Phosphonate-titanium dioxide assemblies: Platform for multimodal diagnostic-therapeutic nanoprobes

Article abstract of DOI:10.1021/jm200449y

Multimodal imaging-therapeutic nanoprobe TiO₂@RhdGd was prepared and successfully used for in vitro and in vivo cell tracking as well as for killing of cancer cells in vitro. TiO₂ nanoparticles were used as a core for phosphonic acid

Full text of DOI:10.1021/jm200449y

ARTICLE
pubs.acs.org/jmc
PhosphonateꢀTitanium Dioxide Assemblies: Platform for Multimodal
DiagnosticꢀTherapeutic Nanoprobes
†
‡
§
†
||
||
ꢀ
Ivan Rehoꢀr, Vanda Vilímovꢁa, Pavla Jendelovꢁa, Vojtꢀech Kubíꢀcek, Daniel Jirꢁak, Vít Herynek,
§
†
‡
†
,†
ꢀ
Miroslava Kapcalovꢁa, Jan Kotek, Jan Cernꢁy, Petr Hermann, and Ivan Lukeꢀs*
†Department of Inorganic Chemistry, Faculty of Science, Charles University in Prague, Hlavova 2030, 128 40 Prague 2, Czech Republic
‡Department of Cell Biology, Faculty of Science, Charles University in Prague, Viniꢀcnꢁa 7, 128 40 Prague, Czech Republic
§Department of Neuroscience, Institute of Experimental Medicine ASCR v.v.i., Vídeꢀnskꢁa 1083, 142 20 Praha 4, Czech Republic
Department of Diagnostic and Interventional Radiology, Institute for Clinical and Experimental Medicine, Vídeꢀnskꢁa 1958,
140 21 Praha 4, Czech Republic
S Supporting Information
b
ABSTRACT: Multimodal imagingꢀtherapeutic nanoprobe
TiO2@RhdGd was prepared and successfully used for in vitro
and in vivo cell tracking as well as for killing of cancer cells in
vitro. TiO2 nanoparticles were used as a core for phosphonic
acid modified functionalities, responsible for contrast in MRI
1
and optical imaging. The probe shows high H relaxivity and
relaxivity density values. Presence of fluorescent dye allows for
visualization by means of fluorescence microscopy. The applic-
ability of the probe was studied, using mesenchymal stem cells,
cancer HeLa cells, and T-lymphocytes. The probe did not exhibit toxicity in any of these systems. Labeled cells were successfully
visualized in vitro by means of fluorescence microscopy and MRI. Furthermore, it was shown that the probe TiO2@RhdGd can be
changed into a cancer cell killer upon UV light irradiation. The above stated results represent a valuable proof of a principle showing
applicability of the probe design for diagnosis and therapy.
’ INTRODUCTION
the depth of detection is its limiting factor. Combined MRIꢀOI
agents have great application potential in selective tumor labeling
for oncological diagnosis and surgery.7,8 This class of probes can
be also effectively used for cell labeling, for example, during stem
cell transplantations.9
Nanosized multimodal medical probes with inorganic nano-
particle core were widely studied. Molecules responsible for
various probe functions are usually attached onto the nanopar-
ticle surface or are incorporated into the nanoparticle structure.
Several types of nanoparticles can provide specific functions
themselves (i.e., quantum dots active in optical imaging, mag-
netic nanoparticles active in MRI, nanoparticles consisting of
drug molecules, etc.).
Nanosized objects like inorganic nanoparticles, liposomes,
polymers, or other nanovesicles exhibit specific behavior in
biological environment.1 They have prolonged circulation time
in the body compared to small molecules. Their ability to
penetrate into tissues and effective endocytosis is documented.
They possess the ability to accumulate in tumor tissue because of
the enhanced permeability and retention (EPR) effect.2 Nano-
vesicles can also carry considerably higher payload of a contrast
agent or a drug compared to low-molecular structures. There-
fore, using nanoobjects for preparation of advanced medical
agents is very advantageous.
Unique biological properties stated above led to use of a wide
spectrum of nanoscale objects for preparation of contrast agents
(CA) for various medical imaging modalities including magnetic
resonance imaging (MRI),3,4 radiomethods,5 optical imaging
(OI),6 and others. The combination of various functions into
one agent gives rise to multimodal imaging and imagingꢀ
therapeutic probes. The presence of two or more agents in one
probe ensures their identical biodistribution and allows for
combining of advantages of particular imaging techniques.
Combination of MRI and optical imaging in one MRIꢀOI probe
is particularly useful. MRI has exclusive spatial resolution but
rather poor sensitivity and specificity. On the other hand, optical
imaging has excellent sensitivity and very good resolution, but
Most MRIꢀOI imaging nanoprobes are based on superpar-
amagnetic iron oxide nanoparticles. These probes provide a
negative signal in MRI; i.e., their presence is indicated by
hypointense area in T2 weighted images (negative, T2 contrast).
