after 24 h incu-
bation at 378C with 500 mm TbL3 was
>90% for all cell lines, except Jurkat
(>70%). All of these properties make
LnL3 complexes interesting potential
probes for bioanalyses.
Keywords: cytotoxicity
transfer lanthanides
cence · macrocyclic ligands
·
energy
lumines-
·
·
Introduction
of trivalent lanthanides (LnIII) in water.[2] The resulting
ennea-coordinate complexes are thermodynamically stable
and kinetically inert and they feature one water molecule
occupying the axial site.[3] These properties, together with
large LD50 values,[4] stirred immediate interest for bioinor-
ganic and medical applications, the first of which was in the
design of contrast agents for magnetic resonance imaging,
The synthesis of the tetracarboxylic acid derivative of cyclen
(1,4,7,10-tetraazacyclododecane), H4dota, and of some of its
d-transition-metal complexes by Stetter and Frank in 1976[1]
was followed three years later by the recognition that this
derivatised macrocycle is ideally suited for the complexation
for example [Gd
A
G
a wealth of developments during the last 20 years.[6] The dis-
tinctive photophysical properties of LnIII ions,[7] which emit
line-like, easily recognisable luminescence spectra and have
long excited-state lifetimes, open the way to highly sensitive
time-resolved bioanalyses.[8,9] These properties can be ade-
quately tuned in complexes with cyclen derivatives, in par-
ticular by grafting sensitising and/or coordinating groups
onto the macrocycle scaffold for light harvesting and subse-
quent energy transfer onto the metal ion. Laporte forbidden
f–f transitions are too weak to be efficient vectors for LnIII
excitation and the cyclen framework in itself is not a good
[a] Dr. A.-C. Ferrand, Dr. D. Imbert, Dr. A.-S. Chauvin,
Dr. C. D. B. Vandevyver, Prof. Dr. J.-C. G. Bünzli
École Polytechnique FØdØrale de Lausanne
Laboratory of Lanthanide Supramolecular Chemistry
BCH 1402, 1015 Lausanne (Switzerland)
Fax:(+41)21-698-9825
A
E-mail: jean-claude.bunzli@epfl.ch
[b] Dr. D. Imbert
Laboratoire de Reconnaissance Ionique
UMR-E3 CEA-UJF, DRFMC
CEA-Grenoble (France)
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the author.
8678
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 8678 – 8687

FULL PAPER
chromophore. The first sensitising group tested was carbos-
tyril 124 (7-amino-4-methyl-2-(1H)-quinolinone),[10] with
modest results as far as the tetraamide derivative of cyclen,
1,4,7,10-tetrakis(carbamoylmethyl)-1,4,7,10-tetraazacyclodo-
decane (dtma), is concerned.[11] Other sensitising groups
were rapidly proposed[12] and rich chemistry and analytical
applications were developed, for instance, for sensing pH,
pO2 and biorelevant anion values[13–16] or for producing
highly luminescent thin films for light-emitting diodes.[17]
Lanthanide-containing responsive luminescent probes tail-
ored from cyclen derivatives are presently ubiquitous tools
in bioanalyses. Lately, they have sustained further engineer-
ing so that in cellulo molecular imaging and sensing have
become feasible, by making use of both visible and near-in-
frared metal-based luminescence.[18,19]
Scheme 1. Ligand formulae
In our laboratories, we pursue the development of visi-
ble[20] and near-infrared[21] luminescent probes through sev-
eral strategies, one of which is based on dtma derivatives
(Scheme 1).[22,23] In particular, we have shown that ligands
L1 and L2 react with lanthanide triflates to form stable 1:1
complexes in water, as shown by the pLn values[24] for L1
(7.5 and 9 for EuIII and TbIII) and by the fact that logK
ꢁ12–13 for L2. The metal-centred luminescence of the
SmIII, EuIII, TbIII and DyIII complexes is sensitised to a varia-
ble extent; the best chromophore for EuIII is the phenylphe-
nacyl group, while the phenacyl group is best suited for
energy transfer on TbIII.[25] Building on the latter work, we
have now replaced one phe-
Results and Discussion
Ligand synthesis and acid–base properties: Ligand synthesis
followed the now-classical regioselective method proposed
for introducing 3+1 substituents on the cyclen scaffold.[29,30]
The carboxylic acid substituent 1 was obtained from bro-
moacetyl bromide and 6-aminocaproic acid (Scheme 2),
while the phenacyl pendant 3 was synthesised as previously
described.[25] One amine function of cyclen was protected by
the method of Platzek et al. for 1,7-substituted cyclens:[31]
N,N-dimethylformamide dimethylacetal (DMF-DMA) was
treated with cyclen to produce 1,4,7,10-tetraazatricyclo-
nacyl substituent with an ali-
phatic chain bearing a carbox-
ylic acid function; this group,
along with primary amine,
chlorosulfonyl and isothiocya-
nate, is among the most
common linking moieties that
can covalently interact with
biologically relevant molecules
such as proteins.[26] The aim of
this study is to examine the in-
fluence of this substitution on
both the thermodynamic and
photophysical properties of the
resulting lanthanide complexes.
In addition, we establish that
LnL3 (Ln: Eu, Tb) complexes
efficiently transfer energy on
cyanine 5 (Cy5), a dye used in
luminescence
resonance
energy transfer (LRET) ex-
periments (for example, in evi-
dencing DNA hybridisation)[27]
and in homogeneous lumines-
cence immunoassays.[28] Finally,
the influence of the TbL3 com-
plex on cell viability is tested
on various malignant and non-
malignant cell lines.
Scheme 2. i) 4m NaOH, H2O, 0 8C; ii) N,N-dimethylformamide dimethylacetal, toluene, 1208C; iii) EtOH, H2O,
ꢀ188C; iv) 3, NEt3, THF; v) HClaq, H2O/acetone; vi) 1, Cs2CO3, MeCN, 24 h, RT.
Chem. Eur. J. 2007, 13, 8678 – 8687
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8679

J.-C. G. Bünzli et al.
A
latter may arise from stabilisation through hydrogen bond-
ing, much as carboxylate groups can assist protonation of
ligand amine groups.[32] This would imply that the carboxylic
acid arm folds over the macrocycle, a hypothesis that may
be substantiated by the hydration numbers determined for
TbL3 (see below). Another conclusion is that the replace-
ment of one amide pendant by the C5 carboxylic acid chain
does not noticeably influence the protonation constants of
the macrocycle amine functions. A comparison of the proto-
nation constants found for L3 and L4 with published data for
dtma is more delicate since only three logKa values are re-
ported for this macrocycle.[32] The smallest protonation con-
stant is comparable to the logKa4 value for L4 and the
middle value is a reasonable match with the logKa3 values
for L3 and L4. On the other hand, the largest value corre-
sponds to the logKa2 value of L3 and L4 and not the logKa1
value. A distribution diagram for L3 is shown in Figure S1 in
the Supporting Information.
careful hydrolysis at low temperature this produced the
formaldehyde-protected material 2b. Use of a quasistoichio-
metric quantity of DMF-DMA (1.05 equivalents) resulted in
avoidance of the formation of the diformylated product. A
classical reaction between the three unprotected amine func-
tions of 2b and N-(phenacyl)bromoacetamide (3) in the
presence of triethylamine afforded 4a in 65% yield. The
final substitution reaction leading to L3 has been optimised
with respect to the choice of solvent (chloroform, THF,
MeCN) and base (no base, triethylamine, Hüning base, po-
tassium and caesium carbonates), as well as the reaction
time. The best experimental conditions turned out to be
with one equivalent of caesium carbonate in acetonitrile: a
24 h reaction time at room temperature afforded L3 in 30%
yield, while most of the starting macrocycle 4b could be re-
cycled and no secondary product was formed. Such products
form as soon as the reaction time is increased, while the ab-
sence of base or the presence of a different base leads to
smaller conversion rates. Ligand L4 was synthesised with the
purpose of confirming the pKa assignment for L3; it was ob-
tained in 43% yield by refluxing L3 overnight in methanol.
