Expanding the Versatility of Dipicolinate-Based Luminescent Lanthanide Complexes: A Fast Method for Antenna Testing

Article abstract of DOI:10.1021/acs.inorgchem.5b01579

A dipicolinate (dpa)-based platform for the rapid testing of potential lanthanide-sensitizing antennae was developed; 4-methyl-7-O-alkylcoumarin-appended dpa could sensitize four lanthanides. The platform could be used to estimate the photophysical proper

Full text of DOI:10.1021/acs.inorgchem.5b01579

Communication
pubs.acs.org/IC
Expanding the Versatility of Dipicolinate-Based Luminescent
Lanthanide Complexes: A Fast Method for Antenna Testing
Julien Andres and K. Eszter Borbas*
Department of Chemistry, Biomedical Center, Uppsala University, 75 123 Uppsala, Sweden
S
* Supporting Information
sensitize europium (Eu) and terbium (Tb).6 Coumarin-
ABSTRACT: A dipicolinate (dpa)-based platform for the
rapid testing of potential lanthanide-sensitizing antennae
was developed; 4-methyl-7-O-alkylcoumarin-appended dpa
could sensitize four lanthanides. The platform could be
used to estimate the photophysical properties of a more
difficult-to-prepare 1,4,7,10-tetraazacyclododecane-1,4,7-
triacetic acid based structure carrying the same antenna.
sensitized Ln complexes are increasingly used as chemical or
biochemical probes.10 The utility of our design lies in its ability to
rapidly identify promising antennae. Thus, we have prioritized
complex accessibility over brightness. Ln sensitization has 1/r6
distance dependence. Our linker with a considerable Ln-antenna
distance facilitates complex assembly, but results in low quantum
yields. Importantly, modestly emissive systems can be success-
fully used for in vivo imaging with a sufficiently low back-
ground.11
Dpa-based ligand L1 is prepared in 4 steps from readily
available building blocks (Scheme 1). Ln complexation by L1 is
virtually instantaneous. A 1:3 Ln-to-ligand stoichiometry gives
ore and more applications take advantage of lanthanide
M
(Ln) luminescence.1 Investigating new Ln architectures
and increasing their sophistication affords versatile systems for
solving a wide scope of physicochemical and biological
problems.2 Ln ions are unique emitters but rely on photo-
sensitizers (antennae) for significant excitation in coordination
complexes.1a,b Despite the abundance of antennae, there is still a
craving for new Ln sensitizers, not least to enable a refined
understanding of Ln photophysics.3
LnL1 in H2O at neutral pH,4,5,6a,8 the formation of which was
3
evidenced by measuring the emission of the EuL1 complex upon
3
titration of L1 with a solution of EuCl3 (Figure S1). The
procedure also provides a means of complexometric titration for
L1. Eu emission of EuL1 is quite stable in the pH 7−10 range and
3
drops below pH 6 (Figure S2).
The photophysical properties of a series of LnL1 (Ln = Eu,
Here, we address the problem of antenna-dependent ligand
synthesis by taking advantage of a dipicolinate (dpa) framework
(Scheme 1). Dpa complexes are well described in the literature
3
Tb, Gd, Sm, and Dy) were then evaluated in H2O (0.01 M Tris,
pH 7.4; Figures 1 and S3−S5). The absorbance spectra showed
that all of the complexes have the same extinction coefficient
(Figure S3). Under coumarin excitation (i.e., at 320 nm), the
molar extinction coefficient is ε320 = 26900 ( 200) M−1 cm−1.
This excitation wavelength was used in most of the subsequent
measurements.
Scheme 1. (a) Synthesis of the dpa Ligand Na2L1 and Its Ln
Complexes [LnL1 ]3− and (b) the DO3A Analogue of LnL1
3
3
Eu, Tb, samarium (Sm), and dysprosium (Dy) are the main
visible-emitting Ln ions. Because of its inaccessible high-energy
excited state, gadolinium (Gd) provides a reference complex
where energy transfer onto the Ln ion is absent while retaining
the possible heavy-atom effect of the metal ion. The complexes of
all four luminescent Ln ions exhibited sensitization upon
coumarin excitation, as shown by the characteristic spectral
fingerprints of the Ln ions besides the coumarin fluorescence
(Figure 1). Eu displayed the strongest luminescence, followed by
Tb, Dy, and finally the barely detectable Sm emission.