Sometimes these areas can be difficult to distinguish from other
signal voids and a loss of spatial resolution is occurring in the
presence of CA.10 Another approach is represented by attach-
ment of Gd complexes on a diamagnetic core. Presence of Gd
chelates results in a hyperintense area in T1 weighted images
Received: April 14, 2011
Published: June 12, 2011
r
2011 American Chemical Society
5185
dx.doi.org/10.1021/jm200449y J. Med. Chem. 2011, 54, 5185–5194
|

Journal of Medicinal Chemistry
ARTICLE
(positive, T1 contrast) and lacks the listed disadvantages of
superparamagnetic nanoparticles. Nevertheless their sensitivity
is lower than that of superparamagnetic nanoparticles. Accumu-
lation of a number of CA molecules in one entity helps to
overcome the MRI sensitivity problem. Furthermore, the r1
relaxivity expressing the efficiency of MRI CA is increased upon
attachment because of the slowing of the tumbling of the chelate
in solution.11
Scheme 1. Structures of Discussed Molecules
The reported T1 MRI and OI active inorganic nanoparticles
use the cores consisting of fluorescent dye-doped silica,12
zeolite,13,14 nanoparticle, or quantum dots15 (usually coated by
silica shell16ꢀ18). The surface architecture, containing Gd che-
lates, is usually created using silyl ester chemistry. Hydrolysis of
silyl ester modified molecules performed in dispersion contain-
ing nanoparticles leads to their grafting onto the surface.12 An
alternative is nanoparticle modification with aminopropyltrialk-
oxysilane and consequent treatment of surface amino groups
with Gd chelates, typically using amidic coupling techniques.13 In
either case, the surface of the particle consists of R-SiO3 units.
The low resistance of this surface architecture against hydrolysis
is well described.19 An alternative to overcoming the problem of
hydrolytical lability is the creation of liposome-like bilayers
around lipid coated silica particles.16,18
Titanium dioxide is an inert and nontoxic material, and thus, it
was utilized for preparation of MRI20 and luminescence21
imaging nanoprobes. However, when TiO2 is irradiated with
UV light, electron excitation occurs and the electron or/and its
energy can be transferred to another molecule and be responsible
for chemical changes. Irradiated TiO2 in aqueous solution
produces highly toxic OH radicals. Therefore, TiO2 nanopar-
3
ticles are effectively used for disinfection in water treatment22,23
and it appears that these properties can also be used for killing of
cancer cells.24ꢀ29
Recently, we published a paper describing the preparation and
promising MRI properties of TiO2 nanoparticles covered with
phosphonated Gd chelates.30 Consequently, we performed a
study on adsorption properties of phosphonic and geminal
bisphosphonic acids onto the surface of nanocrystalline
TiO2.31 This study revealed that bisphosphonate strongly inter-
acts with the surface. The interaction results in formation of a
coordination polymer consisting of -P-O-Ti-O-P- chains, which
covers the surface of nanoparticle. Extremely high loads of
bisphosphonate (almost 4 times higher than that corresponding
to monolayer surface coverage) indicate creation of polymeric
3D structures forming multilayers on the surface. These multi-
layers exhibit extremely high stability against desorption, even in
the presence of competing sorbate (phosphate buffer). Further-
more, it was found that bisphosphonate and phosphonate can be
coadsorbed together, giving rise to similar bisphosphonate multi-
layers with incorporated phosphonate. These outstanding prop-
erties led us to exploit the TiO2ꢀphosphonic acid scaffold for the
preparation of a multimodal nanoagent, presented in this work.
The preparation of the bimodal MRIꢀOI probe proceeds in
one pot by single step coadsorption of both sorbates in an
aqueous solution. The developed method allows for preparation
of monodisperse nanocolloids of 12 nm anatase nanoparticles
modified with phosphonates and geminal bisphosphonates,
which are stable under biological conditions. The results prove
that the procedure is simple and universal and could lead to large
scale production of TiO2 nanoprobes covered with molecules of
various contrast agents. The application potential of the probe
for cell labeling is tested on stem cells, HeLa human cancer cells,
Figure 1. Excitation (left part of the plot) and emission (right part of
the plot) spectra of rhodamine B (dotted line), rhodamineꢀPPA (dash-
dotted line), and TiO2@RhdGd (solid line). The intensities of the
signals are relative and are given by the measurement conditions for each
sample.
and T-lyphocytes, which were labeled and visualized by fluores-
cence microscopy and by MRI. Furthermore, the viability of the
human HeLa cancer cells incubated in the presence of our
nanoagent was followed upon UV irradiation.
’ RESULTS AND DISCUSSION
Preparation and Properties of Covering Molecules. Macro-
cyclic ligand BPAMD, containing geminal bisphosphonate in the
pendent arm (Scheme 1), was prepared in our laboratory.32 The
structure of the ligand is derived from the ligand DOTA
(Scheme 1), whose Gd complexes are used as MRI contrast
agent (CA) in clinical practice. The bisphosphonate group
remains uncoordinated and is free for interaction with solid
substrates33 or metal ions.34 The relaxation properties of Gd-
BPAMD are within the range expected for DOTA-like
complexes.35 Lanthanide complexes of BPAMD exhibit high
stability, and thus, no decomplexation has been observed under
5186
dx.doi.org/10.1021/jm200449y |J. Med. Chem. 2011, 54, 5185–5194

Journal of Medicinal Chemistry
ARTICLE
conditions of adsorption reaction according to NMR and
fluorescence experiments.31
Among many existing fluorescent dyes, rhodamine B was
chosen for several reasons: (i) it is water-soluble, and there-
fore, the surface of modified particles remains hydrophilic
(hydrophobization of the nanoparticle surface can lead to loss
of colloidal stability); (ii) its emission wavelength is in the red
part of the spectra, ensuring good tissue penetration; (iii)
rhodamine B is a widely used dye, and most fluorescence
microscopes can be tuned to its excitation and emission
parameters.