Potentiometric titration of a 1.02 mm aqueous solution of
L3 adjusted to pH 1.5 with hydrochloric acid, was performed
with NaOH 0.1m up to pH 12.5. Data could be satisfyingly
fitted to the following set of equations and protonation con-
stants:
Isolation and characterisation of the complexes: Complexes
with formulae [Ln
(OTf)3(L3)]·nMeCN·mH2O (Ln: Eu, Tb,
A
Gd, Lu; OTf: triflate; n=0.5 or 0; m=6–7; denoted LnL3
below) were readily obtained by mixing equivalent quanti-
ties of the lanthanide salt and the ligand in acetonitrile. To
assess the stability of the 1:1 complexes in water, solutions
containing equivalent amounts of ligand and lanthanide per-
chlorate, with a pH value set between 1.92 and 12.24, were
equilibrated until thermodynamic equilibrium was reached
and then their pH value was controlled and their absorption
spectra were measured. Factor analysis yielded five data
vectors and the spectra were fitted to Equations (6)–(8), in
which the first logK values are for EuIII complexes and the
second are for those with TbIII.
ðL3Þꢀ Hþ Ð ðHL3Þaq
ðHL3Þaq þ Hþ Ð ðH2L3Þþ
ðH2L3Þþaq þ Hþ Ð ðH3L3Þ
ðH3L3Þ aq þ Hþ Ð ðH4L3Þ
þ
½LnðL3ފ2þþOHꢀ Ð ½LnðL3ÞðOHފ
ð6Þ
logðKÞ ¼ 5:9ð4Þ=6:1ð3Þ
ðH4L3Þ aq þ Hþ Ð ðH5L3Þ
2þ
Ln3þaqþðL3Þꢀ Ð ½LnðL3ފ
ð7Þ
Assignment of pKaCarb to the carboxylic acid function of
the pendant arm is straightforward when the logKas are
compared to those obtained under the same experimental
conditions for ligand L4 and to literature values for dtma[32]
(Table 1). Only four protonation constants could be deter-
mined for L4. The first three have values close to the first
three logKa values of L3, while the fourth is close to the
logKa5 value for L3. The value of around 4.5 for L3 is missing
for L4, so it can confidently be assigned as arising from the
carboxylic acid function (logKaCarb). The large value of the
logðKÞ ¼ 17:0ð3Þ=16:3ð2Þ
3þ
½LnðL3ފ2þþHþ Ð ½LnðHL3ފ
ð8Þ
logðKÞ ¼ 26:0ð3Þ=24:1ð2Þ
Convergence was very good for EuIII (s=4.46”10ꢀ3) but
somewhat less reliable for TbIII (s=1.14”10ꢀ2). From these
data and the protonation constants of L3, the following pLn
values[24] could be calculated: pEu=13.4 and pTb=11.6. In
view of the less good adjustment obtained for TbIII, the last
value has to be considered as an estimate only and it is diffi-
cult to comment on the difference with respect to pEu. One
may, however, conclude that ligand L3 leads to reasonable
stability of the 1:1 complexes in solution. The pLn values
are comparable to those for dtma (for example, pGd=12.4
when calculated from the constants reported by Bianchi
et al.[32]), but smaller than the corresponding data for diethy-
lene-trinitrolopentaacetic acid (dtpa) (for example, pGd=
Table 1. Protonation constants for L3 and L4 as determined by potentio-
metric titration and comparison with corresponding data for dtma. Stan-
dard deviations are given within parentheses.
Ligand
L3
L4
dtma[a]
[a] Data taken from reference [32].
8680
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 8678 – 8687

Luminescent Lanthanide Complexes
FULL PAPER
19.8). A distribution diagram for EuIII taking the protona-
tion constants of L3 into account is shown in Figure 1.
Figure 2. Phosphorescence spectra of L3 and its lanthanide complexes in
water/glycerol (95:5) at pH 6.5 and 77 K upon excitation at 40320 cmꢀ1
(248 nm). The spectra are not drawn to scale; the scale of the trace for
the Tb complex has been expanded to show the triplet emission.
Figure 1. Distribution diagram for the system EuIII/L3 as calculated from
spectrophotometric titrations in the pH range 1.92–12.24 at 258C and
with I=0.1m (Me4NCl).
5
EuL3 excited at 466 nm (21460 cmꢀ1, D2 level) is displayed
in Figure 3. It is typical for a species (labelled I) with sym-
metry derived from a tetragonal arrangement.[25] The 0–0
transition appears as
a mostly symmetrical band at
Ligand-centred photophysical properties: The optical prop-
erties of ligand L3 are similar to those reported previously
for L1.[25] The absorption maximum occurs at 247 nm
(35090 cmꢀ1, loge=4.58) and no fluorescence is detected at
295 or 77 K in water/glycerol (95:5). Weak triplet-state emis-
sion is only seen at 77 K with a maximum at 420 nm
(23810 cmꢀ1) and
a 0-phonon component at 390 nm
(25640 cmꢀ1); non observation of singlet-state fluorescence
points to efficient non-radiative and/or intersystem crossing
(isc) processes taking place in this molecule. The triplet-
state energy is not much affected by the complexation; on
the other hand, the phosphorescence spectra display a
better intensity due to the heavy-atom effect of the lantha-
nide ion. All triplet-state emission spectra show vibrational
components with an average spacing at around 1300–
1400 cmꢀ1, which is typical of aromatic compounds. When
recorded in fluorescence mode, the emission spectrum of
LuL3 features a very weak band on the high-energy side of
the triplet-state emission peak (ꢁ370 nm, 27000 cmꢀ1),
which is tentatively assigned to singlet-state emission since
this feature disappears upon enforcement of a time delay
(Figure S2 in the Supporting Information). It is noteworthy
that in addition to the narrow metal-centred emission, trip-
let phosphorescence is also seen for EuIII and TbIII com-
plexes; in the latter case though, the metal-centred bands
have a much larger integrated intensity than the ligand
emission (Figure 2).
Figure 3. High-resolution emission spectrum of microcrystalline EuL3
7
measured at 295 K upon 5D2! F0 excitation (21460 cmꢀ1, 466 nm) and
recorded in time-resolved mode. The inset shows an enlargement of the
7
5D0! F0 transition.
17238 cmꢀ1 (full width at half height (fwhh)=25 cmꢀ1) and
has sizeable oscillator strength with an intensity relative to
5
7
the magnetic dipole transition D0! F1, I0/I1, equal to 0.15.
