Prototypes of Lanthanide(III) Agents Responsive to Enzymatic Activities in Three Complementary Imaging Modalities: Visible/Near-Infrared Luminescence, PARACEST-, and T1-MRI

Article abstract of DOI:10.1021/jacs.5b12084

We report first prototypes of responsive lanthanide(III) complexes that can be monitored independently in three complementary imaging modalities. Through the appropriate choice of lanthanide(III) cations, the same reactive ligand can be used to form complexes providing detection by (i) visible (Tb³⁺) and near-infrared (Yb³⁺) luminescence, (ii) PARACEST- (Tb³⁺, Yb³⁺), or (iii) T₁-weighted (Gd³⁺) MRI. The use of lanthanide(III) ions of different natures for these imaging modalities induces only a minor change in the structure of complexes that are therefore expected to have a single biodistribution and cytotoxicity.

Full text of DOI:10.1021/jacs.5b12084

Communication
pubs.acs.org/JACS
Prototypes of Lanthanide(III) Agents Responsive to Enzymatic
Activities in Three Complementary Imaging Modalities: Visible/Near-
Infrared Luminescence, PARACEST‑, and T1‑MRI
Jiefang He,† Celia S. Bonnet,‡ Svetlana V. Eliseeva,‡ Sara Lacerda,‡ Thomas Chauvin,‡ Pascal Retailleau,†
́
‡
†
,‡
,‡
,†
́
Frederic Szeremeta, Bernard Badet, Stephane Petoud,* Eva Toth,* and Philippe Durand*
́
́
†Institut de Chimie des Substances Naturelles, CNRS UPR 2301, Universite
France
‡Centre de Biophysique Molec
́
Paris-Saclay, 1, av.de la Terrasse, 91198 Gif-sur-Yvette,
́
ulaire, CNRS UPR 4301, Universite
́
d’Orlea
́
ns, Rue Charles Sadron, F-45071 Orleans 2, France
́
S
* Supporting Information
the spacer associated with a change in MRI signal. In the present
work, we set the basis of smart imaging probes, which, in addition
to PARACEST-MRI, allow the detection of enzymatic event by
visible and near-infrared (NIR) luminescence and by T1-MRI.
Bioimaging in the NIR provides a major advantage of enhanced
detection sensitivity due to the limited autofluorescence of
biological tissues in this energy window.12 Several NIR-emitting
Ln3+ ions13 proved attractive for microscopy imaging of living
cells.14 Moreover, Ln3+-based imaging agents possess several
complementary advantages over organic fluorophores, such as
sharp emission bands (for spectral discrimination) and strong
resistance to photobleaching.15
ABSTRACT: We report first prototypes of responsive
lanthanide(III) complexes that can be monitored
independently in three complementary imaging modalities.
Through the appropriate choice of lanthanide(III) cations,
the same reactive ligand can be used to form complexes
providing detection by (i) visible (Tb3+) and near-infrared
(Yb3+) luminescence, (ii) PARACEST- (Tb3+, Yb3+), or
(iii) T1-weighted (Gd3+) MRI. The use of lanthanide(III)
ions of different natures for these imaging modalities
induces only a minor change in the structure of complexes
that are therefore expected to have a single biodistribution
and cytotoxicity.
As Ln3+ cations possess versatile magnetic and optical
properties while having similar reactivities,16 we have designed
a molecular platform that can be made responsive and that is
compatible with the complementary imaging modalities
previously described (Scheme 1). Ligand H3L1 is a DO3A
derivative substituted by a pyridylmethyl moiety. The pyridine
has multiple roles here: (i) to provide a donor atom for stable
Ln3+ complexation,17 (ii) to act as a sensitizer for Ln3+
olecular imaging aims at visualizing biological objects or
Mbiochemical events at physiological, cellular, and
molecular levels within living cells, tissues or whole organisms
for research, diagnostics, or personalized medicine purposes.1 It
relies on selective and/or responsive imaging agents optimized
for a specific imaging modality,2 i.e., magnetic resonance imaging
(MRI),3 optical imaging,4 ultrasonography,5 tomography,6 etc.