The conjugate of rhodamine B and phenylphosphonic acid
(rhodamineꢀPPA, Scheme 1) was prepared from rhodamine B
and diethyl 4-aminophenylphosphonate. Excess of coupling
agent had to be used in the first reaction step to achieve 70%
conversion. The second reaction step proceeds quantitatively,
and no purification of the product was necessary. The change of
luminescent properties occurs upon modification of the rhoda-
mine B (Figure 1). The excitation band moves 60 nm to the
longer wavelength, and the emission band moves 20 nm to the
lower wavelength.
Figure 2. 1H NMRD profiles of the Gd-BPAMD complex in solution
(circles) and TiO2@RhdGd colloid (triangles) (measured at 25 °C).
Preparation and Properties of Nanoprobe TiO2@RhdGd.
Acid-stabilized colloidal TiO2 (anatase, 12 nm diameter) was
used as a substrate for adsorption. The adsorption reaction of
both sorbates Gd-BPAMD and rhodamineꢀPPA onto the TiO2
nanoparticles was performed in one reaction step. The adsorp-
tion reaction proceeds under acidic conditions. The procedure,
involving several dialysis cycles and addition of polyvinylalcohol
(PVA), was developed to remove excess sorbate and to obtain
modified material as a colloid of concentration, 2.1 mg
TiO2@RhdGd per milliter, stable at neutral pH. The procedure
led to completely transparent colloids with pH ∼6.5. Prepared
samples were stable in water, physiological solution (0.9% NaCl,
pH 7.4), and cell incubation solution.
relaxivity values indicate that even though multilayers of covering
molecules are present on the surface, these layers are well
hydrated and water molecules can effectively approach Gd(III)
complexes. The number of the Gd(III) chelates adsorbed on one
12 nm nanoparticle was calculated to be around 4800. Therefore,
r1 relaxivity per one particle can be calculated as approximately
135 000 mmolꢀ1 sꢀ1 (20 MHz, 25 °C). The calculation is
described in Supporting Information.
ꢀ1
The relaxivity per mM Gd(III) solution (28 mMꢀ1s
,
20 MHz, 25 °C) can be recalculated into so-called density of
relaxivity,36 i.e., relaxivity per 1 g of contrast agent dissolved in
1 L of water. The value 18.2 sꢀ1 gꢀ1dm3 is almost 3 times higher
than value 6.7 sꢀ1 ꢀ1dm3 reported for GdDOTA, MRI CA
g
The detailed methodology of sample preparation and the
study of adsorption properties of phosphonate and bisphospho-
nate and of their coadsorption were recently published by our
team.31 The study shows that adsorption of bisphosphonates
leads to formation of hydrolytically stable multilayers on the
surface. When bisphosphonate and phosphonate are coadsorbed,
phosphonate molecules are incorporated into this multilayer
structure. These results were also confirmed for TiO2@RhdGd,
and their discussion can be found in Supporting Information.
The amount of Gd-BPAMD in TiO2@RhdGd was estimated
using ICP-AES to be 1.22 mmol of Gd-BPAMD per 1 g of TiO2.
The amount of adsorbed rhodamine-PPA, 0.18 mmol per 1 g of
TiO2, was estimated indirectly by determination of its content in
dialyzing solutions by means of UVꢀvis spectroscopy.
commonly used in human medicine. The doses of MRI CAs used
in clinical practice are in gram scale, and therefore, the relaxivity
value expressed with respect to weight of contrast agent is crucial
for considering its applicability.
Excitation and emission spectra of the prepared probe were
measured (Figure 1). The excitation band is moved to the
longer wavelengths in comparison with rhodamineꢀPPA and is
influenced by scattering on the colloid particles at longer
wavelengths. The emission band is moved even more to the
shorter wavelengths (λEm,max = 591 nm) than in the case of
rhodamineꢀPPA.
Labeling of Stem Cells. Stem cells have ability to undergo
mitosis and to differentiate into cells of specific tissue. For this
ability, stem cells are used in cell therapy, a very perspective
branch of modern medicine research, where transplantation of
stem cells into damaged tissue leads to its reparation. Tracking of
the cells during and after the transplantation process is essential
for understanding the process as well as for successful treatment.
MRI is a particularly suitable technique for such tracking.37,38 As
a model for testing the suitability of our bimodal probe for stem
cells labeling, we have chosen rat mesenchymal stem cells
(rMSCs) as an example of adult stem cells, since they are easily
derived from bone marrow, can differentiate into a variety of
specialized cell populations, and are efficiently proliferated in
vitro. The results concerning stem cells viability are of particular
importance, as stem cells are very sensitive toward the presence
of any chemical and therefore can serve as a very good test of
probe biocompatibility.
Quantification of the amount of adsorbates shows that the
complex is approximately 6.6 times more abundant on the
surface than fluorescent dye. Surface covered mainly with MRI
contrast agent molecules is a rational design of probe, as the
sensitivity of MRI is much lower than the sensitivity of OI.
The MRI properties of the prepared nanoprobe were studied.
A plot of millimolar r1 relaxivity (corresponding to the MRI
efficiency) vs magnetic field is given in Figure 2. Significant
millimolar relaxivity increase in comparison with free Gd-
BPAMD complex is observed. Also, the shape of the curve
changes and exhibits a maximum (28 mMꢀ1 ꢀ1) at 20 MHz.
s
The relaxivity increase and the presence of a local maximum are
attributed to the immobilization of chelates on the surface, which
causes prolongation of a rotational correlation time.11 High
5187
dx.doi.org/10.1021/jm200449y |J. Med. Chem. 2011, 54, 5185–5194

Journal of Medicinal Chemistry
ARTICLE
Figure 3. Image of labeled rMSCs, obtained by fluorescence microscopy: (A) TiO2@RhdGd-labeled MSCs (red) with cell nuclei colabeled with DAPI
(blue); (B) image obtained immediately after 48 h of incubation with TiO2@RhdGd agent; (C) image obtained 7 days after the contrast agent was
washed out. Cell nuclei were not colabeled to avoid affecting cell viability. Scale bar, 50 μm.
unlabeled cells showed intensive, easily detectable fluorescence
(Figure 4B). The above stated results show very good labeling
ability of our probe as well as long persistence inside the cells.