It sustains the usual red shift (10 cmꢀ1) when the tempera-
ture is lowered to 10 K. Its room-temperature energy can, in
principle, be estimated from the nephelauxetic effect gener-
ated by the coordinated donor groups[33] according to Equa-
tion (9), in which di is the nephelauxetic effect of the donor
group i and ni is the number of coordinated donor groups.
X
nðcmꢀ1, 295 KÞ ¼ 17 374þ
nidi
~
Metal-centred photophysical properties of microcrystalline
complexes: Information on the metal-ion environment has
been gained from high-resolution emission spectra of micro-
crystalline samples of the LnL3 complexes (Ln: Eu, Tb) at
295 and 10 K. The room-temperature emission spectrum of
i
With di
U
G
Chem. Eur. J. 2007, 13, 8678 – 8687
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8681

J.-C. G. Bünzli et al.
Table 2. Ligand-field sublevels (major species I, cmꢀ1) identified for mi-
crocrystalline EuL3 from emission spectra at 295 and 10 K. The origin of
the energy scale is set on the F0 level.
dination position is most probably occupied. The Eu
(5D0)
A
lifetime amounts to 0.94ꢂ0.02 ms at 295 K, whatever the
excitation mode is, through ligand or EuIII levels, so that oc-
cupancy by water can be ruled out; a plausible alternative is
coordination by the carboxylate group of the C5 pendant
7
Level Sublevels at Sublevels at Level
Sublevels at Sublevels at
295 K
10 K
295 K
10 K
7F1
301
302
7F4
2855
2864
arm. As a comparison, the Eu
(5D0) lifetime in EuL1 is small-
E
(cont.)
er, 0.60 ms, in line with the coordination of a water mole-
403
419
2962
3940
4105
4940
5035
5160
17238
2977
3956
4125
4953
5052
5176
17228
cule, as demonstrated by X-ray crystallography for TbL1.[25]
7F2
7F3
1009
1861
1936
2605
2713
2787
1014
1870
1942
2627
2724
n.a.[a]
7F5
7F6
Recalculation of n˜
A
above, scaled down for 9-coordination (ꢀ6%), and with di-
7F4
A
5D0
is in fair agreement with the observed value.
A high-resolution scan of the 0–0 excitation profile at
10 K with the analyzing wavelength set on the various com-
ponents of the transitions to F1 and F2 (Figure S3 in the
Supporting Information) reveals that the compound contains
[a] n.a.: not applicable.
7
7
with competition between the proton and TbIII for the car-
boxylate group and, possibly, with partial decomplexation of
the macrocyclic ligand. The hydration number drops to q=
0.64 at pH 7.5, a value very similar to the one found for
TbL1 (q=0.7 at a pH value close to neutral), a result point-
ing to an equilibrium between a non-hydrated and a mono-
hydrated species. With respect to the microcrystalline
a
minor species, II, with n˜
N
very similar, except for splitting of the D0! F1 transition,
and the lifetimes are alike (1.05ꢂ0.01 ms for II at 10 K, as
compared with 0.98ꢂ0.01 ms for I). Population analysis of
5
7
the D0! F1 transition carried out by comparing the emis-
sion spectra of the selectively excited species with the
broad-band excited spectrum[34] reveals that the minor spe-
cies accounts for only 3% of the total EuIII population. Sev-
eral hypotheses are possible, such as the presence of an
eight-coordinate species devoid of carboxylate coordination
or of an isomer with a different macrocycle conformation;
coordination of water is again excluded in view of the long
sample, the Eu
to the one reported for EuL1, for which a hydration number
of q=0.7 has been found.[25] Both the Eu(5D0) and Tb(5D4)
A
U
ACHTREUNG
lifetimes are almost temperature independent between
room temperature and 77 K (Table 3), which indicates the
absence of temperature-dependent quenching mechanisms.
The quantum yields (Table 3) are smaller than for LnL1
complexes under the same experimental conditions. The
ligand sensitisation efficiency can be estimated for EuL3
from Equation (10).[35]
Eu
(5D0) lifetime recorded for II.
G
Analysis of the ligand-field splitting of species I (Table 2)
confirms C4-derived site symmetry for the metal ion. The 0–
0 transition has sizeable intensity, as is usually the case for
7
Cn or Cnv symmetry. The F1 level is split into two compo-
QLn ¼ hsens ꢃ QLLnn ¼ hsens ꢃ ðtobs=tradÞ,
L
nents (102 cmꢀ1, allowed magnetic dipole transitions to A
and E in the C4 symmetry point group[34]), while the 7F2
level appears to be unsplit although a weak shoulder can be
found on the low-energy side of the corresponding transition
(two transitions allowed in C4 symmetry) which has I2/I1 =
ð10Þ
with 1=trad ¼ 14:65 ꢃ ðItot=IMDÞ ꢃ n3 ½sꢀ1
Š
In our case, Itot/IMD =5.0 and n=1.332, so trad =5.8 ms, QLLnn
=
10.2% and hsens ꢁ19%. This figure is relatively low, in line
with the observation of substantial triplet emission for this
complex; in addition, the intrinsic quantum yield remains
small because of water coordination in solution. Quantum
yields are pH dependent and this is illustrated in Figure S5
in the Supporting Information for TbL3, the quantum yield
5
7
2.40. Finally, D0! F4 is comprised of five components, as
predicted by the symmetry-related selection rules.
The TbIII spectra are less informative (Figure S4 in the
Supporting Information); they feature the usual predomi-
5
7
nance of the D4! F5 transition. As for the EuIII complex,
the luminescence decays are
perfectly single exponential
functions and the Tb
(5D4) life-
A
Table 3. Lifetimes of the Eu
(Ln: Eu, Tb) in water (or water/glycerol (95:5) for measurements at 77 K) and acetonitrile.
G
G
time amounts to 1.30 ꢂ
0.02 ms at 295 K and 1.35ꢂ
0.02 ms at 10 K. Data recorded
at pH values of 3 and 7.5 for
TbL3 confirm the hypothesis of
an interaction between the
metal ion and the C5 carboxyl-
ic pendant: the hydration
number is equal to 2 at acidic
pH values, which is consistent
[a,b]
[a,c]
Complex
EuL3
Solvent
pH value
6
G
t
U
qLn
qLn
QLn [%][d]
L
[e]
[e]
water
MeCN
water
0.67ꢂ0.04
0.68ꢂ0.02
–
–
–
–
1.9ꢂ0.2
[e]
[e]
[e]
–
–
TbL3
3
–
2.1
2.0
[e]
[e]
[e]
6
7.5
1.89ꢂ0.04
–
–
3.4ꢂ0.3
[e]
–
0.56
0.72
–
[a] From lifetime measurements in water and deuterated water. [b] Estimated according to q=5
ACHTREUNG
with Dk=1/tH Oꢀl/tD O. [c] Estimated according to q=4.2 Dk.[53] [d] Absolute quantum yield in aerated water
2
2
upon ligand excitation; average of measurements with respect to two standards (see the Experimental Sec-
tion). [e] Not determined.