As each modality has its advantages and limitations, additional
and less ambiguous information becomes accessible by
sequentially or simultaneously combining several complemen-
tary techniques.7 The broad diversity of molecular events
requires innovative solutions for their monitoring. Among
them, enzymatic activities are of interest since their dysregulation
is involved in many diseases. Smart imaging agents endowed with
a responsive mechanism are required to detect these activities.8
Despite long-term efforts, progress in in vivo responsive imaging
remains limited.9 An appealing strategy to accelerate the
development of molecular imaging relies in the design of
responsive probes suitable for different detection techniques
using versatile molecular platforms, thus optimizing both time
and efforts for synthetic work. Such platforms must combine a
trigger, which can be adapted to a broad variety of molecular
events, and a reporter moiety that can be detected by several
imaging modalities.10 We previously designed a platform suitable
for detection of enzymatic activities by PARACEST-MRI11 in
which an enzyme-specific trigger was coupled to a Ln3+-based
reporter using a self-immolative spacer. We have demonstrated
that the action of the selected enzyme resulted in elimination of
Scheme 1. Ln3+ Complexes as Prototypes of Enzyme
Responsive Probes for Detection by Visible (Tb3+) or Near-
Infrared (Yb3+) Luminescence, by T1-Weighted- (Gd3+), or by
a
PARACEST- (Yb3+, Tb3+) MRI
a
The R in LnL1 can be replaced by an appropriate substrate to afford
enzyme sensitivity and specificity.
Received: November 23, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/jacs.5b12084
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
luminescence (antenna effect),18 and (iii) to introduce additional
functionalities, in particular a self-immolative benzyl carbamate
that will make this reporter responsive to molecular events such
as enzymatic cleavage, as previously reported.11 The enzymatic
cleavage of such probe and the elimination of the self-immolative
arm will induce the transformation of the carbamate into an
amine19 with subsequent changes in the coordination environ-
ment of the Ln3+ ion and modification of its hydration state. We
hypothesized that such switching would be detectable by both
visible and NIR luminescence as well as by T1- and PARACEST-
MRI depending on the nature of the Ln3+. We note that
PARACEST and T1 agents cannot be used simultaneously since
the T1 agent destroys the CEST signal; however, sequential
imaging of the same animal in the two techniques can ascertain
the findings and provide more accurate information. The
luminescence capability provides higher sensitivity and reso-
lution at the cellular level giving complementary information.
LnL1 are synthetically accessible model complexes of the
responsive probes that will bear the self-immolative benzyl
carbamate linker and an enzyme-specific substrate (R). For
instance, β-D-galactopyranoside can be used as a substrate for the
detection of β-galactosidase. LnL2 complexes are the end
product of the enzymatic cleavage of LnL1.
Figure 2. Left: excitation spectra of LnL1 (top) and LnL2 (bottom) (λem
= 545 nm (Tb3+) and 980 nm (Yb3+)). Center and right: emission
spectra of LnL1 (top, λex = 290 nm) and LnL2 (bottom, λex = 310 nm);
100 μM in HEPES buffer, pH 7.4, room temperature. Ln = Tb3+ (green
traces), Yb3+ (magenta traces).
7
2
2
5D4 → FJ (J = 6−0) or F5/2 → F7/2 transitions, respectively
(Figure 2, center and right). The differences in Stark splitting of
the f−f transitions, which are more pronounced in the case of
Yb3+ complexes, reflect changes in coordination environment
around the Ln3+ ion.
In order to test this concept, we synthesized and investigated
Gd3+, Yb3+, and Tb3+ complexes of H3L1 and H3L2 (Scheme 1,
Supporting Information). Synthesis of H3L1 was achieved in
eight steps with a 13% overall yield starting from commercially
available 2-bromo-6-amino pyridine. LnL1 complexes were
formed at room temperature in a pH-controlled aqueous
solution upon addition of the corresponding LnCl3 to the ligand.