Because of their long lifetimes (τLn; Table 1), the Ln peaks can
be discriminated by time-resolved measurements. Under these
conditions, the short-lived coumarin fluorescence disappears,
leaving only the long-lived Ln peaks. Dy and Sm lifetimes are, as
expected, much shorter than those of Eu and Tb (21 and 24 μs vs
1.4 and 1.1 ms, respectively). The resolution of Sm emission
required dpa excitation. Because dpa is closer to the Ln ion, it is a
more efficient sensitizer.6
and provide stable complexes with three ligands coordinated to a
single Ln ion in a tricapped trigonal-prismatic geometry.2c,4−8
Their inner coordination sphere is devoid of H2O molecules,
affording good luminescent properties. The dpa chromophore is
also a good standard sensitizer for multiple Ln ions.4 The
diversity of dpa-based structures alone justifies expanding its
adaptability and functionality.2a,c,5−7,9 As a model antenna, we
chose 4-methyl-7-hydroxycoumarin, which has been shown to
Received: July 15, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/acs.inorgchem.5b01579
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Communication
1a,3
5
7
from the ratio of the D0 → F1 magnetic dipole transition.
These values are also close to those reported for dpa derivatives.
Thus, this triazole-containing linker does not seem to have a
dramatic impact on the sensitization efficiency, which is around
3% for EuL1 . This is beneficial, as CuAAC is increasingly used
3
for Ln complex assembly.7a,b,10b,c,12 A clear quenching of the
coumarin fluorescence is observed when the antenna quantum
yields ΦL of the Eu, Tb, Sm, and Dy complexes are compared
with that of GdL1 . This quenching represents an 8−11% relative
3
decrease of the ΦL value of GdL1 , which is higher than the
3
sensitization efficiency calculated for EuL1 but could include
3
processes other than energy transfer. No significant quenching
difference between the emissive Ln complexes was seen (within
experimental error; see the Supporting Information for a
discussion on ΦLn errors).
At low temperature (77 K), the coumarin triplet excited-state
phosphorescence was observed (Figure 1). This phosphor-
escence is very long-lived, with triplet lifetimes around 1.5(2) s.
The Ln excited-state lifetimes are much shorter than this,
suggesting that this phosphorescent state is not transferring
energy to the Ln ions.
Figure 1. Excitation spectra of Sm, Dy, Tb, Eu, and coumarin
([GdL3]3−) at 298 K (solid lines, left). Emission spectra of LnL13 at 298
K (solid lines, right, black and blue) and time-resolved emission spectra
showing Sm, Dy, Tb, and Eu luminescence (298 K, colored lines) and
coumarin phosphorescence ([GdL3]3−, 77 K, blue dotted line).
Excitation at λex = 320 nm, except the orange line (λex = 256 nm).
1,4,7,10-Tetraazacyclododecane-1,4,7-triacetic acid (DO3A)-
based emissive Ln complexes are commonly used in vitro and in
vivo. However, their syntheses can be lengthy making the testing
of new antennae time-consuming. A noncovalent system has
been reported for the identification of potential antennae, which
relies on chromophore incorporation into the core of Eu3+-
[LnL1 ] = 1.8 × 10−5 M, 0.01 M Tris in H2O, pH 7.4.
3
Table 1. Photophysical Properties of LnL13 and LnL2 in Tris,
pH 7.4, and HEPES, pH 7.0, Buffered Aqueous Solutions,
Respectively, with λex = 322 nm and at Room Temperature
loaded micelles.13 LnL1 was envisioned as an accessible
3
ΦL/%
ΦLn/%
τLn/ms
screening platform that captures key features of the DO3A
complexes, such as antenna and linker type.
[GdL1 ]3−
13.7(2)
12.3(2)
12.6(2)
12.5(2)
12.3(2)
25.3(4)
18.7(3)
23.4(4)
24.4(4)
23.3(4)
3
[EuL1 ]3−
1.08(2)
0.26(3)
0.13(1)
0.05(1)
1.4(1)
DO3A analogues of LnL1 (Scheme 1) were prepared as
3
3
a
[TbL1 ]3−
1.1(1)
3
shown in Scheme S1. Their photophysical properties were
[DyL1 ]3−
0.021(2)
0.024(2)
evaluated as described for LnL1 . As is customary for DO3A-
3
3
[SmL1 ]3−
based complexes, LnL2 were studied in HEPES (0.1 M) buffered
aqueous solutions at pH 7.0 and at [LnL2] = 10−5 M. The
excitation spectra (Figure S6) are consistent with antenna
excitation and, compared to [LnL3]3−, are uncomplicated by the
dpa ligand absorbing below 300 nm. The emission spectra
(Figure S6) are dominated by coumarin emission, which has a
3
[GdL2]
[EuL2]
[TbL2]
[DyL2]
[SmL2]
0.57(1)
0.64(1)
0.10(1)
<0.01
0.47(5)
1.3(1)
0.010(5)
<0.005
a
quantum yield twice as high as in LnL1 (Table 1). This supports
Double exponential decay gives a better fit: 0.5(1) ms (25% of the
3
amplitude); 1.3(1) ms (75% of the amplitude). See the Supporting
Information for details.