8682
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 8678 – 8687

Luminescent Lanthanide Complexes
FULL PAPER
of which steadily increases for pH 2–6.5 and then sharply
decreases for pH values larger than 8 due to the formation
to some chemical interaction taking place between Cy5 and
the lanthanide complex. The corresponding KISV constant
for the TbIII system is about 2.5 times smaller, in line with
the decreased overlap between the donor-emission and the
acceptor-absorption spectra. The collected data allow calcu-
lation of the yields of the energy-transfer processes. From
Figure 5 (EuIII) and Figure S8 in the Supporting Information
(TbIII), one sees that these yields increase sharply up to
about one equivalent of added Cy5 and then stabilise. The
maximum yield is in the range 90–95% for EuL3, while it re-
mains at about 75–80% for the TbIII chelate.
of hydroxo species. The drop in the QTb value between
L
pH 6.5 and 7.5 remains unexplained; it is essentially linked
to a weaker emission, with the absorbance remaining rela-
tively constant over this pH range, and may be related to a
conformational rearrangement due to acid–base equilibria
involving the C5 pendant as such a behaviour is not ob-
served for TbL1. In that case, the quantum yield increases
from pH 5 to 7 and stays constant up to pH 8.5, before de-
creasing at more basic pH values.
In view of the interest for LRET experiments,[36] we have
investigated the variation in the luminescence intensity and
lifetime of 2”10ꢀ6 m LnL3 (Ln: Eu, Tb) complexes in water
upon addition of Cy5. This fluorescent dye absorbs at
649 nm with a large molar absorption coefficient (2.5”
105 mꢀ1 cmꢀ1) and emits at 670 nm. Its excited state can,
therefore, be populated by transfer from the lanthanide ion.
Figure 4 illustrates the transfer from EuL3 while correspond-
ing data for TbL3 are shown in Figure S6 in the Supporting
Information. As expected from the overlap between the Cy5
absorption spectrum and the Ln emission spectra, transfer is
more effective with EuL3 than with TbL3.
Figure 5. Yield of the EuL3-to-Cy5 energy transfer at pH 6.5 calculated
from both emission-intensity and lifetime data.
Influence of TbL3 on cell viability: To assess the potential
use of the LnL3 complexes for in vitro analyses, the cytotox-
icity of the TbIII complex has been determined with respect
to several cell lines, both malignant (5D10 mouse hybrido-
ma, Jurkat human T leukaemia, MCF-7 human breast carci-
noma) and non-malignant (HaCat human keratinocyte). In
a first step, the influence of the TbIII complex on cell prolif-
eration was determined by using the WST-1 test with meas-
urements after 0.5, 1, 2, 3 and 4 h. As shown at the top of
Figure 6 for the MCF-7 and Jurkat cell lines, the absorbance
difference (A450ꢀA650) increases with time, thereby reflecting
the proliferation of the cells. No significant difference can
be observed between the proliferation of the cells in the ab-
sence or presence of the TbIII complex up to 500 mm for the
MCF-7 cell line. The same was true for the HaCat and 5D10
cell lines (Figure S9 in the Supporting Information). On the
other hand, a negative influence on cell proliferation can be
observed for the Jurkat cell line for concentrations of the
complex >125 mm.
These observations were confirmed by the viability of the
cells calculated after 24 h of incubation in the presence of
various concentrations of TbL3. Data for the MCF-7 and
Jurkat cells are displayed at the bottom of Figure 6. Where-
as a negative influence on cell viability was again observed
for the Jurkat cell line when the concentration of TbL3 was
larger than 125 mm, no negative effect could be detected for
the other three cell lines up to 500 mm. This conclusion is
further backed by the lactase dehydrogenase (LDH) assay
of cellular membrane damage, the results of which are re-
ported in Table 4. This assay tests the leakage of LDH,
Figure 4. Variation in the emission spectra (uncorrected; excited at
40320 cmꢀ1, 248 nm) of EuL3 in water at pH 6.5 upon addition of Cy5
(0–3.5 equiv).
The lifetimes of the Ln excited states decrease in parallel
with the decrease in metal-centred luminescence, from 0.6
(no Cy5) to 0.06 ms (3.5 equiv of Cy5) for Eu
ACHTREUNG
1.7 (no Cy5) to 0.16 ms (3 equiv of Cy5) for Tb
AHCTREUNG
sis of the EuIII data in terms of Stern–Volmer plots by using
variations in both the EuIII lifetimes and emission intensities
is presented in Figure S7 in the Supporting Information.
Two concentration domains are evidenced: the first one, ap-
proximately up to the addition of 1.5 equivalents Cy5, corre-
sponds to a Stern–Volmer constant of KISV =(3.5ꢂ0.1)”
107 mꢀ1 sꢀ1 (from emission intensities), while the quenching is
more efficient above this ratio, with a doubling in the KISV
value. On the other hand, the opposite is observed from life-
time data, with the KtSV value being smaller by 1.75 in the
high-concentration domain. This behaviour probably points
Chem. Eur. J. 2007, 13, 8678 – 8687
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8683

J.-C. G. Bünzli et al.
plications of the LnL3 macro-
cyclic chelates, which will be
tested in future work.
Experimental Section
Materials and methods: Reagents and
solvents were purchased from Al-
drich, Fluka Ag or Acros; cyclen was
from Strem. Solvents were dried by
distillation over CaH2 (CH2Cl2,
MeCN), Na/benzophenone (THF) or
KOH (NEt3).[37] Lanthanide tri-
flates[38] and perchlorates[39] were pre-
pared in the usual way, from analyti-
cal-grade triflic or perchloric acids
and high-purity lanthanide oxides
(Rhodia). The Ln content was deter-
mined by complexometric titration
with H2Na2(edta) in presence of xyle-
G
nol orange and urotropine (edta: eth-
ylenediaminetetraacetate).[40]
Figure 6. Top: WST-1 proliferation test of cells in the absence or presence of different concentrations of TbL3.
Bottom: Cell viability values (%), obtained by using the same test, for different concentrations of TbL3 after
24 h incubation at 378C.
1H NMR spectra were recorded on a
Bruker Avance DRX 400 spectrome-
ter. IR spectra were recorded on a
Perkin–Elmer Spectrum One FT
spectrometer equipped with an ATR
Table 4. Cell cytotoxicity values [%] measured by using the LDH assay[a]
accessory. ESI-MS spectra were measured on a Finningan SSQ-710C in-
strument. Elemental analyses were performed by Dr H. Eder from the
University of Geneva.
after incubation of the cells with different concentrations of TbL3.
A
Jurkat
MCF-7
HaCat
50
0.5ꢂ4.6
4.0ꢂ5.7
11.3ꢂ5.0
9.9ꢂ 3.8
ꢀ5.8ꢂ7.5
ꢀ1.9ꢂ9.6
1.1ꢂ1.6
ꢀ11.0ꢂ3.9
ꢀ5.9ꢂ0.1
ꢀ5.5ꢂ0.9
ꢀ5.6ꢂ1.7
Potentiometric titrations: Ligands were titrated in a thermostated (25.0ꢂ
0.18C) glass-jacketed vessel under an Ar atmosphere. The ionic strength
was fixed with Me4NCl (m=0.1m). The solution was acidified to pHꢁ1.5
with 37% HCl 30 min before titration. Titrations were carried out by
using an automatic Metrohm Titrino 736 GP potentiometer (resolution
0.1 mV, accuracy 0.2 mV) linked to an IBM PS/2 computer and by using
constant volume addition. An automatic burette (Metrohm 6.3013.210,
10 mL, accuracy 0.03 mL) was used along with a Metrohm 6.0238.000
glass electrode. A standard base solution (NaOH 0.1 m) was added inside
the solution through a capillary tip attached to the automatic burette.