Their hydrogenolysis over palladium-charcoal in water led to the
formation of the corresponding LnL2 chelates.
Luminescence lifetimes and quantum yields have been
determined in H2O and D2O solutions (Table S1). In H2O,
quantum yields and luminescence lifetime values of TbL1 are
larger than those observed for TbL2 (37(1) vs 22.2(2)% and
2.78(1) vs 1.94(4) ms) but are similar when placed in D2O
(∼36% and 3 ms), reflecting the role of O−H vibrations in the
latter complex. Indeed, comparison of τobs in H2O and D2O using
phenomenological equations20 confirms that the Tb3+ ion is
monohydrated in TbL2, while no water molecule is coordinated
to Tb3+ in TbL1. For Yb3+ complexes, the situation is different
due to the intrinsically higher impact of vibrations on the
nonradiative quenching of luminescence.13,21 Quantum yield and
lifetime values are lower even in D2O by 1.4−1.5 times for YbL2
compared to YbL1 (0.181(6) vs 0.261(5)% and 5.02(2) vs 7.7(2)
μs). This result can be explained by a stronger nonradiative
quenching contribution of N−H vibrations in YbL2. Moreover,
in H2O, YbL2 exhibits a biexponential luminescence decay with
lifetimes of 1.44 and 0.77 μs. We can hypothesize the presence of
an equilibrium between mono- and nonhydrated species in this
case.
Absorption spectra of the ligands H3L1 and H3L2 (Figure 1)
display broad bands in the UV range centered at 232, 279, and
To analyze the abilities of the Tb3+ and Yb3+ complexes to
operate as luminescence responsive probes upon elimination of
the benzyl carbamate spacer, emission spectra of LnL1 and LnL2
were measured under ligand excitation at 340 nm (Figure 3). The
choice of the excitation wavelength was dictated by the
differences in absorption for the two series of complexes (Figure
1). The integrated emission intensities of TbL2 and YbL2 were
found to be 20- and 6.5-times higher, respectively, than that of
the corresponding LnL1 complexes. Thus, conversion of visible
(Tb3+) or NIR (Yb3+) emitting probes bearing the enzyme-
specific substrate to LnL2 complexes will lead to a turn-on
luminescence response, which will be the first example of a NIR-
emitting Ln3+ complex for monitoring enzymatic reactions.
To assess the potentiality of the Gd3+ analogues as enzyme
responsive T1-MRI agents, longitudinal water proton relaxivities,
r1, were determined for GdL1 and GdL2 (0−80 MHz, 25 and 37
°C). At all fields and both temperatures, the r1 values are
considerably (90−100%) larger for GdL2 than for GdL1
(Supporting Information; r1 = 1.91 and 3.77 mM−1 s−1 for
GdL1 and GdL2, respectively; 60 MHz, 25 °C). This relaxivity
Figure 1. Absorption spectra of H3L1 and H3L2 and their corresponding
Yb3+ and Tb3+ complexes (100 μM in HEPES buffer, pH 7.4, room
temperature).
234, and 303 nm, respectively, due to π → π* transitions. Upon
formation of the Tb3+ or Yb3+ complexes, the low-energy
absorption bands undergo a bathochromic shift of 5−10 nm. It is
important to note a pronounced red-shift (25 nm, i.e., 320 vs 345
nm) of the cutoff wavelengths of the absorption bands upon
conversion of LnL1 to LnL2, which can be used to our advantage
in the design of responsive probes.