To confirm the results obtained from rat MSCs and to prove
nontoxicity of our probe in more clinically relevant cell model, we
performed real time cell proliferation analysis using RTCA
Instrument on human mesenchymal stem cells. Cells were
seeded in E-Plates 16. The solution of TiO2@RhdGd was added
to cells after 6 h, i.e., in the proliferation curve plateau. The cells
in the second experiment were incubated in the presence of
TiO2@RhdGd from the beginning. There is no effect of probe
presence on cell proliferation and viability observed (for plots see
Supporting Information), indicating any cytotoxic effect on
hMSC. The labeling efficiency of hMSCs was 82%, i.e., was
comparable with results obtained with rMSCs.
Labeling of HeLa cells. It is well described that nanoscale
objects are accumulated in solid tumors because of the enhanced
permeability and retention (EPR) effect.2 As our probe is a
typical nanoparticle of a significant size, it is expected to
accumulate in cancer tissue. Accumulation in the solid tumor is
the only prerequisite for efficient antitumor activity. There must
Figure 4. Visualization of rat mesenchymal stem cell suspensions: (A)
magnetic resonance image of vials with labeled cells (left) and control
unlabeled ones (right); (B) fluorescence from labeled cells (left) and
control unlabeled ones (right).
also be efficient entry of the potential anticancer substance/
nanoparticle into the cell. Therefore, pilot experiments studying
the interaction of TiO2@RhdGd with a typical human adeno-
The cells were labeled by cultivation for 48 h in medium
containing TiO2@RhdGd agent. After incubation, the cells were
washed with culture medium. The viability of rMSCs labeled
with TiO2@RhdGd agent was not affected, reaching 95% and
suggesting the biocompatibility (nontoxicity) of TiO2@RhdGd
agent. Also, the growth of cells in comparison with an unlabeled
control was not affected. The labeling efficiency was determined
by fluorescence microscopy (Figure 3A), which showed that
more than 89% of cells was labeled. To check the possible efflux
of the internalized TiO2@RhdGd agent out of the cells, the
labeled cells (Figure 3B) were cultured for another 7 days in
medium not containing TiO2@RhdGd. Even after this period of
time the labeled cells were very well visible via fluorescence
microscopy (Figure 3C).
Labeled cells were harvested, and their suspensions were
centrifuged in vials to form a pellet on the bottom on the vial.
A T1-weighted MRI slice through the settled cells showed a
noticeable contrast between suspensions of the labeled and the
control cells (Figure 4A). a fluorescence camera was also used for
labeled cells visualization. The image of capillary with labeled and
carcinoma cell line were performed (HeLa). The experiments
were designed to prove the ability of TiO2@RhdGd to enter the
cells, the vesicular pathway of storage and accumulation inside
the cell, and also the ability to kill cancer cells containing
TiO2@RhdGd upon UV irradiation.
Although our probe did not exhibit toxicity on sensitive stem
cells, the viability experiments on HeLa cells, which are much
more robust and less sensitive to various substances toxic to the
primary cell lines, were also performed, as potential toxicity can
distort interpretation of appropriate results. Cells were incubated
in medium containing TiO2@RhdGd. DAPI (impermeable into
the normal live cells) was used for identification of the damaged/
dead cells. The numbers of living and dead cells were quantified
using flow cytometer and compared to control group, incubated
in the absence of TiO2@RhdGd. The viability of HeLa cells was
not influenced by the presence of TiO2@RhdGd and was about
96% in all samples.
Nanoparticles typically enter the cell via endocytosis using
various types of endosomes. Following their trafficking inside the
cell is crucial for the predictions of the time they spend inside the
5188
dx.doi.org/10.1021/jm200449y |J. Med. Chem. 2011, 54, 5185–5194

Journal of Medicinal Chemistry
ARTICLE
Figure 5. Fluorescence microscopy, colocalization of nanoparticles with holotransferrin-FITC and lysosomal marker LAMP-1, nuclei stained with
DAPI (blue) in all images: (A) colocalization of nanoparticles (red) with LAMP-1 (green) in HeLa cells; (B) lysosomal pattern of LAMP-1 staining only
(green) in HeLa cells (displayed same sample location as in (A); (C) colocalization plot (intensity of pixels in both color channels) between
nanoparticles (red) and LAMP-1 specific staining (green) exhibiting extremely high colocalization values; (D) colocalization of nanoparticles (red) and
holotransferrin-FITC (green) in HeLa cells; (E) localization of hotransferrin-FITC (green) in HeLa cells, typical pattern of early and recycling
endosomes (displayed same sample location as in (D)); (F) colocalization plot between nanoparticles (red) and holotrasferrin-FITC exhibiting lack of
colocalization.
cell before are cleared away. Therefore, we performed the
colocalization studies of the fluorescence signal from internalized
nanoparticles with typical markers of early and recycling endo-
somes (fluorescently labeled holotransferrin) and, in separate
experiment, of lysosomes (anti-LAMP-1 antibody) as shown in
Figure 5.
events indicating the entry pathway for the particle internaliza-
tion. The results of these experiments are in a good agreement
with experiments on stem cells, where 1 week after incubation
only negligible nanoparticle clearance was observed. Successful
internalization and long-term storage of nanoparticles inside the
cells increase their value as biomedical probes and spread the
possibilities of their application. TiO2@RhdGd nanoparticles
could be used in some special applications as nontoxic fixable
probes for lysosomes useful for in vivo imaging.