The data (200 points, drift<1 mVminꢀ1) were mathematically treated in
the program HYPERQUAD2000[41] by using a Marquardt algorithm, and
the distribution of species was calculated by using the program HYSS2.
Calibration of the pH meter and the electrode system was performed
prior to each measurement by using a standardised HCl solution at
25.08C: the latter (10 mL) was titrated with a standardised 0.1 m NaOH
solution and the electrode potential readings were converted into pH
values. The ion product of water (pKw =13.91) and the electrode poten-
tial were refined by using the program Scientist by Micromath (Ver-
sion 2.0).
125
250
500
2.6ꢂ6.3
[a] At 228C and after 30 min of incubation.
which increases if the cell membrane is damaged, that is, if
the cell is destroyed. The measured LDH leakage 30 min
after incubation with concentrations of TbL3 ꢄ250 mm was
around 10% for Jurkat cells, while it is close to zero or even
slightly negative for the two other cell lines tested, MCF-7
and HaCat. This result means that the latter cell lines do
not sustain membrane damage under these experimental
conditions.
Conclusion
UV/Visible spectra were measured in 1-, 0.2- or 0.1 cm quartz cuvettes
on a Perkin–Elmer Lambda 900 spectrometer and the data were analysed
with the Specfit software.[42] Speciation of the complexes was determined
by batch titration of a 3.3”10ꢀ5 m ligand solution containing lanthanide
perchlorate (Ln: Eu, Tb; 1 equiv). 30–40 samples were prepared with pH
values ranging from 1.92–12.24; the ionic strength was kept constant with
0.1m Me4NCl. In order to reach thermodynamic equilibrium, samples
were kept at 558C for one week followed by one day at 258C; the pH
values and absorption spectra were then measured at this temperature.
The data were analysed with the Specfit software.
Modification of L1 by replacing one chromophoric pendant
arm with a C5 alkyl chain bearing a carboxylic function in L3
does not influence the stability of the resulting lanthanide
chelate LnL3. On the other hand, the photophysical proper-
ties are affected, but the EuIII and TbIII complexes are suffi-
ciently emissive to act as potential luminescent stains. In
particular, an efficient LRET phenomenon has been ob-
served and quantified with Cy5. Moreover, interaction of
the C5 pendant with the metal ion is pH dependent and
modulates the photophysical properties. The latter, com-
bined with the non-toxicity for several cell lines that has
been demonstrated for TbL3, points to potential in vitro ap-
Luminescence spectra were recorded on a Fluorolog FL-3-22 spectrome-
ter from Spex–Jobin–Yvon–Horiba; quartz cells with optical paths of 1,
0.2 or 0.1 cm were used for room-temperature spectra, while 77 K meas-
urements were carried out on samples put into quartz capillaries with
3 mm diameter. Quantum yields were determined in aerated water with
respect to several references: quinine sulphate in 0.5m sulphuric acid
8684
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 8678 – 8687

Luminescent Lanthanide Complexes
FULL PAPER
(QS, lexc =247 nm, Q=54.6%[43]) and cresyl violet in methanol (CV,
lexc =258 nm, Q=54%[44]) for the Tb complex and rhodamine 101 in eth-
anol (R101, lexc =255 nm, Q=100%[45]) and CV for the Eu complex. The
absorbance of the samples and references was usually kept under 0.1.
The complex concentration was 2”10ꢀ6 m and Equation (11) was used,[46]
in which A is the absorbance, n2D0 is the refractive index (1.338, 1.361,
1.328 and 1.332 for QS, R101, CV and the Ln samples, respectively), and
S is the integrated, corrected emission area.
drying under vacuum: 1H NMR (400 MHz, CDCl3, Me4Si): d=2.8 (s,
3
10H), 2.9 (s, 6H), 3.3 (s, 4H), 3.3 (s, 2H), 4.5 (d, 4H, J=5.4 Hz), 4.6 (d,
2H, 3J=5.5 Hz), 7.4–7.9 (m, 15H), 8.0 ppm (s, 1H); IR (ATR): n˜ =3290
(w, br, nNꢀH), 3060 (w, nCꢀH(ar)), 2924, 2831 (w, nCH), 1695 (s, nC O(phen)),
=
1652 (s, nC O(am)), 1596 (m, nC C), 1579 (w, nC C), 1525 (s, nC C, nC O(am)),
=
=
=
=
=
1448 (s, dCꢀH), 1221 (s, dCꢀO), 755 (s, dCꢀH(ar)), 688 cmꢀ1 (s, dCꢀH(ar)); ESI-
MS: m/z: 698 [4b+H]+, 349 [4b+2H]2+
.
1,4,7-Tris[(N-4-phenylacetyl)carbamoylmethyl]-10-[N-(carboxypentyl)car-
bamoylmethyl]-1,4,7,10-tetraazacyclododecane (L3): Compound 4b
(1.383 g, 2 mmol) was dissolved with CsCO3 (0.646 g, 2 mmol) in anhy-
drous acetonitrile (200 mL) and the mixture was stirred under a control-
led atmosphere for 30 min. A solution of 1 (0.474 g, 2 mmol) in anhy-
drous acetonitrile (100 mL) was added and the mixture was stirred at
room temperature for 3 d. A white precipitate was filtered from the re-
sulting yellow solution and the solvent was evaporated to leave a yellow
solid which was purified on a cationic Dowex 50W X 8 resin followed by
an anionic chloride resin (Amberlite): Yield=0.516 g (30%); 1H NMR
(400 MHz, CDCl3, Me4Si): d=1.1 (m, 2H), 1.2 (m, 2H), 1.4 (m, 2H), 2.1
ꢀ
ꢁ
ꢀ
ꢁ
2
1ꢀ10ꢀAðref:Þ
n2D0ðsampleÞ
n2D0ðref:Þ
SðsampleÞ
Sðref:Þ
sample
abs
ref:
abs
ð11Þ
Q
¼
ꢃ
ꢃ
ꢃ Q
1ꢀ10ꢀAðsampleÞ
N-hydroxysuccinimide-activated Cy5 (abbreviated Cy5) was bought from
Amersham and used for the energy-transfer experiments as received.