Excitation spectra recorded upon monitoring of the main
transitions of Tb3+ and Yb3+ ions at 545 and 980 nm, respectively,
match the shapes of the absorption ones, indicating that the
sensitization of Ln3+ ions is occurring through the organic ligands
(Figure 2, left). Upon excitation into ligand-centered absorption
bands at 290−310 nm, TbL and YbL complexes exhibit
characteristic emission in the visible or the NIR range due to
B
DOI: 10.1021/jacs.5b12084
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
ppm; pH 7.4, 25 °C). They are similar or slightly higher
compared to values reported for Ln3+ tetraamide DOTA-
derivatives with PARACEST effect.24
To further demonstrate the utility of our platform system, we
have acquired T1-weighted and CEST MR images of the Gd3+
and Yb3+ complexes, respectively (Figure 5). These images are in
Figure 3. Demonstration of off−on luminescence switch from LnL1
(dotted line) to LnL2 (plain line) under excitation at 340 nm (100 μM in
HEPES buffer, pH 7.4, room temperature). Ln = Tb3+ (green traces),
Yb3+ (magenta traces).
difference should lead to a remarkable turn-on response in T1-
MRI upon enzymatic activation of the probe. The low relaxivity
of GdL1 points to a nonhydrated state, in accordance with the
crystal structure (Figure S8) and the luminescence lifetimes (see
above), while the higher relaxivity of GdL2 is consistent with a
monohydration of the Ln3+.
CEST spectra were recorded for TbL1, TbL2 and YbL1, YbL2
complexes in aqueous solution by applying a 3 s selective
saturation at 25 μT in 1 ppm increments from −90 to +90 ppm
(Yb3+) or −500 to +400 ppm (Tb3+) and plotted as normalized
water signal intensity (Mz/M0 %) against frequency offset (ppm).
A single CEST effect (∼25%) was observed at +44 ppm for YbL1,
while two CEST peaks appear at −105 and −115 ppm for TbL1
(Figure S2), which were assigned to carbamate protons. While
amide protons are a common source of PARACEST in Ln3+
complexes, this is the first example of carbamate protons that
generate PARACEST effect. In contrast, no CEST is produced
by LnL2 analogues (Figures 4 and S2). Thus, the conversion of
Figure 5. T1-weighted and CEST MR images of phantoms containing
complexes of H3L1 and H3L2 with Gd3+ (2.86 mM, pH 7.8) and Yb3+
(20.8 mM; pH 7.0). T1-weighted images were acquired at 7 T using spin
echo sequence with TE = 11.0 ms and TR = 200 ms. CEST difference
images were obtained by subtracting on-resonance (+44 ppm) from off-
resonance (−44 ppm) images at 9.4 T (irradiating pulse of 25 μT, 4 s).
accordance with the relaxivities of GdL1 and GdL2 and with the
CEST effect observed for YbL1. They evidence the potential of
obtaining MRI detectable responses upon the transformation of
the carbamate to the corresponding amine derivative complexes,
representing off−on or on−off responses in T1- or CEST-MRI,
respectively.
In conclusion, we have designed, synthesized and charac-
terized the physicochemical properties of complexes formed with
a ligand system that can be tuned for specific detection of a wide
variety of enzymatic activities using complementary detection
modalities including visible and NIR optical imaging and T1- or
PARACEST-MRI depending on the nature of the Ln3+ ion. A
common triggering molecular mechanism is used for the
operation of different imaging modalities. LnL1 and LnL2
complexes described here model the states prior- and
postenzymatic cleavage occurring in these probes.
This is the first example where a single molecular system can
be used as a responsive imaging agent in three different
independent modalities. In these compounds, we have evidenced
a remarkable variation in relaxivity (Gd3+), in PARACEST as well
as in luminescence properties (Tb3+ and Yb3+), which open the
way toward multimodal enzyme detection by using a single
ligand.
Figure 4. CEST spectra of YbL1 (red) and YbL2 (blue); 500 MHz, pH
7.5, 37 °C, irradiation power 25 μT, saturation time 3 s, [YbLi]i=1,2 = 20
mM.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/jacs.5b12084.
■
S
the carbamate complexes to the amine analogues upon
enzymatic reaction will lead to a turn-off PARACEST response.
For TbL1, the two CEST peaks likely belong to two isomers, as it
was reported for YbHPDO3A.22 Indeed, Ln3+ complexes of
cyclen-based ligands may exist as two diastereomeric pairs,23
which are observable in the 1H NMR spectrum of TbL1 (Figure
S23). The two CEST signals belonging to the same molecule can
be later exploited for a ratiometric calibration that reports on
enzyme activities irrespective of probe concentration.