self-quenching in the dpa3 structures, as reported for similar
complexes.6a
The striking difference between the emission spectra of LnL1
3
and LnL2 is the dramatic decrease of Eu emission in EuL2. EuL2
The superior Ln sensitization efficiency of the dpa unit is clear
from the excitation spectra. The excitation at 250 nm is stronger
than that at 320 nm, whereas the extinction coefficient at 250 nm
is lower than that at 320 nm (Figure S3). The good efficiency of
the dpa sensitizer also creates a glitch in the excitation spectra,
which appears as an unusual excitation band at ∼500 nm with
significant intensity. This band is not observed in blank solutions
or in solutions of complexes without dpa-coordinating groups.
Filtering the excitation source by a UV cutoff filter removes the
abnormal band, leaving only the sharp Ln direct absorption
spikes (Figure S4). The absorption range of the individual dpa
and coumarin moieties can be found in Figure S5. The quantum
yields were measured (Table 1) and are valid above 300 nm,
where the dpa moiety does not absorb. Below 300 nm, the
excitation of both the dpa and coumarin moieties varies with the
wavelength, and ΦLn is continuously changing, as illustrated in
Figure S5 for ΦEu.6a The quantum yields of Ln emission (ΦLn) of
EuL1 (1.1%) and TbL1 (0.3%) are comparable to those of
exhibits nearly half of the quantum yield of EuL1 . Sm emission
3
was lowered to the point that it could not be quantified anymore,
whereas Dy emission is barely affected by the different
environment. Unexpectedly, Tb experienced a major increase
of 250% of its quantum yield in TbL2.
The lifetimes are significantly shorter in LnL2, except for TbL2.
Hydration numbers (q) of 1.5 or 1.7 for EuL2 and 1.5 for TbL2
were calculated,1a consistent with reported DO3A Ln com-
plexes.14 The inner-sphere H2O molecules can account for
quenching of the Eu ion, as evidenced by the calculated intrinsic
quantum yield in EuL1 and EuL2 (41% vs 10%). The
3
sensitization efficiency is significantly different in EuL1 and
3
EuL2 [2.6(3)% vs 5.2(6)%] and seems to indicate a more
efficient sensitization from the coumarin in EuL2, in line with the
enhanced ΦL and the shorter sensitizer−Ln distance in the
structure lacking the pyridine moiety.
The improved Tb photophysics in TbL2 can be rationalized by
3
3
similar systems.4−6 The intrinsic quantum yield (41%) and
radiative rate constant (290 s−1) of the Eu ion were calculated
the smaller energy gap between the Tb D4 spectroscopic level
5
and the ligand excited state, which probably induces more back
B
DOI: 10.1021/acs.inorgchem.5b01579
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Communication
energy transfer than in the Eu complexes. The three dpa units in
REFERENCES
■
TbL1 would thus be detrimental to the Tb ion because the
3
(1) (a) de Bettencourt-Dias, A. Luminescence of lanthanide ions in
coordination compounds and nanomaterials; John Wiley & Sons Inc.:
Chichester, U.K., 2014. (b) Hanninen, P.; Harma, H.; Ala-Kleme, T.
Lanthanide luminescence: photophysical, analytical and biological aspects;
Springer: Berlin, 2011. (c) Andres, J.; Hersch, R. D.; Moser, J. E.;
Chauvin, A. S. Adv. Funct. Mater. 2014, 24, 5029. (d) Dissanayake, P.;
Mei, Y. J.; Allen, M. J. ACS Catal. 2011, 1, 1203.
(2) (a) Butler, S. J.; Delbianco, M.; Lamarque, L.; McMahon, B. K.;
Neil, E. R.; Pal, R.; Parker, D.; Walton, J. W.; Zwier, J. M. Dalton Trans.