High-resolution emission spectra and lifetimes were measured on a previ-
ously described instrumental setup.[47]
Ligand synthesis (Scheme 2)
3
(t, 2H, J=5.5 Hz), 2.8–3.0 (m, 18H), 3.4 (s, 2H), 3.5 (s, 4H), 3.5 (s, 2H),
Carboxylic acid substituent 1: This was synthesised according to the
method of Yamada et al.[48] 6-Aminocaproic acid (13.12 g, 0.10 mol) was
dissolved in water (100 mL). After being cooled at 08C, bromoacetyl bro-
mide (9.55 mL, 0.11 mol) was added dropwise, while the pH value was
maintained at 7 with 4m NaOH (a total of 50 mL was needed). The mix-
ture was stirred for 15 min and the pH value was adjusted to 2 with aque-
ous HBr (47%). The resulting solution was concentrated to a volume of
30 mL and extracted with ethyl ether. The combined organic phases were
dried over MgSO4, the solvent was evaporated to dryness, diethyl ether
(50 mL) was added and the solution was stored at ꢀ188C for 2 d. The
white precipitate was filtered and dried (2.52 g, 0.01 mol, yield=10%):
m.p. 728C; 1H NMR (400 MHz, CDCl3, Me4Si): d=1.5–1.8 (m, 6H), 2.3
4.4 (d, 2H, 3J=5.5 Hz), 4.5 (d, 4H, 3J=5.3 Hz), 7.4–7.8 ppm (m, 15H);
13C NMR (800 MHz, MeOD) d=196.16 (Ph C=O), 173.78 (CO2H),
ꢀ
ꢀ
172.59 (NH C=O), 136.70, 136.65 (Car), 135.51, 135.44, 135.63 (CHar),
133.70 (CHar), 131.08 (CHar), 130.48, 130.43, 130.25 (CHar), 129.74,
129.58, 129.49 (CHar), 63.09 (COCH2N), 58.83, 58.74 (COCH2N), 58.74
(COCH2), 50.00, 49.83, 49.78 and 49.69 (CH2-CH2), 47.85 (COCH2NH),
40.72, 40.23 (CH2), 36.46, 35.85 (CH2), 30.76, 30.48 (CH2), 28.05, 28.01
(CH2), 26.52, 26.44 ppm (CH2); IR (ATR): n˜ =3300 (w, br, nOꢀH, nNꢀH),
3062 (w, nCꢀH(ar)), 2933, 2854, 2831 (w, nCꢀH), 1694 (m, nC O(Phen)), 1657 (s,
=
nC O(ac)), 1652 (s, nC O(am)), 1596 (m, nC C(ar)), 1579 (w, nC C(ar)), 1544 (sh,
=
=
=
=
nC O(ac)), 1530 (s, nC C(ar), nC O(am)), 1448 (m, dCꢀH), 1222 (s, dCꢀO), 755,
=
=
=
688 cmꢀ1 (s, dCꢀH(ar)); ESI-MS: m/z: 869 [L3+H]+, 435 [L3+2H]2+; ele-
mental analysis: calcd (%) for L3·2H2O·HCl (C46H65ClN8O11): C 58.68, H
6.96, N 11.90; found: C 58.73, H 7.01, N 12.06.
3
(t, 2H, J=14.4 Hz), 3.3 (m, 2H), 3.9 (s, 2H), 6.7 ppm (br, 1H).
1,4,7,10-Tetraazatricyclo[5.5.1.0]tridecane (2a) and 1-formyl-1,4,7,10-tet-
N
raazacyclododecane (2b): Cyclen (5 g, 29 mmol) and N,N-dimethylforma-
mide dimethylacetal (3.92 mL, 30.5 mmol) were dissolved in toluene
(70 mL) and put into a flask equipped with a Dean–Stark condenser
under a nitrogen atmosphere. The methanol/toluene azeotrope formed
was distilled until complete elimination of toluene had occurred. The re-
sulting yellow oil (2a) was dried overnight at 708C. It was then cooled to
08C and a water/methanol mixture (1:1 v/v; 25 mL) was added dropwise.
The mixture was warmed to room temperature and stirred for 24 h. The
solvents were evaporated and the residue was redissolved in a minimum
amount of acetonitrile, which was subsequently evaporated; this proce-
dure was repeated twice to completely eliminate water and to yield 2b as
1,4,7-Tris[(N-4-phenylacetyl)carbamoylmethyl]-10-[N-(methanopentyl)car-
bamoylmethyl]-1,4,7,10-tetraazacyclododecane (L4): Ligand L3 (0.1 g,
0.11 mmol) was dissolved in anhydrous ethanol (40 mL) and the solution
was refluxed overnight. The solvent was evaporated and the product was
recrystallised in hot acetonitrile (0.042 g, 43% yield): 1H NMR
(400 MHz, CDCl3, Me4Si): d=1.1 (m, 2H), 1.2 (m, 2H), 1.3 (m, 2H), 2.1
3
(t, 2H, J=5.5 Hz), 2.8–3.0 (m, 18H), 3.4 (s, 2H), 3.5 (s, 4H), 3.5 (s, 2H),
3
3
3.67 (s, 3H), 4.4 (d, 2H, J=5.5 Hz), 4.5 (d, 4H, J=5.3 Hz), 7.4–7.8 ppm
(m, 15H); ESI-MS: m/z: 884 [L4+H]+, 443 [L4+2H)2+; elemental analy-
sis: calcd (%) for L4·H2O·HCl (C47H65ClN8O10): C 60.21, H 6.99, N 11.95;
found: C 60.40, H 7.03, N 11.99.
1
a white solid (5.58 g, 95% yield): H NMR (400 MHz, CDCl3, Me4Si): d=
2.6–2.7 (m, 8H), 3.4 (dt, 8H, 2J=20.2, 3J=5.5 Hz), 8.0 ppm (s, 1H); IR
Synthesis of the complexes: Ln
with ligand (1 equiv) in acetonitrile. The solution was refluxed for 3 d
and the [Ln
(OTf)3L3]·nMeCN·mH2O complexes were then precipitated
A
(ATR): n˜ =3312 (w, nNꢀH), 2811 (m, nCꢀH), 1656 (s, nC O), 1444 (m, dCH),
=
1227 cmꢀ1 (m, dC O).
AHCTREUNG
=
by addition of CH2Cl2 and dried under vacuum. Elemental analyses of
the products are reported in Table 5.
1-Formyl-4,7,10-tris[(N-4-phenylacetyl)carbamoylmethyl]-1,4,7,10-tetra-
azacyclododecane (4a): Compound 2b (3.78 g, 189 mmol) was dissolved
in THF(100 mL), distilled triethylamine (8.16 mL, 586 mmol) was added
and the solution was stirred for 1 h under an inert atmosphere. A solu-
tion of 3 (15.0 g, 585 mmol) in anhydrous THF(100 mL) was added and
the mixture was stirred at room temperature for 2 d. A white precipitate
of triethylbromide was filtered off and the solution was concentrated.
The beige product 4a (9.37 g, 65% yield) precipitated upon addition of
diethyl ether: 1H NMR (400 MHz, CDCl3, Me4Si): d=2.8–3.3 (m, 16H),
3.7 (s, 1H), 4.6 (d, 2H, 3J=4.4 Hz), 4.7 (d, 2H, 3J=5 Hz), 4.7 (d, 2H,
3J=5.2 Hz), 7.3–8.0 (m, 15H), 8.1 ppm (s, 1H); ESI-MS: m/z: 726
[4a+H]+, 748 [4a+Na]+.