Photocatalytic Activity of TiO2@RhdGd Used for Killing of
HeLa Cells. The tests of photocatalytical activity leading to
cancer cell death were performed for TiO2@RhdGd. HeLa cells
were incubated in medium containing TiO2@RhdGd, were
extensively washed with phosphate buffer, and were illuminated
with UV light. The absorption edge of TiO2 is around 360 nm. As
the toxicity of UV light increases with decreasing wavelength, the
filter from the soft glass (Petri dish) was used to filter off
wavelengths under 350 nm. The irradiation was performed for
7ꢀ20 min. For precise estimation of destructive UV impact,
including delayed effects, cells were incubated for another 3 h
after illumination. Viability of harvested cells is plotted in
Figure 6. The plot shows that cells labeled with nanoparticles
are dying much more in longer exposure times than the cells in
the control group; however, the population in the control group
also suffers from UV exposure.
The obvious overlap between the fluorescence signals of
nanoparticles and Lamp-1 in Figure 5A indicates localization of
nanoparticles inside the lysosomes. The colocalization is for-
malized in Figure 5C, where the fluorescence of TiO2@RhdGd is
localized on the x axis while the LAMP-1 fluorescence is localized
on the y axis. It is obvious that most of the individual fluorescent
events exhibit both types of fluorescence; i.e., most of the signal
density is localized in the middle of the diagram. In contrast, the
position of fluorescence signals of TiO2@RhdGd does not
correspond to the signals of holotransferrin-FITC (Figure 5D,E).
Again, quantitative information is very visible in the diagram in
Figure 5F, where the majority of the signal density is distributed
along the x and y axes, indicating that only a negligible amount of
TiO2@RhdGd is localized in early and recycling endosomes.
The results from colocalization experiments show that nano-
particular probe is localized exclusively in lysosomes. This means
that nanoparticles follow the endocytic pathway, and so they are
not excluded from the cell via recycling endosomes. Therefore,
one can find them accumulated in terminally differentiated
lysosomes (the same behavior is exhibited by dextrane probes
widely used to mark lysosomal compartment). A small amount of
nanoparticles localized in endosomes (colocalizing with the
holotransferrin) probably belongs to the ongoing endocytic
Still the penetration depth of UV light in tissue is very small (it
is estimated to be less than 0.1 mm, whereas the typical
penetration depth in living tissue of the red light used for
photodynamic therapy ranges between 1 and 3 mm) and a shift
5189
dx.doi.org/10.1021/jm200449y |J. Med. Chem. 2011, 54, 5185–5194

Journal of Medicinal Chemistry
ARTICLE
Figure 6. Viability of UV illuminated cells, incubated in the presence
of TiO2@RhdGd (ꢁ, full line) and comparison with control group
(+, dashed line). Experiment was performed in triplicate; errors are
expressed in form of standard deviation.
Figure 8. Fluorescent microphotography of cryocut lymph nodes from
control mouse (A) and from the mouse with injected TiO2@RhdGd
labeled lymph node derived cells (B). Red signal indicates fluorescence
of TiO2@RhdGd accumulated in lysosomal compartment of homed
leukocytes. Blue signal indicates nuclei labeled with DAPI.
cryocuts prepared from inguinal lymph nodes (obtained from
control mouse and from mouse injected with the TiO2@RhdGd
treated cell suspension derived from lymph nodes) clearly
showed that TiO2@RhdGd treated cell population injected into
recipient mouse was able to home back to lymph node. This
ability was not altered by TiO2@RhdGd and the cells could be
visualized using fluorescence microscopy (Figure 8).
Figure 7. Mice inguinal lymph node containing splenocytes labeled
with TiO2@RhdGd (right) and control lymphnode (left). Image
obtained from fluorescence binolupe.
’ CONCLUSION
A novel dual MRI + OI nanoprobe also showing photoactive
properties was synthesized and tested for living systems. The
probe core consists of a TiO2 nanoparticle, and its surface is
covered with Gd(III) chelates responsible for contrast in MRI
and fluorescent rhodamine B derivatives. The Gd(III) chelator
contains the bisphosphonic acid group in the side chain, and
rhodamine contains the phosphonic acid group. Through the
phosphorus functionalities, the MRI and optical imaging contrast
agents were linked to the surface of TiO2 nanoparticles. This
functionalization proceeds in aqueous medium and allows for
grafting of both functionalities onto the surface in one step. The
formation of well hydrated multilayers of bisphosphonateꢀTi
coordination polymers on the surface is responsible for very high
loads of modifying molecules; however, these remain accessible
from the solution. The bisphosphonate covers exhibit very high
hydrolytical stability, which allows application of the nanoprobe
for long-term tasks in the human body. The prepared MRIꢀOI
probe has excellent relaxivity as well as relaxivity density, and the
fluorescence intensity is sufficient for observation of probe
behavior in living systems.
of this wavelength into the visible range of spectra using a suitable
chromophore is a challenging task, allowing effective use of the
prepared system in vivo.
Homing of TiO2@RhdGd Labeled Leukocytes in Vivo. As
was shown and discussed before, TiO2@RhdGd probe is non-
toxic, is efficiently endocytosed, enters the lysosomal storage
compartment, and can potentially label the appropriate cell for
the long period. We also tested the potential of the
TiO2@RhdGd to become a long-term probe for bimodal
(MRI/fluorescence) labeling of migratory cell populations
in vivo. We decided to use cell suspension derived from lymph
nodes to test the hypothesis that primary leukocytes treated in
vitro with TiO2@RhdGd can be introduced back into the
recipient animal and that these cells are able consequently to
home and find their appropriate locations in situ in the secondary
lymphoid organs. The cells derived from lymph nodes are
complex suspensions (composed mostly of T-lymphocytes,
together with B-lymphocytes and antigen presenting cells)
containing cells with high endocytic rates. Figure 7 shows that
cells endocytosed TiO2@RhdGd probe in vitro and after injec-
tion into the caudal vein were able to home into the original
location, i.e., the lymph node. Microscopic investigation of
Three types of cells were used in our study: mesenchymal
stem cells, adenocarcinoma HeLa cells, and cells derived from the
lymph node. The probe did not exhibit toxicity in any of these
5190
dx.doi.org/10.1021/jm200449y |J. Med. Chem. 2011, 54, 5185–5194

Journal of Medicinal Chemistry
ARTICLE
systems. The nanoprobe penetrates into the cells in an amount
sufficient for cell tracking. In the case of HeLa cells, storage in the
lysosomal compartment was confirmed. Furthermore, it was
shown that the TiO2@RhdGd remains inside the cells for a
number of days, which is essential for cell tracking applications.
The core of the probe, a photoactive TiO2 nanoparticle, although
is inert under normal conditions, upon UV irradiation produces
was placed on a carbon coated copper grid (SPI 3630C-MB) and was left
to dry freely.
Luminescence. The luminescence measurements were performed
on an Edinburgh Instruments FS900 spectrofluorimeter, equipped with a
450 W xenon arc lamp, a microsecond flash lamp, and a red-sensitive
photomultiplier (300ꢀ850 nm).
1H Relaxivity. Relaxivities were measured on a Bruker Minispec
MQ20 relaxometer (Bruker, Germany; 20 MHz, 25 °C). The NMRD
profile was measured on a Stelar FFC spectrometer operating at 0.01ꢀ20
MHz and a Stelar SpinMaster spectrometer operating at 20ꢀ70 MHz.
Fluorescence Microscopy. The cells were visualized using an
Olympus IX81 CellR fluorescence microscope equipped with a Hamamatsu
C4742-80-12AG digital camera under a 63ꢁ oil-immersion lens and Tx
Red (572 mn), DAPI (350 mn), and GFP filters (540 mn).
highly toxic OH radicals, which can cause the death of the
3
surrounding living tissue. In an experiment with cancer HeLa cells,
we demonstrated that nontoxic, biocompatible probe TiO2@
RhdGd can be changed into a cancer cell killer upon UV light
irradiation. The TiO2@RhdGd probe was designed to possess MRI
and fluorescence, diagnostic modalities that allow exact tumor
localization. When exact tumor position is known, the UV irradia-
tion can be performed directly in the place of the tumor, avoiding
shielding from surrounding tissue. We have performed experiments
proving that the probe can visualize tissues via MRI and fluores-
cence, as well as kill cancer cells under in vitro conditions.
Optical Imaging. Optical imaging was performed using a Caliper
IVIS Lumina XR imager (used excitation wavelength 535 nm).
Preparation of RhodamineꢀPPEt2 Conjugate. Diethyl 4-ami-
nophenylphosphonate (96 mg, 0.42 mmol) was mixed with dimethylami-
nopyridine (100 mg, 0.82 mmol), 1-(hydroxy)benzotriazole (HOBT)
(56 mg, 0.57 mmol), N,N-(diisopropyl)ethylamine (110 mg, 0.85 mmol),
and [9-(2-carboxyphenyl)-6-diethylamino-3-xanthenylidene]diethylammo-
nium chloride (rhodamine B) (200 mg, 0.42 mmol). The mixture was
dissolved in dry MeCN (20 mL), and powder N,N,N0,N0-tetramethyl-
o-(benzotriazol-1-yl)uronium tetrafluoroborate (TBTU, 540 mg, 1.68
mmol) was added. The reaction mixture was stirred at room temperature
in the dark for 12 h. The solvent was evaporated on a rotavap, and the
product was purified by column chromatography on silica (mobile phase of
EtOAc/AcOH/EtOH = 100:2:5 {vol}; Ref = 0.5). Crude product, contain-
ing remains of HOBT and diethyl 4-aminophenylphosphonate, was purified
by column chromatography (mobile phase of CH2Cl2/MeOH = 20:1) to
obtain pure product in the form of a purple oil (yield 140 mg, 51%). 1H
NMR (CD3OD): δ = 1.11 ppm (12H, t, CH3-CH2-N), 1.22 (6H, td, CH3-
CH2-O), 3.33 ppm (8H, q, CH2-N), 4.00 ppm (4H, m, CH2-O), 6.30
ppm (2H, d, Ar), 6.35 (1H, d, Ar), 6.38 (1H, d, Ar) 6.52 (1H, s, Ar), 6.55
(1H, s, Ar), 7.03 (2H, m, Ar), 7.08 (1H, m, Ar), 7.47ꢀ7.58 (4H, m, Ar),
7.95 (1H, m, Ar). 31P{1H} NMR (CD3OD): δ = 18.6. MS: calculated
654.3, observed 654.3 (M+)
’ EXPERIMENTAL SECTION
Materials. Diethyl 4-aminophenylphosphonate and ligand BPAMD
(Scheme 1) and its complexes were prepared according to the published
procedures.30,32,39 Nanocrystalline anatase (average particle diameter
12 nm) was used as a sorbent.40 The material was used in the form of
stable transparent colloidal solution (pH ∼2.5, TiO2 concentration of
1.8 g/L according to ICP-AES, specific surface ∼180 m2/g). Polyvinyl
alcohol (Mw = 80 kDa, 86ꢀ89% hydrolyzed, Wacker) was used for
colloid stabilization. Standard grade chemicals and deionized water were
used for adsorption and desorption experiments; for the preparation of
ICP samples, high-purity grade sulfuric acid, nitric acid, hydrogen
peroxide, and reverse osmosis purified water were used.
