Gold nanoparticles functionalized with gadolinium chelates as high-relaxivity MRI contrast agents

Article abstract of DOI:10.1021/ja904094t

(Graph Presented) Water-dispersible gold nanoparticles functionalized with paramagnetic gadolinium have been fully characterized, and the NMRD profiles show very high relaxivities up to 1.5 T. Characterization using TEM images and dynamic light scattering

Full text of DOI:10.1021/ja904094t

Published on Web 07/15/2009
Gold Nanoparticles Functionalized with Gadolinium Chelates as
High-Relaxivity MRI Contrast Agents
Lo¨ıck Moriggi,† Caroline Cannizzo,† Eddy Dumas,‡ Ce´dric R. Mayer,*,§ Alexey Ulianov,§ and
Lothar Helm*,†
Institut des Sciences et Inge´nieurie Chimique, Ecole Polytechnique Fe´de´rale de Lausanne (EPFL),
CH 1015 Lausanne, Switzerland, Institut LaVoisier de Versailles, UniVersité de Versailles, UMR 8180 CNRS,
45 aVenue des Etats-Unis, F 78035 Versailles, France, and Institut de Mine´ralogie et Ge´ochimie, UniVersite´ de
Lausanne, CH 1015 Lausanne, Switzerland
Received May 20, 2009; E-mail: lothar.helm@epfl.ch
Functionalized nanoparticles (NPs) are presently under intensive
study aimed at creating useful tools for molecular diagnosis, therapy,
and biotechnology.1,2 Especially, applications in magnetic resonance
molecular imaging (MRI), which suffer from the intrinsically low
sensitivity of this technique, rely on the development of efficient
contrast agents.3-5 If we consider gadolinium-based agents, the
efficiency can be augmented by increasing the relaxation enhance-
ment induced by a single paramagnetic Gd3+ ion and by bringing
a large number of these ions to the molecular target. From
theoretical considerations, a relaxation enhancement (relaxivity) of
∼100 mM-1 s-1 induced by 1 mM Gd3+ can be expected as a
Figure 1. (left) Diameter size distributions obtained from (black) TEM
maximum for T1 contrast agents.3 However, the number of Gd3+
images and (red) DLS. (right) Chelating unit Dt complexed with gadolinium
ions that can be delivered to a specific volume element is not
restricted by stringent physics rules but by the challenging synthesis
(Gd-Dt).
of complex compounds and more soft biological and chemical
limitations.
of 3.6 was measured. The Gd-DtNP solutions were further analyzed
using scanning transmission electron microscopy coupled to energy-
dispersive X-ray analysis (STEM-EDX; see Figure S2 in the SI).
Size distributions were determined from Feret’s diameters (Figure
1, black histogram) of particles obtained from TEM images (Figure
S1 in the SI) and dynamic light scattering (DLS; Figure 1, red
histogram). It has to be kept in mind that the TEM distribution
was obtained from a deposit of particles on a carbon-coated copper
grid while the DLS distribution represents the hydrodynamic
diameter in solution. The distribution from DLS is narrower, but
the overall accordance between the results is good. From the
distributions, we calculated a mean particle diameter of 4.8 nm.
A route for creating nanosized units that can be loaded with
chelating ligands involves nanocrystals made of noble metals such
as gold or silver. Gold NPs are easily functionalized by thiol
derivatives, opening multimodal perspectives. To design a potential
MRI contrast agent, nanocrystals coated with Gd3+ chelates present
the advantage of a rigid core that minimizes internal degrees of
freedom. Another interesting aspect is the high electron density of
these heavy-metal objects, promising gold NPs as agents for X-ray
imaging.4-7
Recent studies have revealed that gold NPs show intrinsic
magnetization of Au in thiol-capped gold NPs with a permanent
magnetism at room temperature.8-12 With respect to MRI, it is
conceivable that the metallic core of an NP coated with gadolinium
chelate thiol derivatives can contribute to the relaxivity of the bulk
water molecules, in addition to the contribution from the electron
spin of the Gd3+ ions.
In the present study, a thiol derivative DTTA13 chelate (called
Dt) was used as the protective agent for the Au NPs. Batches of
gold Dt-coated NPs (DtNP) were synthesized using different
HAuCl4/Dt ratios [Table S1 in the Supporting Information (SI)].
DtNPs were complexed with gadolinium(III) and yttrium(III) by
mixing solutions of DtNPs with a slight excess of aqueous solutions
of LnCl3 (Ln ) Gd, Y). The resulting solutions were filtered with
a 0.2 µm hydrophilic syringe filter and purified with a size-exclusion
chromatography column (Sephadex LH20). Gold, gadolinium, and
yttrium concentrations in solution were determined by ICP-MS
analysis. In parallel, different dilutions of the Gd-DtNP solution
([Gd] from 201 to 34 µM) were analyzed, and a mean Au/Gd-ratio
Figure 2. Partially optimized structure (MM3 force field) of Y-DtNP
containing 201 gold atoms, 56 Y-Dt chelates, and 112 water molecules.
† Ecole Polytechnique Fe´de´rale de Lausanne.
‡ Universite´ de Versailles.
To obtain a microscopic picture of our functionalized gold NPs,
we used partially optimized molecular mechanics with the MM3
§ Universite´ de Lausanne.
9
10828 J. AM. CHEM. SOC. 2009, 131, 10828–10829
10.1021/ja904094t CCC: $40.75  2009 American Chemical Society