2015, 44, 4791. (b) Cross, J. P.; Lauz, M.; Badger, P. D.; Petoud, S. J. Am.
Chem. Soc. 2004, 126, 16278. (c) Debroye, E.; Ceulemans, M.; Vander
Elst, L.; Laurent, S.; Muller, R. N.; Parac-Vogt, T. N. Inorg. Chem. 2014,
53, 1257. (d) Eliseeva, S. V.; Song, B.; Vandevyver, C. D. B.; Chauvin, A.
S.; Wacker, J. B.; Bunzli, J. C. G. New J. Chem. 2010, 34, 2915. (e) Lehr,
J.; Beer, P. D.; Faulkner, S.; Davis, J. J. Chem. Commun. 2014, 50, 5678.
(f) Sørensen, T. J.; Kenwright, A. M.; Faulkner, S. Chem. Sci. 2015, 6,
2054. (g) Delbianco, M.; Sadovnikova, V.; Bourrier, E.; Mathis, G.;
Lamarque, L.; Zwier, J. M.; Parker, D. Angew. Chem., Int. Ed. 2014, 53,
10718.
higher static concentration of coumarin increases the back-
transfer probability (i.e., rate). This explanation is supported by
an improved sensitization of Tb by blue-shifted fluorophores and
the associated longer lifetimes of the Tb 5D4 level compared to
that of 4-methylumbelliferone.6b
̈
̈
̈
A comparison of the emission spectra of LnL1 and LnL2
3
(Figures S6−S10) revealed that coumarin fluorescence of LnL1
3
is broader, extending farther into the red than that of LnL2. This
extension of the fluorescence band was also observed during
titration of L1 by Eu and correlates with the appearance of the 1:3
species. The extra component is probably due to the proximity of
three coumarins that may interact with each other (self-
quenching supports that the chromophores are within
interaction range), or even form exciplexes within LnL1 . The
3
bathochromic shift of the fluorescence tail is unfavorable for Tb
luminescence because it further shortens the donor−acceptor
excited-state energy gap. Therefore, the sensitization efficiency in
TbL2 could be even better than the 2-fold improvement
calculated from EuL31 to EuL2.
(3) Bunzli, J.-C. G. Coord. Chem. Rev. 2015, 293−294, 19.
̈
(4) (a) Aebischer, A.; Gumy, F.; Bunzli, J. C. G. Phys. Chem. Chem.
Phys. 2009, 11, 1346. (b) Chauvin, A. S.; Gumy, F.; Imbert, D.; Bunzli, J.
C. G. Spectrosc. Lett. 2004, 37, 517.
Overall, complexes LnL1 and LnL2 are interesting because of
3
their versatility: sensitization of four Ln ions, concomitant
coumarin fluorescence, and time-resolved predisposition.
Antennae sensitizing more than two Ln ions is still quite
rare.1a,3,15 For some applications (e.g., multiplex detection16),
the ability to sensitize multiple Ln ions is an important
(5) Gassner, A. L.; Duhot, C.; Bunzli, J. C. G.; Chauvin, A. S. Inorg.
Chem. 2008, 47, 7802.
(6) (a) Andres, J.; Chauvin, A. S. Eur. J. Inorg. Chem. 2010, 2010, 2700.
(b) Andres, J.; Chauvin, A. S. Phys. Chem. Chem. Phys. 2013, 15, 15981.
(7) (a) Candelon, N.; Hadade, N. D.; Matache, M.; Canet, J. L.;
Cisnetti, F.; Funeriu, D. P.; Nauton, L.; Gautier, A. Chem. Commun.
2013, 49, 9206. (b) Chamas, Z. E.; Guo, X. M.; Canet, J. L.; Gautier, A.;
Boyer, D.; Mahiou, R. Dalton Trans. 2010, 39, 7091. (c) D’Aleo, A.;
Picot, A.; Baldeck, P. L.; Andraud, C.; Maury, O. Inorg. Chem. 2008, 47,
10269.
parameter. For LnL1 , the “double sensitization” capability
3
below 300 nm allows a comparison of the new antenna to an
internal standard, which is useful for predicting Ln sensitization
by the antenna in another setting, i.e., in DO3A-based systems.
Deviations from strict analogy between LnL1 and LnL2 are
3
(8) Grenthe, I. J. Am. Chem. Soc. 1961, 83, 360.
moderate and are readily explained. The synthesis of the dpa-
based platform is easy and high-yielding and should be adaptable
to a wide range of antennae beyond coumarins. The complexes
are well-suited to applications at physiological pH and are useful
for the screening of antennae for Ln-based biological probes. The
dpa platform, hence, greatly facilitates the development of state-
of-the-art architectures that are difficult to synthesize and purify.
(9) Regueiro-Figueroa, M.; Bensenane, B.; Ruscsak, E.; Esteban-
Gomez, D.; Charbonniere, L. J.; Tircso, G.; Toth, I.; de Blas, A.;
Rodriguez-Blas, T.; Platas-Iglesias, C. Inorg. Chem. 2011, 50, 4125.
(10) (a) Halim, M.; Tremblay, M. S.; Jockusch, S.; Turro, N. J.; Sames,
D. J. Am. Chem. Soc. 2007, 129, 7704. (b) Pershagen, E.; Borbas, K. E.
Angew. Chem., Int. Ed. 2015, 54, 1787. (c) Pershagen, E.; Nordholm, J.;
Borbas, K. E. J. Am. Chem. Soc. 2012, 134, 9832. (d) Szijjarto, C.;
Pershagen, E.; Ilchenko, N. O.; Borbas, K. E. Chem. - Eur. J. 2013, 19,
3099.
(11) Foucault-Collet, A.; Gogick, K. A.; White, K. A.; Villette, S.;
Pallier, A.; Collet, G.; Kieda, C.; Li, T.; Geib, S. J.; Rosi, N. L.; Petoud, S.
Proc. Natl. Acad. Sci. U. S. A. 2013, 110, 17199.
(12) (a) Tropiano, M.; Record, C. J.; Morris, E.; Rai, H. S.; Allain, C.;
Faulkner, S. Organometallics 2012, 31, 5673. (b) Molloy, J. K.; Kotova,
O.; Peacock, R. D.; Gunnlaugsson, T. Org. Biomol. Chem. 2012, 10, 314.
(c) Szijjarto, C.; Pershagen, E.; Borbas, K. E. Dalton Trans. 2012, 41,
7660.
ASSOCIATED CONTENT
* Supporting Information
■
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.inorg-
chem.5b01579.
Synthesis procedures, titration curves, and additional
photophysical characterization (PDF)
(13) Bonnet, C. S.; Pellegatti, L.; Buron, F.; Shade, C. M.; Villette, S.;
́
̌ ́
k, V.; Guillaumet, G.; Suzenet, F.; Petoud, S.; Toth, E. Chem.
Kubíce
AUTHOR INFORMATION
Corresponding Author
*E-mail: eszter.borbas@kemi.uu.se.
■
Commun. 2010, 46, 124.
(14) Peters, J. A.; Djanashvili, K.; Geraldes, C. F. G. C.; Platas-iglesias,
C. The Chemistry of Contrast Agents in Medical Magnetic Resonance
Imaging; John Wiley & Sons, Ltd.: New York, 2013; p 251.
(15) (a) Dickins, R. S.; Howard, J. A. K.; Maupin, C. L.; Moloney, J. M.;
Parker, D.; Riehl, J. P.; Siligardi, G.; Williams, J. A. G. Chem. - Eur. J.
Author Contributions
The manuscript was written through contributions of all authors.
All authors have given approval to the final version of the
manuscript.
1999, 5, 1095. (b) Petoud, S.; Cohen, S. M.; Bunzli, J.-C. G.; Raymond,
̈
K. N. J. Am. Chem. Soc. 2003, 125, 13324. (c) Zhang, J.; Badger, P. D.;
Geib, S. J.; Petoud, S. Angew. Chem., Int. Ed. 2005, 44, 2508. (d) Lewis,
D. J.; Glover, P. B.; Solomons, M. C.; Pikramenou, Z. J. Am. Chem. Soc.
2011, 133, 1033. (e) de Bettencourt-Dias, A.; Barber, P. S.; Bauer, S. J.
Am. Chem. Soc. 2012, 134, 6987.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
(16) Pershagen, E.; Borbas, K. E. Coord. Chem. Rev. 2014, 273−274, 30.
We thank Vetenskapsradet (Project Grant 2013-4655 to K.E.B.)
̊
and Stiftelsen Olle Engkvist Byggmastare (to K.E.B.) for funding.
̈
C
DOI: 10.1021/acs.inorgchem.5b01579
Inorg. Chem. XXXX, XXX, XXX−XXX