The exchange rate of the mobile carbamate protons has been
assessed by QUEST measurements for YbL1 (kex = 6100 100
s−1) and by the omega plot method for the two isomers of TbL1
(kex = 4300 300 s−1 at −115 ppm and 5000 400 s−1 at −105
1
Synthetic procedures, compounds characterization, H
NMR, NMRD, CEST spectra, photophysical measure-
ments, ORTEP plot, and HPLC analyses (PDF)
Crystallographic information (CIF)
AUTHOR INFORMATION
■
Corresponding Authors
*philippe.durand@cnrs.fr
*eva.jakabtoth@cnrs.fr
*stephane.petoud@inserm.fr
C
DOI: 10.1021/jacs.5b12084
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
(17) Hancock, R. D. Chem. Soc. Rev. 2013, 42, 1500.
Notes
(18) (a) de Bettencourt-Dias, A.; Barber, P. S.; Viswanathan, S. Coord.
Chem. Rev. 2014, 273−274, 165. (b) Uh, H.; Petoud, S. C. R. Chim.
2010, 13, 668.
(19) Alouane, A.; Labruere, R.; Le Saux, T.; Schmidt, F.; Jullien, L.
̀
Angew. Chem., Int. Ed. 2015, 54, 7492.
(20) Beeby, A.; Clarkson, I. M.; Dickins, R. S.; Faulkner, S.; Parker, D.;
Royle, L.; de Sousa, A. S.; Williams, J. A. G.; Woods, M. J. Chem. Soc.,
Perkin Trans. 2 1999, 493.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
This research was supported by ANR (ENZYMAGE), La Ligue
Contre le Cancer and carried out within European COST actions
■
́ ̂
TD1004 and CM1006, and Canceropole Grand Ouest. S.P.
acknowledges support from INSERM and J.H. acknowledges the
French research ministry for her grant.
(21) Bunzli, J.-C. G.; Eliseeva, S. V. J. Rare Earths 2010, 28, 824.
̈
(22) Delli Castelli, D.; Terreno, E.; Aime, S. Angew. Chem., Int. Ed.
2011, 50, 1798.
REFERENCES
■
(23) Platas-Iglesias, C. Eur. J. Inorg. Chem. 2012, 2023.
(24) Zhang, S. R.; Michaudet, L.; Burgess, S.; Sherry, A. D. Angew.
Chem., Int. Ed. 2002, 41, 1919.
(1) (a) Jung, K.-H.; Lee, K.-H. J. Pathol. Transl. Med. 2015, 49, 5.
(b) Hussain, T.; Nguyen, Q. T. Adv. Drug Delivery Rev. 2014, 66, 90.
(c) Molecular Imaging: Principles and Practice; Weissleder, R.; Ross, B.
D.; Rehemtulla, A.; Gambhir, S. S., Eds.; People’s Medical Publishing
House: Shelton, 2010.
(2) The Chemistry of Molecular Imaging; Long, N. J.; Wong, W.-T., Eds.;
Wiley: New York, 2015.
(3) (a) Hingorani, D. V.; Bernstein, A. S.; Pagel, M. D. Contrast Media
Mol. Imaging 2015, 10, 245. (b) Boros, E.; Gale, E. M.; Caravan, P.
Dalton Trans. 2015, 44, 4804. (c) Bonnet, C. S.; Tei, L.; Botta, M.; Toth,
E. In Chemistry of Contrast Agents in Medical Magnetic Resonance Imaging,
́
́
2nd ed.; Merbach, A., Helm, L., Toth, E., Eds.; John Wiley & Sons Ltd.,
2013; p 343. (d) Terreno, E.; Castelli, D. D.; Aime, S. In Chemistry of
Contrast Agents in Medical Magnetic Resonance Imaging, 2nd ed.;
́
́
h, E., Eds.; John Wiley & Sons Ltd., 2013;
Merbach, A., Helm, L., Tot
p 387.