C O M M U N I C A T I O N S
force field, as implemented in the Scigress Explorer 7.7 program.
Barnard et al.14 found evidence for the formation of a discrete
sequence of truncated octahedral morphological motifs for gold NPs
over 1.5 nm. The particle formed is therefore described by the
formula Au201[Y-Dt(H2O)2]56 (Figure 2). The Au/Y ratio used was
3.6, corresponding to the mean Au/Gd ratio determined by
ICP-MS.
Visual inspection of a space-filling model (see Figure S4 in the
SI) shows that the spherical shell formed by the Y-Dt chelates is
densely packed. From the simple model, it can be deduced that the
thickness of the spherical shell is 1.3 ( 0.3 nm, corresponding in
this particular case roughly to the radius of the gold core. The total
diameter of the modeled particle is ∼4.9 nm and corresponds to
the mean diameter of the Gd-DtNP measured by DLS.
respectively) (Table S3 in the SI). The results for the transient ZFS
are close to the values found for the Ru-based metallostar
21
{Ru[Gd2bpy-DTTA2(H2O)4]3}4-
,
which is built using the same
chelating unit. However, the amplitude of the static ZFS is ∼30%
lower than that for {Ru[Gd2bpy-DTTA2(H2O)4]3}4-. A global
298
rotational correlation time τR ) 1.2 ns was calculated for Gd-
DtNPs. The dense packing of the Gd-Dt units on the surface of the
NPs makes them very rigid. The absence of internal rotation of the
Gd-chelating units leads to a very good fit, even at high NMR
frequencies (ν g 200 MHz).
In conclusion, we have developed small, stable, water-dispersible,
DTTA thiol-functionalized gold NPs complexed with paramagnetic
gadolinium or diamagnetic yttrium rare-earth ions. Characterizations
using TEM images, DLS, and STEM-EDX analysis indicated a
particle size distribution from 1 to 13 nm. Bulk magnetic suscep-
tibility measurements at high magnetic field showed the absence
of a significant magnetic contribution from the gold core. NMRD
profiles of Gd-DtNP solutions at 25 °C showed very high relax-
ivities with marked relaxivity humps between 10 and 60 MHz,
indicating slow rotational motion.
Thiol-covered gold nanoparticles are known to show magnetic
behavior.11 In view of the goal of creating particles that potentially
can act as MRI contrast agents, any magnetic property of the particle
is of natural interest. We determined the effective magnetic moment
1
(µeff) per gadolinium ion in different dilutions of Gd-DtNP by H
NMR analysis using Evans’ method.15,16 The µeff values of the
solutions obtained from the experimental chemical shifts (Table
S2 in the SI) are in close accordance with the effective magnetic
moment of Gd3+ [µeff(Gd3+) ) 7.94]. No magnetic moment was
detected from 1H NMR shifts of the Y-DtNP solution, even at the
high magnetic field of 18.8 T. The gold core of the NP therefore
does not contribute significantly to the overall magnetic moment
of Gd-DtNPs.
Acknowledgment. This work was supported by the Swiss NSF
and the French Agence Nationale de la Recherche (ANR-07-JCJC-
0040) and was performed in the frame of the EU COST Action
D38. We thank Prof. Hon. Andre´ Merbach and Dr. Alain Borel
for very helpful discussions and Dr. Marco Cantoni and Anas Mouti
for TEM images and the STEM-EDX experiment.
Supporting Information Available: Tables S1-S3 and Figures
S1-S5. This material is available free of charge via the Internet at
http://pubs.acs.org.
References
(1) McCarthy, J. R.; Weissleder, R. AdV. Drug DeliVery ReV. 2008, 60, 1241–
1251.
(2) Debbage, P.; Jaschke, W. Histochem. Cell Biol. 2008, 130, 845–875.
(3) Caravan, P.; Farrar, C. T.; Frullano, L.; Uppal, R. Contrast Med. Mol.
Imaging 2008, 4, 89–100.
(4) Hainfeld, J. F.; Dilmanian, F. A.; Slatkin, D. N.; Smilowitz, H. M. J. Pharm.
Pharmacol. 2008, 60, 977–985.
(5) Debouttiere, P.-J.; Roux, S.; Vocanson, F.; Billotey, C.; Beuf, O.; Favre-
Reguillon, A.; Lin, Y.; Pellet-Rostaing, S.; Lamartine, R.; Perriat, P.;
Tillement, O. AdV. Funct. Mater. 2006, 16, 2330–2339.
(6) Alric, C.; Taleb, J.; Le Duc, G.; Mandon, C.; Billotey, C.; Le Meur-Herland,
A.; Brochard, T.; Vocanson, F.; Janier, M.; Perriat, P.; Roux, S.; Tillement,
O. J. Am. Chem. Soc. 2008, 130, 5908–5915.
(7) Park, J.-A.; Reddy, P. A. N.; Kim, H.-K.; Kim, I.-S.; Kim, G.-C.; Chang,
Y.; Kim, T.-J. Bioorg. Med. Chem. Lett. 2008, 18, 6135–6137.
(8) Carmeli, I.; Leitus, G.; Naaman, R.; Reich, S.; Vager, Z. J. Chem. Phys.
2003, 118, 10372–10375.
(9) De La Fuente, J. M.; Alca´ntara, D.; Eaton, P.; Crespo, P.; Rojas, T. C.;
Ferna´ndez, A.; Hernando, A.; Penade´s, S. J. Phys. Chem. B 2006, 110,
13021–13028.
(10) Guerrero, E.; Rojas, T. C.; Multigner, M.; Crespo, P.; Munoz-Marquez,
M. A.; Garcia, M. A.; Hernando, A.; Fernandez, A. Acta Mater. 2007, 55,
1723–1730.
Figure 3. 1H NMRD profile for Gd-DtNPs at 298 K and the fit using the
modified Florence approach.
The nuclear magnetic relaxation dispersion (NMRD) profiles of
Gd-DtNP were measured at 25 °C (Figure 3). Both Gd-DtNP
batches led to very similar relaxivities (r1). A very high relaxivity
maximum of ∼60 mM-1 s-1 at 30 MHz was found, corresponding
to a relaxation enhancement per NP of more than 3000 mM-1 s-1
.
Tillement6 and Kim7 both used ligands chelating several Gd3+ ions,
leading to lower relaxivities but higher relaxation enhancements
per NP. Various theoretical approaches have been developed to
analyze NMRD profiles of gadolinium compounds.17 We used the
approach of Bertini et al.18 and Kruk et al.,19 which has been
implemented in the software package developed in Florence and
is called the “modified Florence method”.20 This method is well-
(11) Garitaonandia, J. S.; Insausti, M.; Goikolea, E.; Suzuki, M.; Cashion, J. D.;
Kawamura, N.; Ohsawa, H.; De Muro, I. G.; Suzuki, K.; Plazaola, F.; Rojo,
T. Nano Lett. 2008, 8, 661–667.
(12) Crespo, P.; Guerrero, E.; Mu´noz-Ma´rquez, M. A.; Hernando, A.; Ferna´ndez,
A. IEEE Trans. Magn. 2008, 44, 2768–2771.
(13) Moriggi, L.; Cannizzo, C.; Prestinari, C.; Berrie`re, F.; Helm, L. Inorg. Chem.
2008, 47, 8357–8366.
(14) Barnard, A. S.; Curtiss, L. A. ChemPhysChem 2006, 7, 1544–1553.
(15) Evans, D. F. J. Chem. Soc. 1959, 2003–2005.
(16) Evans, D. F.; Fazakerley, G. V.; Phillips, R. F. J. Chem. Soc. A 1971,
1931–1934.
suited for fitting the experimental NMRD profiles for slowly rotating
(17) Kowalewski, J.; Kruk, D.; Parigi, G. AdV. Inorg. Chem. 2006, 57, 41–104.
(18) Bertini, I.; Kowalewski, J.; Luchinat, C.; Nilsson, T.; Parigi, G. J. Chem.
Phys. 1999, 111, 5795–5807.
7
complexes of gadolinium(III), an S )
/ ion characterized by
2
relatively low static zero-field splitting (ZFS).
(19) Kruk, D.; Nilsson, T.; Kowalewski, J. Phys. Chem. Chem. Phys. 2001, 3,
4907–4917.
The data fitting was performed by fixing most of the parameters
to values typical for Gd complexes with the DTTA chelating
unit. The four parameters fitted were the overall rotational correla-
tion time (τR298), the correlation time for the transient ZFS (τv298),
and the amplitudes of the transient and static ZFS (∆t2 and ∆s2,
(20) Bertini, I.; Galas, O.; Luchinat, C.; Parigi, G. J. Magn. Reson. 1995, 113A,
151–158.
(21) Moriggi, L.; Aebischer, A.; Cannizzo, C.; Sour, A.; Borel, A.; Bu¨nzli, J. C.;
Helm, L. Dalton Trans. 2009, 2088–2095.
JA904094T
9
J. AM. CHEM. SOC. VOL. 131, NO. 31, 2009 10829