Cell culture: Cell viability was tested on the following cell lines: mouse
hybridoma 5D10, human T leukaemia Jurkat (ATCC TIB152), human
breast adenocarcinoma MCF-7 (ATCC HTB-22) and non-malignant epi-
thelial HaCat (human keratinocytes). Cells were cultivated in 75 cm2 cul-
Table 5. Elemental
analyses
for
the
isolated
[Ln(OTf)3
C
Ln
Eu
n
m
Formula
C [%]
H [%]
N [%]
1,4,7-Tris[(N-4-phenylacetyl)carbamoylmethyl]-1,4,7,10-tetraazacyclodode-
cane (4b): Compound 4a (7.41 g, 102 mmol) was dissolved in a mixture
of acetone (75 mL) and water (40 mL) and cooled to 08C before concen-
trated HCl (37%; 45 mL) was added dropwise. The resulting solution
became orange and was stirred for 4 d at room temperature. A saturated
solution of NaHCO3 was added until the pH value reached 7 (ꢁ500 mL)
and the compound was extracted with CH2Cl2 (3”400 mL). The com-
bined organic phases were dried over MgSO4 and filtered, then the sol-
vent was evaporated. An orange oil was collected which solidified after
0
6
C51H75F9N9O24S3Eu
37.52
4.79
6.95
(37.34)
(4.60)
4.40
(7.11)
7.55
Tb 0.5 6.5
C
100H149F18N17O49S6Tb2 36.99
(37.24)
37.33
(37.21)
36.45
(4.66)
4.75
(4.59)
4.76
(7.38)
6.96
(7.08)
6.54
Gd
Lu
0
0
6
7
C49H72F9N8O24S3Gd
C49H74F9N8O25S3Lu
(36.39)
(4.61)
(6.93)
Chem. Eur. J. 2007, 13, 8678 – 8687
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8685

J.-C. G. Bünzli et al.
ture flasks by using RPMI 1640 medium supplemented with 10% foetal
calf serum (FCS), 2 mm l-glutamine, 1 mm sodium pyruvate, 1% non-es-
sential amino acids and 1% sodium [4-(2-hydroxyethyl)piperazine-1-
ethanesulfonate monosodium salt (HEPES; all from Gibco Cell Culture,
Invitrogen, Basel, Switzerland). Cultures were maintained at 378C under
a 5% CO2 and 95% air atmosphere. The growth medium was changed
every other day until the time of use of the cells. The cell density and via-
bility, defined as the ratio of the number of viable cells over the total
number of cells, of the cultures were determined by trypan blue staining
and use of a Neubauer improved hemacytometer (Blau Brand, Wer-
theim, Germany). Prior to each viability test, the cells were harvested
and diluted at a density of 7.5”105 cellsmLꢀ1. The cell suspension was
seeded into 96-well plates at a concentration of 100 mL per well and incu-
bated overnight at 378C and under 5% CO2 prior to addition of the
TbL3 complex and the WST-1 or LDH reagent.
Acknowledgements
This research is supported by the Swiss National Science Foundation. We
are indebted to Dr. T. Peymann for preliminary work on the ligand syn-
thesis and to Dr. Anjum Dadabhoy for valuable advice. We thank Prof.
F. Wurm (École Polytechnique FØdØrale de Lausanne, Switzerland) for
the use of his Spectra MAX340 ELISA reader and Cytofluor series 4000
fluorimeter. The 5D10 hybridoma cell line was a gift from the Biomedical
Research Institute “Dr. L. Willems Instituut” (University of Hasselt, Bel-
gium).
[1] H. Stetter, W. Frank, Angew. Chem. 1976, 88, 760; Angew. Chem.
Int. Ed. Engl. 1976, 15, 786.
[2] J. F. Desreux, Bull. Acad. R. Belg. 1979, 64, 814–814.
[3] C. A. Chang, L. C. Francesconi, M. F. Malley, K. Kumar, J. Z. Gou-
goutas, M. F. Tweedle, D. W. Lee, L. J. Wilson, Inorg. Chem. 1993,
32, 3501–3508.
[4] A. Bianchi, L. Calabi, F. Corana, S. Fontana, P. Losi, A. Maiocchi,
L. Paleari, B. Valtancoli, Coord. Chem. Rev. 2000, 204, 309–393.
[5] J. F. Desreux, P. P. Barthelemy, Nucl. Med. Biol. 1988, 15, 9–15.
[6] P. Caravan, Chem. Soc. Rev. 2006, 35, 512–523.
[7] J.-C. G. Bünzli, C. Piguet, Chem. Soc. Rev. 2005, 34, 1048–1077.
[8] I. Hemmilꢁ, V. M. Mukkala, Crit. Rev. Clin. Lab. Sci. 2001, 38, 441–
519.
[9] S. Faulkner, J. L. Matthews in Comprehensive Coordination Chemis-
try II, Vol. 9 (Ed.: M. D. Ward), Elsevier Pergamon, Amsterdam,
2004, pp. 913–944.
[10] M. Li, P. R. Selvin, J. Am. Chem. Soc. 1995, 117, 8132–8138.
[11] D. Parker, J. A. G. Williams, J. Chem. Soc. Perkin Trans. 2 1996,
1581–1586.
[12] A. Beeby, D. Parker, J. A. G. Williams, J. Chem. Soc. Perkin Trans. 2
1996, 1565–1579.
[13] D. Parker, Chem. Soc. Rev. 2004, 33, 156–165.
[14] D. Parker, J. A. G. Williams in Responsive luminescentlanhtanide
complexes (Eds.: A. Sigel, H. Sigel), Metal Ions in Biological Sys-
tems, Vol. 40, Marcel Dekker, New York, 2003.
[15] D. Parker, R. S. Dickins, H. Puschmann, C. Crossland, J. A. K.
Howard, Chem. Rev. 2002, 102, 1977–2010.
WST-1 cell viability test: Cell-proliferation reagent WST-1 (Roche, Ger-
many) is a slightly red tetrazolium salt that is reduced only in living met-
abolically active cell mitochondria to yield a dark-red formazan dye
which can be quantified spectrophotometrically.[49] Cells were seeded in a
96-well tissue-culture microplate at a concentration of 7.5”104 cells per
well in culture medium (100 mL) and then incubated overnight at 378C
and under 5% CO2. The TbL3 complex was dissolved in fresh RPMI
medium at 378C and at a concentration of 500 mm. The medium was re-
moved from the cell cultures and the TbL3 chelate was added (100 mL
per well; final concentrations: 500, 250, 125 and 50 mm). WST-1 reagent
(10 mL) was added to each well and the plate was shaken for 1 min on a
microtiter-plate shaker (450 rpm). The plate was further incubated at
378C and under 5% CO2 and the absorbance of the formazan product
was measured at 450 nm with an ELISA reader (Spectra MAX 340, Mo-
lecular Devices, Sunnyvale, CA, USA). The cell viability was calculated
from the absorbance values according to Equation (12), in which
(A450ꢀA650) is the difference between the absorbances at 450 and 650 nm
for the cells put into contact with TbL3 (exp) or with the medium only
(medium). Results are expressed as averages over four nominally identi-
cal measurements.
ðA450ꢀA650Þexp
ð12Þ
viabilityWST ½%Š ¼
ꢃ 100
ðA450ꢀA650Þmedium
LDH cell toxicity test: Lactate dehydrogenase (LDH) leakage was mea-
sured by using the CytoTox 96 non-radioactive cytotoxicity assay kit
(Promega Corporation, Madison, WI, USA).[50,51] Cells were seeded in a
96-well tissue-culture microplate at a concentration of 7.5”104 cells per
well in culture medium (100 mL) and then incubated overnight at 378C
and under 5% CO2. The TbL3 complex was dissolved in fresh RPMI
medium at 378C and at a concentration of 500 mm. The medium was re-
moved from the cell cultures and the TbL3 complex was added (100 mL
per well; final concentrations: 500, 250, 125 and 50 mm). The substrate
mixture solution (50 mL) was then added. The plates were incubated at
228C for 30 min. The LDH released from the cells catalyses the conver-
sion of resazurin into resorufin. After incubation of the mixture, the fluo-
rescence values were measured with an excitation wavelength of 530 nm
and an emission wavelength of 590 nm in a microplate fluorometer (Cy-
tofluor series 4000, Perceptive Biosystems, Framingham, MA, USA). The
results are expressed as averages of four nominally identical measure-
ments. The maximum LDH release after cell lysis with Triton X-100
(9 vol%) and the spontaneously released LDH in cells not in contact
with the TbL3 complex were also measured. The average fluorescence
values of the culture-medium background were subtracted from all fluo-
rescence values of experimental wells. Cytotoxicity caused by the TbL3
complex was calculated according to Equation (13).