(4) (a) Lou, Z.; Li, P.; Han, K. Acc. Chem. Res. 2015, 48, 1358.
(b) Kobayashi, H.; Choyke, P. L. Acc. Chem. Res. 2011, 44, 83.
(5) van Rooij, T.; Daeichin, V.; Skachkov, I.; de Jong, N.; Kooiman, K.
Int. J. Hyperthermia 2015, 31, 90.
(6) Van de Bittner, G. C.; Ricq, E. L.; Hooker, J. M. Acc. Chem. Res.
2014, 47, 3127.
(7) (a) Seibold, U.; Wangler, B.; Schirrmacher, R.; Wangler, C. BioMed
̈
̈
Res. Int. 2014, 2014, 13 (Article ID 153741). (b) Chiang, S. Frontiers in
Neurology 2015, 6, 74. (c) Kobayashi, H.; Longmire, M. R.; Ogawa, M.;
Choyke, P. L. Chem. Soc. Rev. 2011, 40, 4626.
(8) (a) Srivastava, A. K.; Kadayakkara, D. K.; Bar-Shir, A.; Gilad, A. A.;
McMahon, M. T.; Bulte, J. W. M. Dis. Models & Mech. 2015, 8, 323.
(b) Hingorani, D. V.; Yoo, B.; Bernstein, A. S.; Pagel, M. D. Chem. - Eur.
J. 2014, 20, 9840. (c) Razgulin, A.; Ma, N.; Rao, J. Chem. Soc. Rev. 2011,
40, 4186.
(9) (a) Reilly, R. M.; Lam, K.; Chan, C.; Levine, M. Bioconjugate Chem.
2015, 26, 625. (b) Pierre, V.; Allen, M.; Caravan, P. JBIC, J. Biol. Inorg.
Chem. 2014, 19, 127. (c) Heffern, M. C.; Matosziuk, L. M.; Meade, T. J.
Chem. Rev. 2014, 114, 4496. (d) Wolfbeis, O. S. Angew. Chem., Int. Ed.
2013, 52, 9864.
(10) (a) Chen, K.; Chen, X. Y. Curr. Top. Med. Chem. 2010, 10, 1227.
(b) Pershagen, E.; Borbas, K. E. Angew. Chem., Int. Ed. 2015, 54, 1787.
(11) Chauvin, T.; Durand, P.; Bernier, M.; Meudal, H.; Doan, B. T.;
Noury, F.; Badet, B.; Beloeil, J. C.; Toth, E. Angew. Chem., Int. Ed. 2008,
47, 4370.
(12) Smith, A. M.; Mancini, M. C.; Nie, S. Nat. Nanotechnol. 2009, 4,
710.
(13) Comby, S.; Bunzli, J.-C. G. In Handbook on the Physics and
̈
Chemistry of Rare Earths; Gschneidner, K. A., Jr., Bunzli, J.-C. G.,
̈
Pecharsky, V. K., Eds.; Elsevier Science B.V.: Amsterdam, 2007; Vol. 37,
Ch. 235, p 217.
(14) (a) Foucault-Collet, A.; Shade, C. M.; Nazarenko, I.; Petoud, S.;
Eliseeva, S. V. Angew. Chem., Int. Ed. 2014, 53, 2927. (b) 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.
(15) (a) Bunzli, J.-C. G.; Eliseeva, S. V. In Comprehensive Inorganic
̈
Chemistry II; Yam, V. W.-W., Ed.; Elsevier B.V.: Amsterdam, 2013; Vol.
8, p 339. (b) Trivedi, E. R.; Eliseeva, S. V.; Jankolovits, J.; Olmstead, M.
M.; Petoud, S.; Pecoraro, V. L. J. Am. Chem. Soc. 2014, 136, 1526.
(16) Amoroso, A. J.; Pope, S. J. A. Chem. Soc. Rev. 2015, 44, 4723.
D
DOI: 10.1021/jacs.5b12084
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX