1H NMR relaxivity of aqueous suspensions of titanium dioxide nanoparticles coated with a gadolinium(III) chelate of a DOTA-monoamide with a phenylphosphonate pendant arm

Article abstract of DOI:10.1039/b817065g

A new efficient and easy method to anchor Gd(III) chelates onto a TiO ₂ surface in order to prepare a nanoparticulate T₁ MRI contrast agent is reported. For this purpose, a new ligand, DOTAPP, a DOTA-monoamide derivative containing a

Full text of DOI:10.1039/b817065g

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/materials | Journal of Materials Chemistry
1H NMR relaxivity of aqueous suspensions of titanium
dioxide nanoparticles coated with a gadolinium(III) chelate
of a DOTA-monoamide with a phenylphosphonate pendant arm†
a
a
a
a
ˇ ab
ꢁ c
Ivan Rehor, Vojtech Kubıcek, Jan Kotek, Petr Hermann, Ivan Lukes, Jirina Szakova,
ˇ
d
ˇ
ˇ
*
´ꢀ
ꢀ
ꢁ
d
b
*
Luce Vander Elst, Robert N. Muller and Joop A. Peters
Received 29th September 2008, Accepted 12th December 2008
First published as an Advance Article on the web 30th January 2009
DOI: 10.1039/b817065g
A new efficient and easy method to anchor Gd(III) chelates onto a TiO2 surface in order to prepare
a nanoparticulate T1 MRI contrast agent is reported. For this purpose, a new ligand, DOTAPP,
a DOTA-monoamide derivative containing a phenylphosphonic acid function in the amide side chain,
was synthesized. As expected, its Ln(III) complexes exhibit structure analogy to the Ln(III) complexes
of DOTA with one water molecule bound in the apical position. The Gd(III)-DOTAPP complex shows
similar relaxometric properties as Gd(III) complexes of other DOTA-monoamides. Gd(III)-DOTAPP
was anchored onto the surface of nanocrystalline TiO2 via its phosphonate moiety. The oxide surface is
completely covered by the complexes and a strong interaction between the TiO2 surface and the
phosphonate group results in stable chemisorption. The millimolar relaxivity (r1) of the Gd(III)-
DOTAPP complex adsorbed on the surface is substantially higher than that of the free (non-anchored)
complex.
rapid progress in biochemistry has provided detailed insight into
Introduction
molecular recognition processes. This opened the way to the use
of MRI CAs for molecular imaging, the visualization of cellular
and molecular events in normal and pathological processes.
However, due to the relatively low sensitivity of the method, this
requires high local concentrations of CAs, which can only be
achieved if a targeting vector delivers a high payload of Gd(III)
chelates, each with optimal relaxation enhancing capabilities, to
a site of interest.
Magnetic resonance imaging (MRI) is one of the most widely
applied diagnostic tools in medicine. The contrast in an MRI
image is dependent on differences in the water content among
1
tissues and on the H NMR relaxation rate of water, which is
influenced by pH, viscosity, concentration of salts, proteins and
other components in the body fluids, etc. To improve the capa-
bility of MRI, contrast agents (CAs) are often applied; about
40% of all MRI examinations are made after administration of
a CA. For this purpose, complexes of paramagnetic ions or
magnetic nanoparticles are utilized as they accelerate the NMR
relaxation of water protons and, thus, enhance the contrast
among different types of tissues. Nowadays, the majority of CAs
are based on Gd(III) chelates which are used to enhance contrast
mainly in T1-weighted MR images.1,2 During recent decades, the
The longitudinal relaxation rate enhancement of the water
protons (the efficiency of a T1 contrast agent) is commonly
expressed as the millimolar relaxivity, r1. One of the most
important parameters determining the value of the relaxivity of
a Gd(III) complex is its rotational correlation time (tR),1 which is
determined mainly by its molecular volume (proportional to the
molecular weight). In general, increasing the molecular weight
leads to slower rotation and, consequently, to higher relaxivity.
Various approaches have been used to increase the molecular
weight of CAs; they involve binding of the Gd(III) chelates to
macromolecules such as inulins,3,4 dendrimers5,6 or proteins (e.g.
HSA7). Unfortunately, the relaxivity enhancement of these
conjugates was significantly lower than expected from simula-
tions. This has been ascribed to local motions caused by flexi-
bility of the macromolecular carrier and/or flexibility of the
linker anchoring the complex moiety to the carrier.
aDepartment of Inorganic Chemistry, Universita Karlova (Charles
University), Hlavova 2030, 128 40 Prague, Czech Republic
bBiocatalysis and Organic Chemistry, Department of Biotechnology, Delft
University of Technology, Julianalaan 136, 2628 Delft, BL, The
Netherlands
cDepartment of Agroenvironmental Chemistry and Plant Nutrition, Faculty
of Agrobiology, Food and Natural Resources, Czech University of Life
Sciences Prague, Kamyꢁckaꢁ 129, 165 21 Prague, Czech Republic
dDepartment of General, Organic and Biomedical Chemistry, NMR and
Molecular Imaging Laboratory, University of Mons Hainaut, B-7000
Mons, Belgium
† Electronic supplementary information (ESI) available: 1H NMR
spectra of Eu-DOTAPP, Eu-DOTA and Eu-BPAMD, full set of the
relaxometric parameters of Gd-DOTAPP, NMRD profile of TiO2
suspension, NMRD profile of GdDOTAPP + TiO2 suspension before
and after subtraction of diamagnetic contribution, simulation of
Gd(III)-DOTAPP + TiO2 NMRD profile optimized procedure for
synthesis of diethyl 4-acetamidophenylphosphonate. See DOI:
10.1039/b817065g
An alternative way to slow down the tumbling of CA mole-
cules is by anchoring the Gd(III) complexes to the surface of
oxide nanoparticles. Several studies dealing with T1 contrast
agents based on nanoparticles bearing Gd(III) chelates have
recently been published. Dye-doped silica particles were covered
with Gd(III) complexes of DTPA-/DOTA-amide.8,9 A Gd(III)-
DOTA complex was anchored onto silica-coated quantum dots
and on gold nanoparticles.10 The conjugates obtained were
1494 | J. Mater. Chem., 2009, 19, 1494–1500
This journal is ª The Royal Society of Chemistry 2009

View Article Online
proposed as dual modality probes for cell labeling. Similarly,
cells have been efficiently labeled with silica particles coated with
the Gd(III) complex of a DTPA-monoamide.11 A Gd(III)
complex of DO3A, with two water molecules in the first coor-
dination sphere of Gd(III), was attached to the surface of
boehmite through a silane pendant arm.12 In all cases mentioned
above, the Gd(III) chelates were bound to the surface of the
nanoparticles via hydrolysis of –Si(OR)3 groups linked to the
chelate. Furthermore, a methanethiosulfonate group was used
for covalent anchoring of the Gd(III) complex of a DOTA
derivative on a –SH modified silica surface through an S–S
bridge.13 A sulfide bond was also used for modification of gold
nanoparticles with Ln(III) complexes to prepare dual MRI/CT
CAs14 or luminescence probes.15
(DOTAPP) (Scheme 1) and to use it for the surface modification
of TiO2 nanoparticles. The phenylphosphonate-amide was
selected as a linker, because it is expected that its rigidity will
limit the local motions of the chelate. In that way optimal benefit
may be obtained from the slow global tumbling of the TiO2-
DOTA conjugate and thus a significant T1 relaxivity increase
may be expected upon the conjugation of the chelate to the TiO2
nanoparticle.
Experimental
Materials
Commercially available chemicals were used as received. MeCN
was dried by distillation from P2O5. t-Bu3DO3A$HBr23 and
tetraethyl 4-acetamidophenylphosphonate24 were synthesized
according to modified published procedures. TiO2 was purchased
from Degussa (type P25, primary particle diameter 30 nm,
specific surface area ꢀ50 m2/g).
Titanium dioxide may be a suitable carrier for Gd(III) chelates
since it is a biocompatible material, which is stable towards
chemical and biological degradation and which can be prepared
in a well-defined nanocrystalline form. During the preparation of
this manuscript, a paper on a Gd(III) complex of a DOTA-
monoamide derivative bound to a TiO2 surface was published;
the pyrocatechol group was used to anchor the complex.16
The phosphonic acid group has been found to be very effective
in anchoring different molecules to metal oxide surfaces.17–19 The
surface of titanium dioxide has been modified with a well-
ordered monolayer of phosphonic acids.20,21 In particular, reac-
tion of phenylphosphonic acid and its derivatives with TiO2
particles has led to the formation of very stable covalent Ti–O–P
bonds on the surface even under mild conditions.22 Therefore, we
decided to study the Gd(III) complex of a DOTA-monoamide
containing a phenylphosphonate group in a pendant arm
Methods
NMR spectroscopy. 1H (399.95 or 300.16 MHz), 13C (100.58 or
75.48 MHz), 31P (161.9 or 121.5 MHz) and 17O (54.22 or 40.69
MHz) NMR spectra were acquired at 25 ꢁC (unless stated
otherwise) with Varian Unity Inova-400 and 300 spectrometers,
1
using 5 mm sample tubes. For the H and 13C measurements in
D2O, the methyl signal of t-BuOH was used as an internal
standard (d ¼ 1.2 and 31.2 ppm, respectively). The 31P chemical
shifts were measured with respect to 1% H3PO4 in D2O as an
external reference. The pH values of the samples were measured
at ambient temperature using a Corning 125 pH-meter with an
NMR electrode (Aldrich) calibrated with standard buffers. For
the 17O NMR measurements, 10 ml of 17O-enriched water (10%
H217O) was added to the samples. 1H NMR spectra of the
samples containing H217O were acquired using water suppression
by means of presaturation of the water resonance. Variable-
temperature 17O NMR measurements of the Gd(III)-DOTAPP
complex (92 mM, pH ¼ 7.5) were performed without frequency
lock in the temperature range 7–94 ꢁC. Concentrations of Ln(III)
in the samples were determined with bulk magnetic susceptibility
(BMS) measurements.25 The 17O NMR chemical shifts were
corrected for the BMS as described previously.26 The diamag-
netic contributions to the 17O NMR shifts and relaxation rates
were determined with a sample of water at pH 5.9 in the absence
of paramagnetic complexes. The longitudinal (R1) and trans-
versal (R2) relaxation rates were obtained by the inversion
recovery method,27 and the Carr–Purcell–Meiboom–Gil pulse
sequence,28 respectively.
The 31P NMR MAS spectrum was obtained on a Varian Unity
Inova-400 spectrometer using a 7 mm VT CP–MAS probe head,
spinning frequency 7 kHz. Ca3(PO4)2 was used as an external
standard.
1
Proton NMRD profiles. The H NMRD profiles (NMRD ¼
nuclear magnetic relaxation dispersion) of an aqueous solution
of the Gd(III)-DOTAPP complex (4.6 mM, pH ¼ 7.5) and of the
TiO2 suspension with the adsorbed Gd(III)-DOTAPP complex
(0.52 mM, pH ¼ 3.5) were measured in the magnetic field range
2.34 ꢂ 10ꢃ4–0.35 T using a Stelar SpinMaster FFC-2000
Scheme 1 Structures of discussed ligands.
This journal is ª The Royal Society of Chemistry 2009
J. Mater. Chem., 2009, 19, 1494–1500 | 1495

View Article Online
relaxometer. Measurements at 0.47 and 1.42 T were performed
with a Minispec mq-20 and a Minispec mq60 (Bruker), respec-
tively.
was warmed to room temperature over 3 h and then stirred
overnight. The solids were filtered off and volatiles were removed
by evaporation. The resulting oil was three times evaporated with
toluene (10 mL) to give the crude product as yellowish crystals
(0.90 g, 97%); it was used without further purification in the next
step. dH(400 MHz, CD3OD) 1.32 (6H, t, 3JHH ¼ 7.2 Hz, 2 ꢂCH2–
CH3), 4.09 (4H, m, 2 ꢂ O–CH2), 4.21 (4H, s, Cl–CH2), 7.77 (4H,
m, CH arom.). dP(121 MHz, CD3OD) 19.3 (s).
Data evaluation. The experimental 1H NMRD data were fitted
simultaneously with 17O NMR data by means of a least-square
fitting procedure using the Micromath Scientist program version
2.0 (Salt Lake City, UT).
Mass spectrometry. Mass spectra were acquired on a Bruker
Esquire 3000 spectrometer equipped with an electrospray ion
source.
DOTAPP 10-({N-[4-(dihydroxyphosphoryl)phenyl]carbamoyl}
methyl)-1,4,7,10-tetraazacyclododecane-1,4,7-triacetic
acid.
Dried K2CO3 (0.79 g, 5.72 mmol), t-Bu3DO3A$HBr (0.682 g,
1.15 mmol), crude 3 (0.367 g, 1.20 mmol) and dry MeCN (30 mL)
were mixed in a 50 ml flask. The reaction mixture was stirred at
room temperature for 16 h. The solids were removed by filtration
and the volatiles were removed by evaporation. The resulting
yellow oil was dissolved in dry MeCN (20 mL) and trimethylsilyl
bromide (3.2 g, 21 mmol) was added dropwise. The reaction
mixture was stirred at room temperature for 16 h in the dark.
Volatiles were removed by evaporation and the crude product
was purified on a cation exchange resin (Dowex 50, H+-form,
elution with 5% pyridine in water : ethanol ¼ 1 : 1). The eluate
was concentrated by evaporation to ꢀ5 ml and kept at 4 ꢁC for
16 h. The precipitate was filtered off and purified on an anion
exchange resin (Dowex 1, OHꢃ-form, elution with 5% aq. HCl).
The volatiles were removed by evaporation and the resulting
white powder was dried under vacuum. The product was
obtained as DOTAPP$0.75HCl$4.5H2O (0.48 g, 70%). (Found:
C, 39.54; H, 6.46; N, 10.35; Cl, 4.08. Calcd. for:
C22H43.75Cl0.75N5O14.5P: C, 39.56; H, 6.60; N, 10.49, Cl, 3.98%);
ICP-AES. The concentrations of Gd, La, and Ti were deter-
mined with an ICP-AES spectrometer VistaPro (Varian) in axial
plasma configuration, equipped with an autosampler SPS-5
(Australia), an inert parallel flow nebuliser, an inert spray
chamber and a demountable torch with an inert injector tube.
The samples were digested in concentrated H2SO4 (170 ꢁC, 24
hours) and measured after dilution as solutions in 5.5% H2SO4
and 1% HNO3.
TEM. Images were obtained with a Philips EM 201 trans-
mission electron microscope, using an accelerating voltage of
80 kV.
Luminescence. The Eu(III) luminescence measurements were
performed on an Edinburgh Instruments FS900 spectrofluo-
rimeter, equipped with a 450 W xenon arc lamp, a microsecond
flash lamp and a red-sensitive photomultiplier (300–850 nm). The
luminescence spectra were obtained after excitation of the
Eu(III) 5L6 ) 7F0 band (394 nm).
ꢁ
dH(400 MHz, D2O, 90 C) 3.52 (8H, bs, CH2 cyclen), 3.61 (8H,
bs, CH2 cyclen), 3.86 (2H, s, N–CH2–CO), 3.99 (4H, s, N–CH2–
CO), 4.11 (2H, s, N–CH2–CO), 7.98 (2H, m, NH–C–CH arom.),
7.25 (2H, m, P–C–CH arom.); dC(100 MHz, D2O, 70 ꢁC) 7.2
Molecular modeling. Molecular modeling was performed with
the computer program HyperChem for Windows, release 7.5
(Hypercube Inc, Gainsville FL, USA).
2
(bm), 64.4 (bs), 127.7 (s, N–C–CH arom.), 138.0 (d, JCP ¼ 9.1,
1
P–C–CH arom.), 143.8 (d, JCP ¼ 14.8, P–C arom.), 145.6 (s,
NH–C arom.), 178.5 (s), 181.5 (bs), 184.4 (s); dP (121 MHz, D2O)
12.0 (t, 3JPH ¼ 11.1 Hz).
Synthesis
Diethyl 4-aminophenylphosphonate (2). Sodium (0.24 g, 10
mmol) was dissolved in dry EtOH (34 mL, 0.58 mol) under Ar. A
solution of 1 (1.00 g, 3.6 mmol) in dry EtOH(20 mL) was added
in one portion and the reaction mixture was refluxed under an
argon atmosphere for 12 h. Then, the volatiles were removed
with a rotavapor. The oily residue was dissolved in CHCl3 (100
mL), the obtained solution was washed with H2O (100 mL) and
dried with Na2SO4. The CHCl3 was removed by rotary evapo-
ration. The crude product was purified by column chromatog-
raphy (silica, EtOH, Rf ¼ 0.8). Amine 2 was obtained as a white
Preparation of Ln(III) complexes of DOTAPP (general
procedure). Ligand 4 (62.5 mg, 0.11 mmol) was dissolved in hot
H2O (0.5 mL). The pH of the solution was adjusted to 7.5 with
30% aq. NaOH and then a solution of LnCl3$nH2O (0.10 mmol)
in H2O (0.2 mL) was added dropwise. The resulting suspension
ꢁ
of the Ln(III) phosphonate precipitate was heated at 70 C for
12 h yielding a clear solution. The pH of this solution was
adjusted to 6.5, after which the solution was heated again at 70
ꢁC for 12 h. Then the pH was adjusted to 8.5 and the solution was
ꢁ
heated at 70 C overnight.
powder (0.70 g, 82%). dH(400 MHz, CD3OD) 1.29 (6H, t, 3JHH
¼
6.8 Hz, 2 ꢂ CH2–CH3), 4.02 (4H, m, 2 ꢂ O–CH2), 6.71 (2H, m,
CH arom.), 7.44 (2H, m, 2 ꢂ CH arom.). dP(121 MHz, CD3OD)
27.4 (s).
Adsorption of the Ln(III) DOTAPP complexes on the surface of
TiO2 (general procedure). TiO2 (0.1 g, 1.25 mmol) was suspended
in H2O (50 mL) and the suspension was sonicated in an ultra-
sonic bath for 20 min. Then, a solution of a Ln(III) DOTAPP
complex (22 mmol) in H2O (1.5 mL) was added. The pH of the
obtꢁained suspension was adjusted to 3.5 and then it was stirred at
70 C for 3 d followed by 4 h of cooling to room temperature.
The resulting suspension was washed with water six times. In
each washing step, the suspension was concentrated from 50 to
Diethyl 4-(chloroacetamido)phenylphosphonate (3). A suspen-
sion of dried K2CO3 (1.26 g, 9.1 mmol) in a solution of chlor-
oacetyl chloride (0.75 mL, 9.25 mmol) in dry MeCN (10 mL) was
cooled to ꢃ30 ꢁC and then a solution of 2 (0.70 g, 3.04 mmol) in
dry MeCN (10 mL) was added dropwise. The reaction mixture
1496 | J. Mater. Chem., 2009, 19, 1494–1500
This journal is ª The Royal Society of Chemistry 2009

View Article Online
10 ml in an ultrafiltration cell (membrane cutoff: 50 kDa) and
40 ml of H2O was added. The final volume of the suspension was
10 mL and the pH was adjusted to 3.5.
upon final heating of the solution at pH 8–9 was confirmed by the
1
characteristic H NMR spectrum of the Eu(III) complex (for
comparison of 1H NMR of Eu(III)-DOTAPP with spectra of the
Eu(III) complexes of other DOTA-like ligands, see ESI†).
Increasing of the pH in two steps is necessary as a one-step
adjustment led to precipitation of the Ln(III) hydroxide.
Results and discussion
Synthesis
Coordination properties
The ligand DOTAPP was synthesized in good yield by a multi-
step procedure starting from 4-bromoacetanilide (Scheme 2).
Reaction with diethylphosphite using Pd(PPh3)4 as the catalyst
gave ester 124 (for the optimized reaction conditions see ESI†).
The acetyl group was cleaved off with NaOEt and then the free
amine 2 was treated with chloroacetyl chloride to give 3. Treat-
ment with t-Bu3DO3A afforded the target ligand in the ester
form. After full de-esterification with an excess of Me3SiBr, the
ligand DOTAPP obtained was purified on a strong cation
exchange resin and crystallized as the pyridinium salt.
1
The H NMR spectra of the Ln(III)-DOTAPP (Ln ¼ Nd, Eu,
Dy, Yb) complexes (see ESI†) are very similar to those of the
corresponding spectra of Ln(III) complexes of other DOTA-like
ligands. Therefore, it may be concluded that the Ln(III) ions are
coordinated with four nitrogen atoms and four oxygen atoms of
carboxylic/amide groups forming two planes above and below
the central ion, and a water oxygen atom. The mutual orientation
of the ‘‘nitrogen’’ and ‘‘oxygen’’ planes in the coordination
polyhedron leads to two diastereomeric forms, a square-anti-
prism (SA, also called M) and a twisted square-antiprism (TSA,
also called m).2 These two diastereomers can be recognized as
Finally, pyridine was removed on a strong anion exchange
resin and the ligand was obtained as a non-stoichiometric
hydrochloride hydrate.
1
two sets of resonances in the H NMR spectrum. The chemical
shifts and intensities of signals of four protons of the cyclen ring33
were used to determine the diastereoisomeric ratio of the
complexes in solution. The TSA/SA ratio changes along the
lanthanide series (see Table 1); for the light lanthanides the TSA
is predominant whereas towards the end of the series the SA
isomer is preferred. The trend in the TSA/SA ratios is very
similar to that generally observed in Ln(III) complexes of DOTA
and its derivatives.29
The preparation of the Ln(III) complexes of DOTAPP pro-
ceeded in a three-step process. Immediately upon mixing of the
ligand and Ln(III) chloride, a precipitate was formed while the
pH of the resulting suspension dropped below 2. The precipita-
tion may be ascribed to the formation of a Ln(III)-phosphonate
complex as Ln(III) phosphonates are known to be mostly poly-
meric and insoluble.30,31 This precipitate dissolved during heating
at pH ꢀ 5. Analogously to complexes of other DOTA-like
ligands, the intermediate formation of an ‘‘out-of-cage’’ complex
with coordination of phosphonic acid and carboxylic groups is
expected.32 The formation of the ‘‘in-cage’’ DOTA-like complex
The number of water molecules directly coordinated to the
Eu(III) ion (q) was confirmed to be 1 by the values of lumines-
cence lifetimes measured for the complex in H2O (t ¼ 580 ms) and
in D2O (t ¼ 1830 ms). These values are in good agreement with
those observed previously for complexes containing one water
molecule in the inner coordination sphere (q ¼ 1).34–36
Relaxometric study of the Gd(III)-DOTAPP complex in solution
Variable-temperature 17O relaxation rates (measured at pH 7.5,
sample concentration 92 mM) and 1H NMRD profiles (measured
at 5 mM sample at pH 7.5) were measured to evaluate parameters
governing relaxivity of the Gd(III)-DOTAPP complex (Fig. 1).
The data were fitted simultaneously with a set of equations
commonly used.37 To limit the number of adjustable parameters
in the fitting procedure some of them were fixed. The diffusion
coefficient of the complex was fixed at the value obtained with
a semi-empirical method from its molecular weight and the
corresponding activation energy was fixed at 18.2 kJ molꢃ1 38
. The
distance of a water proton to the Gd(III) nucleus (rGdH) was fixed
Table 1 TSA/SA ratio in the Ln(III)-DOTAPP complexes
TSA/SA ratio
Lanthanide ion
DOTAPP
DOTA29
Scheme 2 Synthesis of DOTAPP. i) HPO(OEt)2, Pd(PPh3)4, Et3N,
90 ꢁC, under Ar, 36 h, 58%; ii) EtONa/EtOH, reflux, 12 h, 82%; iii)
ClCH2COCl, K2CO3, dry MeCN, ꢃ30 ꢁC / RT, 12 h, 97%; iv) t-
Bu3DO3A$HBr, K2CO3, dry MeCN, RT, 16 h; v) Me3SiBr, dry MeCN,
RT, 16 h, 70% (from 3).
Nd
Eu
Dy
Yb
56:44
30:70
12:88
9:91
55:45
20:80
4:96
8:92
This journal is ª The Royal Society of Chemistry 2009
J. Mater. Chem., 2009, 19, 1494–1500 | 1497

View Article Online
et al. for Gd-DOTA.39 The calculated relaxivity data are pre-
sented as curves in Fig. 1, and a selection of the best-fit param-
eters are compared in Table 2 with values obtained previously for
related complexes (for a full set of best-fit parameters and
a detailed discussion, see ESI†). The rotational correlation time
(
298tRH) was found to be 135 ps, which is a factor of 1.8 higher
than the value for the Gd(III)-DOTA complex.37 With molecular
models, it can be estimated that the molecular volume increases
by a factor of 1.4 upon attaching the phenylphosphonate group.
For spherical molecules, the molecular volume, according to the
Stokes–Debye–Einstein relation, is linearly proportional to the
298
rotational correlation time. The relatively large
t
value for
RH
Gd(III)-DOTAPP obtained can be attributed to the shape of the
molecule, which is far from spherical. The same value for 298tRH
has been reported for Gd-(DOTA-pBn-NH2).40
The water residence time (298tM ¼ 1.00 ms) is significantly
longer than that reported for the Gd(III)-DOTA complex, which
is typical for complexes of DOTA-amides.2 The electronic
parameters governing the relaxivity, the mean-square zero-field
splitting energy, D2, and the correlation time for the modulation
of the zero-field splitting interaction, 298tv, are also very similar to
related DOTA complexes with an amide pendant arm. The
relaxivity (6.17 sꢃ1 mMꢃ1; 20 MHz, 298 K, pH 7.5) was found to
be higher than the value reported for the Gd(III)-DOTA
complex, which can mainly be attributed to the slower tumbling
of the Gd-DOTAPP complex.
Nanocrystalline TiO2 modified with the Ln(III)-DOTAPP
complexes—synthesis and relaxometry
The Ln(III)-DOTAPP complexes were adsorbed onꢁthe surface
of nanocrystalline TiO2 under mild conditions (70 C, pH 3.5)
using a five fold excess of the adsorbate compared to the amount
needed for full surface coverage. The theoretical amount of
adsorbate needed for full surface coverage was estimated from
the specific surface area of the material (50 m2/g) and the area of
2
ꢂ
surface occupied by one complex molecule (200 A as estimated
from molecular modeling).
The pH appeared to be crucial for the stability of the colloidal
suspension. The optimal pH for the studied TiO2-complex
systems was found to be 3.5. At that pH, the colloidal suspen-
sions are stable for at least several days. A lower pH led to
dissociation of the Ln(III)-DOTAPP complexes whereas, at
higher pH, the colloid was not stable in the long term.
Comparison of the TEM images of the TiO2 nanoparticles before
and after sorption (Fig. 2) shows that the diameter and the shape
of the nanoparticles (30–80 nm) remain unchanged. Electron
diffraction patterns of both materials correspond to that of
anatase.
Fig. 1 Temperature dependences of the 17O NMR parameters. (a) T1:
1
squares and T2: triangles; (b) chemical shifts; (c) H NMRD profiles of
the Gd(III)-DOTAPP complex in solution (circles) and Gd(III)-
DOTAPP + TiO2 suspension (squares) at 25 ꢁC (filled shapes) and 37 ꢁC
(empty shapes).
The 31P MAS NMR spectrum of a freeze-dried suspension of
the La(III)-DOTAPP modified material shows a broad peak with
a center at 14 ppm and a half-width of 15 ppm, indicating surface
sorption of the complex. The broadening of the peak originates
from different modes of adsorption.41,42 The spectrum also
indicates a negligible recrystallisation of the material as the
recrystallized material, i.e. the titanium(IV)-phenylphosphonate
phase, usually exhibits a 31P resonance at ꢃ4 ppm with a smaller
half-width.,20,22 Furthermore, recrystallisation would lead to
ꢂ
at 3.10 A and the distance of closest approach of a proton of an
ꢂ
outer sphere water molecule (aGdH) to 3.50 A, the same values as
in most of the previous studies. The ratio of the the Gd inner-
sphere water O and Gd inner-sphere water H rotational vector
(tRH and tRO) was fixed at 1.5, the value determined by Merbach
1498 | J. Mater. Chem., 2009, 19, 1494–1500
This journal is ª The Royal Society of Chemistry 2009

View Article Online
Table 2 Comparison of relaxometric parameters of the gadolinium(III) complexes of DOTAPP and related ligands (Scheme 1)
r1 (20 MHz, 298 K)
[sꢃ1 mMꢃ1
]
Ligand
D2 [1020 sꢃ2
]
298tv [ps]
298tRH [ps]
298tM [ms]
DOTAPP
BPAMD43
BPAPD44
DOTA37
0.32 ꢄ 0.3
0.37
1.22
21 ꢄ 1
17
27
135 ꢄ 4
88
85
1.00 ꢄ 0.08
1.18
1.1
6.17
5.3
5.0
4.8
0.16
11
77
0.244
have been formed (i. e. chemisorption),21 since any physico-sor-
bed complex would have been removed during the washings
applied during the synthesis of the material.22
NMRD profiles of a suspension of the Gd(IIIꢁ)-DOTAPP
modified TiO2 (Fig. 1) were measured at 25 and 37 C (Gd(III)
concentration 0.52 mM, pH 3.5). The diamagnetic contribution
to the relaxivity, originating from the oxidic material, was sub-
tracted (see ESI†). The profiles show a local maximum at 20
MHz, which is typical for slowly tumbling CAs.
The resulting NMRD profile has an uncommon shape in the
region 0.1–1 MHz, which may be due to the inadequacy of the
correction for the diamagnetic contribution to the relaxation.
Due to the uncommon shape of the NMRD profile an evaluation
of the parameters governing the relaxivity by fitting the profile
with the theoretical equations was not possible. The value of
298tRH for a nanoparticle with a diameter of 30 nm, as estimated
with the Debye–Stokes–Einstein relation, is about 3 ms. For such
slowly tumbling systems, the relaxivity is independent of varia-
tions in the value of tRH. An NMRD profile, simulated using
a tRH value of 3 ms and with the other parameters the same as
those of free Gd(III)-DOTAPP (see ESI†), has a maximum
which is of about the same magnitude as that observed for
Gd(III)-DOTAPP + TiO2. This suggests that the Gd(III)
chelates are effectively immobilized on the nanoparticles. The
low field part of the simulated profile, which is mainly deter-
mined by the electronic relaxation parameters, is higher than that
of the experimental one. Probably, the electronic relaxation times
for Gd(III) on the particles are somewhat higher than in free
Gd(III)-DOTAPP.
As can be seen in Fig. 1(c), the relaxivity of the suspension
increases upon increasing the temperature. This confirms that the
relaxivity is no longer governed by the rotational correlation
time, since that would be reflected in a decrease of relaxivity upon
increasing the temperature. The observed increase can be
attributed to the increase in exchange rate of water between the
Gd(III) chelate and the bulk at higher temperature.
Fig. 2 TEM images of the nanoparticles before (A) and after (B)
adsorption.
For a particle with a diameter of 30 nm with a surface fully
covered with Gd-DOTAPP chelates, the number of Gd(III) ions
per particle can be estimated to be 3300. Therefore, the relaxivity
per particle is about 75 000 mMꢃ1 sꢃ1(20 MHz, 25 ꢁC). It may be
concluded that such particles, when stabilised at neutral pH, will
be suitable for application in molecular imaging.
a change of size and shape of the particles as well, which was not
observed in the TEM images (Fig. 2).
The concentration of TiO2 in the colloid and the adsorbed
amount of the Ln(III)-DOTAPP complex were quantified by
ICP-AES. The content of TiO2 was 10.0 g/L and the adsorbed
amount of the Ln(III)-DOTAPP complex was 52 mmol/g TiO2.
This value is a factor of 1.2 higher than the value calculated for
a surface covered by a monolayer, suggesting that the TiO2
surface is fully covered by the Ln(III)-DOTAPP complex. The
last filtrate obtained after ultrafiltration of the suspension con-
tained a negligible amount of Gd(III) (<3% of the adsorbed
amount), which indicates that the material is highly stable under
the conditions applied. This suggests that stable Ti–O–P bonds
Conclusion
A new DOTA-monoamide derivative containing a phenyl-
phosphonic acid group in the side chain (DOTAPP) was
synthesized. The Gd(III)-DOTAPP complex shows relaxometric
properties similar to other DOTA-monoamides. The rigid
This journal is ª The Royal Society of Chemistry 2009
J. Mater. Chem., 2009, 19, 1494–1500 | 1499

View Article Online
15 J. Massue, S. J. Quinn and T. Gunnlaugsson, J. Am. Chem. Soc.,
2008, 130, 6900–6901.
16 P. J. Endres, T. Paunesku, S. Vogt, T. J. Meade and G. E. Woloschak,
J. Am. Chem. Soc., 2007, 129, 15760–15761.
17 A. Vioux, J. Le Bideau, P. H. Mutin and D. Leclerq, Top. Curr.
Chem., 2004, 232, 145–174.
18 P. G. Mingalyov and G. V. Lisichkin, Russ. Chem. Rev., 2006, 75,
541–557.
19 B. Bujoli, H. Roussiere, G. Montavon, S. Laib, P. Janvier, B. Alonso,
F. Fayon, M. Petit, D. Massiot, J.-M. Bouler, J. Guicheux,
O. Gauthier, S. M. Lane, G. Nonglaton, M. Pipelier, J. Leger,
D. R. Talham and C. Tellier, Prog. Solid State Chem., 2006, 34,
257–266.
phenylphosphonate linker is very efficient for anchoring the
Ln(III) complexes of the ligand onto the surface of TiO2 nano-
particles. The surface of the particles is completely covered and
a strong interaction between TiO2 surface and phosphonate
results in stable chemisorption. The approach represents an
effective and simple approach to nanoparticle-Gd(III) chelate
based CAs. The resulting particles will be able to carry a payload
of Gd(III) chelates which is sufficient for application in mole-
cular imaging.
20 W. Gao, L. Dickinson, C. Grozinger, F. G. Morin and L. Reven,
Langmuir, 1996, 12, 6429–6435.
Acknowledgements
21 S. Marcinko and A. Y. Fadeev, Langmuir, 2004, 20, 2270–2273.
22 G. Guerrero, P. H. Mutin and A. Vioux, Chem. Mater., 2001, 13,
4367–4373.
23 A. Dadabhoy, S. Faulkner and P. G. J. Sammes, Chem. Soc., Perkin
Trans. 2, 2002, 348–357.
24 T. Hirao, T. Masunaga, Y. Ohshiro and T. Agawa, Synthesis, 1981, 1,
56–57.
25 D. M. Corsi, C. Platas-Iglesias, H. van Bekkum and J. A. Peters,
Magn. Reson. Chem., 2001, 39, 723–726.
26 E. Zitha-Bovens, L. Vander Elst, R. N. Muller, H. van Bekkum and
J. A. Peters, Eur. J. Inorg. Chem., 2001, 3101–3105.
27 R. L. Vold, J. S. Waugh, M. P. Klein and D. E. Phelps, J. Chem.
Phys., 1968, 48, 3831–3832.
Thanks are due to the EU for financial support via a Marie Curie
Early Stage Training fellowship (MEST-CT-2004-7442), to the
Grant Agency of the Academy of Science of the Czech Republic
(No. KAN201110651), and Long-Term Research Plan of the
Ministry of Education of the Czech Republic (No.
MSM0021620857) and project OC 179. This work was done in
the frame of COST Action D38 and the EU Network of Excel-
lence European Molecular Imaging Laboratory (EMIL, LSCH-
2004-503569) and DiMI (LSHB-2005-512146).
28 S. Meiboom and D. Gill, Rev. Sci. Instrum., 1958, 29, 688–691.
29 S. Aime, M. Botta, M. Fasano, M. P. M. Marques,
C. F. G. C. Geraldes, D. Pubanz and A. E. Merbach, Inorg. Chem.,
1997, 36, 2059–2068.
References
1 The Chemistry of Contrast Agents in Medical Magnetic Resonance
ꢁ
Imaging, ed. A. E. Merbach and E. Toth, Wiley, Chichester, 2001.
2 P. Hermann, J. Kotek, V. Kubıcek and I. Lukes, Dalton Trans., 2008,
3027–3047.
3 D. M. Corsi, L. Vander Elst, R. N. Muller, H. van Bekkum and
J. A. Peters, Chem. Eur. J., 2001, 7, 64–71.
30 J. G. Mao, Coord. Chem. Rev., 2007, 251, 1493–1520.
31 A. Clearfield, J. Alloys Compnd., 2006, 418, 128–138.
ꢁ
´ꢀ
ꢀ
ꢁ
32 E. Toth, E. Brucher, I. Lazar and I. Toth, Inorg. Chem., 1994, 33,
ꢁ
4070–4076.
€
ꢁ
33 M. P. M. Marques, C. F. G. C. Geraldes, A. D. Sherry,
A. E. Merbach, H. Powell, D. Pubanz, S. Aime and M. Botta,
J. Alloys Compnd., 1995, 225, 303–307.
ꢀ
ꢁ
4 P. Lebduskova, J. Kotek, P. Hermann, L. Vander Elst, R. N. Muller,
ꢀ
I. Lukes and J. A. Peters, Bioconjugate Chem., 2004, 15, 881–889.
34 W. DeW. Horrocks Jr. and D. R. Sudnick, J. Am. Chem. Soc., 1979,
101, 334–340.
35 R. M. Supkowski and W. DeW. Horrocks Jr., Inorg. Chim. Acta,
2002, 340, 44–48.
ꢁ
ꢀ
5 J. Rudovsky, M. Botta, P. Hermann, K. I. Hardcastle, I. Lukes and
S. Aime, Bioconjugate Chem., 2006, 17, 975–987.
ꢁ
6 P. Lebduskova, A. Sour, L. Helm, E. Toth, J. Kotek, I. Lukes and
ꢀ
A. E. Merbach, Dalton Trans., 2006, 3399–3406.
ꢁ
ꢁ
ꢀ
36 A. Beeby, I. M. Clarkson, R. S. Dickins, S. Faulkner, D. Parker,
L. Royle, A. S. de Sousa, J. A. G. Williams and M. Woods,
J. Chem. Soc., Perkin Trans. 2, 1999, 493–503.
37 D. H. Powell, O. M. N. Dhubhghaill, D. Pubanz, L. Helm,
Y. S. Lebedev, W. Schlaepfer and A. E. Merbach, J. Am. Chem.
Soc., 1996, 118, 9333–9346.
7 P. Caravan, N. J. Cloutier, M. T. Greenfield, S. A. McDermid,
S. U. Dunham, J. W. M. Bulte, J. C. Amedio Jr., R. J. Looby,
R. M. Supkowski, W. DeW Horrocks Jr., T. J. McMurry and
R. B. Lauffer, J. Am. Chem. Soc., 2002, 124, 3152–3162.
8 W. J. Rieter, J. S. Kim, K. M. L. Taylor, H. An, Wei Lin, T. Tarrant
and Wen. Lin, Angew. Chem. Int. Ed., 2007, 46, 3680–3682.
9 J. S. Kim, W. J. Rieter, K. M. L. Taylor, H. An, Wei. Lin and
Wen. Lin, J. Am. Chem. Soc., 2007, 129, 8962–8963.
38 L. Vander Elst, A. Sessoye, S. Laurent and R. N. Muller, Helv. Chim.
Acta, 2005, 88, 574–587.
39 F. A. Dunand, A. Borel and A. E. Merbach, J. Am. Chem. Soc., 2002,
124, 710–716.
10 D. Gerion, J. Herberg, R. Bok, E. Gjersing, E. Ramon, R. Maxwell,
J. Kurhanewicz, T. F. Budinger, J. W. Gray, M. A. Shuman and
F. F. Chen, J. Phys. Chem. C, 2007, 111, 12542–12551.
11 P. Voisin, E. J. Ribot, S. Miraux, A.-K. Bouzier-Sore, J.-F. Lahitte,
V. Bouchaud, S. Mornet, E. Thiaudiere, J.-M. Franconi, L. Raison,
C. Labrugere and M.-H. Delville, Bioconjugate Chem., 2007, 18,
1053–1063.
12 E. Delgado-Pinar, J. C. Frias, L. J. Jimenez-Borreguero, M. Teresa
Albeda, J. Alarcon and E. Garcia-Espana, Chem. Commun., 2007,
3392–3394.
ꢁ
40 Z. Jaszberenyi, L. Moriggi, P. Schmidt, C. Weidensteiner, R. Kneuer,
A. E. Merbach, L. Helm and E. Toth, J. Biol. Inorg. Chem., 2007, 12,
ꢁ
ꢁ
406–420.
41 S. Pawsey, M. McCormick, S. De Paul, R. Graf, Y. S. Lee, L. Reven
and H. W. Spiess, J. Am. Chem. Soc., 2003, 125, 4174–4184.
42 J. Randon, P. Blanc and R. Paterson, J. Membr. Sci., 1995, 98, 119–
129.
´ꢀ
43 V. Kubıcek, J. Rudovsky, J. Kotek, P. Hermann, L. Vander Elst,
R. N. Muller, Z. I. Kolar, H. T. Wolterbeek, J. A. Peters and
ꢁ
13 D. Thonon, V. Jacques and J. F. Desreux, Contrast Media Mol.
Imaging, 2007, 2, 24–34.
ꢀ
I. Lukes, J. Am. Chem. Soc., 2005, 127, 16477–16485.
´ꢀ
44 T. Vitha, V. Kubıcek, P. Hermann, L. Vander Elst, R. N. Muller,
Z. I. Kolar, H. T. Wolterbeek, W. A. P. Breeman, I. Lukes and
14 C. Alric, J. Taleb, G. Le Duc, C. Mandon, C. Billotey, A. Le Meur-
Herland, T. Brochard, F. Vocanson, M. Janier, P. Perriat, S. Roux
and O. Tillement, J. Am. Chem. Soc., 2008, 130, 5908–5915.
ꢀ
J. A. Peters, J. Med. Chem., 2008, 51, 677–683.
1500 | J. Mater. Chem., 2009, 19, 1494–1500
This journal is ª The Royal Society of Chemistry 2009