[16] D. Parker, Coord. Chem. Rev. 2000, 205, 109–130.
[17] L. Armelao, G. Bottaro, S. Quici, M. Cavazzani, M. C. Raffo, F. Bar-
igelletti, G. Accorsi, Chem. Commun., 2007, 1911–2913.
[18] S. Faulkner, S. J. A. Pope, B. P. Burton-Pye, Appl. Spectrosc. Rev.
2005, 40, 1–31.
[19] S. Pandya, J. H. Yu, D. Parker, Dalton Trans. 2006, 2757–2766.
[20] J.-C. G. Bünzli, Acc. Chem. Res. 2006, 39, 53–61.
[21] “Lanthanide Near-Infrared Luminescence in Molecular Probes and
Devices”: S. Comby, J.-C. G. Bünzli in Handbook on the Physics and
Chemistry of Rare Earths, Vol. 37 (Eds.: K. Gschneidner, Jr., J.-C. G.
Bünzli, V. Pecharsky), Elsevier Science, Amsterdam, 2007, Chap-
ter 235.
[22] G. Zucchi, R. Scopelliti, P. A. Pittet, J.-C. G. Bünzli, R. D. Rogers, J.
Chem. Soc. Dalton Trans. 1999, 931–938.
[23] G. Zucchi, R. Scopelliti, J.-C. G. Bünzli, J. Chem. Soc. Dalton Trans.
2001, 1975–1985.
[24] K. N. Raymond, G. Müller, F. Matzanke, Top. Curr. Chem. 1984,
123, 49–102.
[25] G. Zucchi, A.-C. Ferrand, R. Scopelliti, J.-C. G. Bünzli, Inorg. Chem.
2002, 41, 2459–2465.
[26] I. Hemmilꢁ, T. Stꢂhlberg, P. Mottram, Bioanalytical Applications of
Labelling Technologies, Wallac Oy, Turku, 1995.
IexpꢀImedium
[27] P. R. Selvin, Annu. Rev. Biophys. Biomol. Struct. 2002, 31, 275–302.
[28] H. Bazin, E. Trinquet, G. Mathis, Rev. Mol. Biotechnol. 2002, 82,
233–250.
cytotoxicityLDH ½%Š ¼
ꢃ 100
ð13Þ
Iexp,maxꢀImedium
Iexp is the fluorescence value of the cells in contact with different concen-
trations of the TbL3 complex, Imedium is the fluorescence value for the
spontaneous release of LDH by cells not in contact with the TbL3 com-
plex and Iexp,max is the fluorescence value for lysed cells with maximum
LDH release.
[29] C. Glogard, R. Hovland, S. L. Fossheim, A. J. Aasen, J. Klaveness, J.
Chem. Soc. Perkin Trans. 2 2000, 1047–1052.
[30] J. H. Yu, D. Parker, Eur. J. Org. Chem. 2005, 4249–4252.
[31] J. Platzek, P. Blaszkiewicz, H. Gries, P. Luger, G. Michl, A. Muller-
Fahrnow, B. Raduchel, D. Sulzle, Inorg. Chem. 1997, 36, 6086–6093.
8686
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 8678 – 8687

Luminescent Lanthanide Complexes
FULL PAPER
[32] A. Bianchi, L. Calabi, C. Giorgi, P. Losi, P. Mariani, Paoli, P. Rossi,
B. Valtancoli, M. Virtuani, J. Chem. Soc. Dalton Trans. 2000, 697–
705.
[33] S. T. Frey, W. deW. Horrocks, Jr., Inorg. Chim. Acta 1995, 229, 383–
390.
[34] J.-C. G. Bünzli in Lanthanide Probes in Life, Chemical and Earth
Sciences: Theory and Practice (Eds.: J.-C. G. Bünzli, G. R. Choppin),
Elsevier Science, Amsterdam, 1989, pp. 219–293.
[35] M. H. V. Werts, R. T. F. Jukes, J. W. Verhoeven, Phys. Chem. Chem.
Phys. 2002, 4, 1542–1548.
[36] “Lanthanide Chelates as Luminescent Labels in Biomedical Analy-
ses”: K. Nishioka, K. Fukui, K. Matsumoto in Handbook on the
Physics and Chemistry of Rare Earths, Vol. 37 (Eds.: K. Gschneid-
ner, Jr., J.-C. G. Bünzli, V. Pecharsky), Elsevier Science, Amsterdam,
2007, Chapter 234.
[37] D. D. Perrin, W. L. F. Armarego, Purification of Laboratory Chemi-
cals, Pergamon Press, Oxford, 1988.
[38] J.-C. G. Bünzli, F. Pilloud, Inorg. Chem. 1989, 28, 2638–2642.
[39] J.-C. G. Bünzli, C. Mabillard, Inorg. Chem. 1986, 25, 2750–2754.
[40] G. Schwarzenbach, Complexometric Titrations, Chapman & Hall,
London, 1957.
[42] H. Gampp, M. Maeder, C. J. Meyer, A. D. Zuberbühler, Talanta
1986, 33, 943–951.
[43] S. R. Meech, D. C. Phillips, J. Photochem. 1983, 23, 193–217.
[44] D. F. Eaton, Pure Appl. Chem. 1988, 60, 1107–1114.
[45] T. Karstens, K. Kobs, J. Phys. Chem. 1980, 84, 1871–1872.
[46] B. Fanget, O. Devos, M. Draye, Anal. Chem. 2003, 75, 2790–2795.
[47] R. Rodriguez-Cortinas, F. Avecilla, C. Platas-Iglesias, D. Imbert, J.-
C. G. Bünzli, A. de Blas, T. Rodriguez-Blas, Inorg. Chem. 2002, 41,
5336–5349.
[48] H. Yamada, F. Uozumi, A. Ishikawa, T. Imoto, J. Biochem. 1984, 95,
503–510.
[49] M. Ishiyama, K. Sasamoto, M. Shiga, Y. Ohkura, K. Ueno, K. Nish-
iyama, I. Taniguchi, Analyst 1995, 120, 113–116.
[50] C. Korzeniewski, D. M. Callewaert, J. Immunol. Methods 1983, 64,
313–320.
[51] T. Decker, M. L. Lohmannmatthes, J. Immunol. Methods 1988, 115,
61–69.
[52] 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, 493–503.
[53] W. deW. Horrocks, Jr., D. R. Sudnick, J. Am. Chem. Soc. 1979, 101,
334–335.
[41] L. Alderighi, P. Gans, A. Ienco, D. C. Peters, A. Sabatini, A. Vacca,
Coord. Chem. Rev. 1999, 311–318.
Received: May 30, 2007
Published online: September 14, 2007
Chem. Eur. J. 2007, 13, 8678 – 8687
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8687