Dialysis. An ultrapor membrane with cutoff of 6ꢀ8 kDa was used for
dialysis. The ratio between volumes inside the membrane and the
washing medium was 1:250 for all experiments. One dialysis step was
performed for 12 h. The dialyzing solution was stirred.
MS. Mass spectra were acquired on a Bruker Esquire 3000 spectro-
meter equipped with an electrospray ion source.
Preaparation of RhodamineꢀPPA. Conjugate rhodamineꢀ
PPEt2 (100 mg, 0.153 mmol) was dissolved in dry MeCN (5 mL),
and trimethylsilyl bromide (612 mg, 4 mmol) was added. The reaction
mixture was stirred at room temperature in the dark for 12 h. Volatiles
were evaporated on a rotavap. EtOH (10 mL) was added to the
nonvolatile residue, and the resulting solution was evaporated on a
rotavap. The procedure was repeated three times. The product was
obtained in the form of a purple solid (89 mg, 97%, purity of >95%
according to elemental analysis) and was not further purified. EA found:
C, 57.11; H, 5.90; N, 5.85; Br 10.99, 4.08. Calcd for C34H42BrN3O7P+:
NMR. 1H (399.95 MHz), 13C (100.58 MHz), and 31P (161.9 MHz)
NMR spectra were acquired at 25 °C (unless stated otherwise) with a
Varian Unity Inova-400 spectrometer, using 5 mm sample tubes. For the
1H and 13C measurements in D2O, the methyl signal of t-BuOH was
used as an internal standard (δ = 1.2 and 31.2 ppm, respectively). The
31P chemical shifts were measured with respect to 1% H3PO4 in D2O as
an external reference.
ICP-AES. The concentrations of Gd and Ti were determined with an
ICP-AES spectrometer VistaPro (Varian) in axial plasma configuration,
equipped with an autosampler SPS-5, an inert parallel flow nebulizer, an
inert spray chamber, and a demountable torch with an inert injector
tube. The samples (200 uL) were digested in concentrated H2SO4
(300 uL) and concentrated H2O2 (100 μL) at 170 °C for 24 h. For
measurement, the solutions were diluted with 1% HNO3 to a volume
of 10 mL.
UVꢀVisible. The concentration of rhodamineꢀPPA in dialyzing
solutions was determined with a Unicam UV300 UV/vis spectrophot-
ometer using 1 cm quartz tubes. Dialyzing solutions (1 L each) were
concentrated using rotary evaporator to approximately 100 mL each.
The obtained solutions were mixed and concentrated using rotary
evaporation to 71.5 g. The pH of the resulting sample was 1.4.
Absorbance of the sample was measured at 570 nm.
1
C, 57.06; H, 5.92; N, 5.87; Br, 11.17. H NMR (CD3OD): δ = 1.16
ppm (12H, t, CH3), 3.71 ppm (8H, q, CH2), 6.78 ppm (2H, dd, Ar),
7.25 ppm (2H, d, Ar), 7.37 ppm (1H, m, Ar), 7.44 ppm (2H, dd, Ar),
7.55 ppm (2H, dd, Ar), 7.66 ppm (2H, d, Ar), 7.76 ppm (2H, m, Ar),
8.09 ppm (1H, m, Ar). 31P{1H}NMR (CD3OD): δ = 14.1 ppm (1P, s).
13C{1H} NMR (CD3OD): δ = 10.8 ppm (4C, s, CH3), 54.9 ppm (4C, s,
CH2), 95.3 (1C, s, Ar) 113.2 (2C, s, Ar), 120.0 (2C, s, Ar), 123.0 (2C, s,
Ar), 125.4 (2C, d, Ar), 128.0 (1C, s, Ar), 128.3 (1C, s, Ar), 131.6 (2C, d,
Ar), 132.4 (2C, s, Ar), 132.7 (1C, s, Ar), 132.8 (1C, s, Ar), 133.2 (1C, d,
Ar), 135.8 (1C, s, Ar), 140.0 (1C, d, Ar), 140.3 (2C, s, Ar), 152.2 (1C, s,
Ar), 153.3 (2C, s, Ar), 169.1 (1C, s, CdO). MS: calculated 598.2,
observed 596.1 (M ꢀ 2H+)
Adsorption of Ln-BPAMD and RPPA onto the TiO2 Sur-
face. RhodamineꢀPPA (2 mg, 3.3 μmol) was dissolved in a 60 mM Gd-
BPAMD solution (212 μL, 12 μmol of Gd-BPAMD, pH 2.5). The
resulting solution was mixed with 4 mL of TiO2 colloidal solution. The
reaction mixture was stirred at 70 °C for 3 days. The reaction mixture
TEM, HR-TEM. The morphology and size of the particles were
investigated by means of TEM (TecnaiG2 SpiritTwin 12, 120 kV) and
HRTEM (Jeol JEM 3010, 300 kV), for which a drop of diluted sample
5191
dx.doi.org/10.1021/jm200449y |J. Med. Chem. 2011, 54, 5185–5194

Journal of Medicinal Chemistry
ARTICLE
was dialyzed three times against HCl solution (1 L, pH 2.5) for 12 h to
remove excess of sorbate. Then PVA solution (1.33 mL, 66.5 mg of
PVA) was added, and the mixture was stirred in the vial for 12 h. Then
the reaction mixture was dialyzed three times against pure water (1 L)
for 12 h. The resulting pH was around 6.5.
Experiments with Stem Cells. To isolate rat bone marrow stromal
cells (rMSCs), femurs were dissected from 4-week-old Wistar rats. The
ends of the bones were cut, and the marrow was extruded with 5 mL of
Dulbecco's modified Eagle’s medium (DMEM) with L-glutamine (PAA,
Pasching, Austria) using a needle and syringe. Marrow cells were plated in
80 cm2 tissue culture flasks in DMEM/10% fetal bovine serum (FBS) with
100 U/mL penicillin and 0.1 mg/mL streptomycin. After 24 h, the
nonadherent cells were removed by replacing the medium. The medium
was changed every 2ꢀ3 days as the cells grew to confluence. The cells were
lifted by incubation with 0.25 wt % trypsin solution.
using 0.5 mL of trypsine/EDTA solution in PBS. Cells released into
suspension were merged with the nonadherent ones floating in the
incubation medium collected previously. The mixture was stained with
DAPI distinguishing the living and the dead cells and analyzed by the
flow cytometer LSR II (Becton Dickinson). Data were analyzed using
BD Diva software measuring the mean fluorescence intensity of the cells
in the Cy5 channel as well. Numbers of events for the unambiguously
living NL and the unambiguously dead cells ND were recorded, whereas
the viability was determined as the NL/(NL + ND) ratio.
Colocalization of Nanoparticles with Endosomal Marker. Cover-
slips with cells, incubated according to standard procedure, were co-
incubated with 5 μg/mL holotransferrin-FITC (EXBIO) for 30 min to
saturate appropriate early and recycling endosomes with the corre-
sponding fluorescence signal. After relevant treatments, coverslips with
approximately 500 000 cells were rinsed twice with PBS (0.01 M PBS,
pH 7.4) and fixed with 3.8% formaldehyde in PBS (room temperature,
20 min, directly in the multiwell plate). Fixation was followed by washing
with PBS, neutralization with 0.015 M NH4Cl in PBS, and mounting in
Mowiol containing DAPI (1 μg/50 mL).
Colocalization of Nanoparticles with Lysozomal Marker. Coverslips
with cells, incubated according to standard procedure, were permeabilized
using 0.1% Triton X-100 first, blocked with 2% BSA in PBS, incubated
with mouse anti-LAMP-1 antibody (1:1000, Santa Cruz, CA) in 1% BSA
in PBS and with the secondary goat anti-mouse Alexa Fluor 488 (1:1000,
Molecular Probes) labeled antibody. Specimens were imaged using an
Olympus IX-81 CellR microscope equipped with a Hamamatsu C4742-
80-12AG digital camera under a 63ꢁ oil-immersion lens.
Human MSCs (hMSCs) were obtained from Bioinova, s.r.o. Cells
were plated in a 75 cm2 tissue culture flask and cultured under the same
culture conditions as rMSC.
Cell Labeling and Cell Viability. rMSCs, at 100 000 cells/mL medium,
were cultured in a 12-well culture dish in a culture medium containing
TiO2@RhdGd agent (0.21 mg/mL) for 48 h. The nanoparticles were
washed out by PBS. Cells were harvested by trypsin/EDTA (Invitrogen)
and counted. The viability of rMSCs was determined using the trypan
blue exclusion test (0.1 wt %). To measure the effect of TiO2@RhdGd
agent on growth and proliferation of hMSCs, we used the xCELLigence
real time cell analyzer (RTCA) instrument. An amount of 10 000 h
MSCs per well was seeded in E-Plates 16 in quadruplicate in normal
culture medium or in medium containing TiO2@RhdGd agent from the
beginning of experiment or with addition of TiO2@RhdGd agent after
6 h, when the proliferation curve has reached plateau. The final volume
of each well was 200 μL. Dynamic cell proliferation was monitored at 15
min intervals from time of plating until the end of the experiment (53 h).
The cell index value (dimensionless parameter) was derived from the
change in electrical impedance as the living cells interact with the
biocompatible microelectrode surface in the E-Plate well.
Labeling Efficiency. To check the intensity of fluorescent staining, an
Axioplan Imaging II fluorescence microscope was used at a magnifica-
tion of 200ꢁ. Cell nuclei were labeled with DAPI (Sigma, 100 μL/ml).
Labeling efficiency was determined by manually counting the total
number of cell nuclei and cell nuclei in fluorescent cells in 10 optical
fields from three wells. The scanned images with manually labeled cells
were processed by the image analysis toolbox of MATLAB software
(The MathWorks, MA, U.S.). The presence or absence of a label inside
the cells was expressed as the percentage of labeled cells.
Illumination with UV Light. Viability Study. Cells cultivated and
incubated according to standard procedure were washed in PBS
(medium was collected so that no dead cells were lost) and illuminated
for 7, 15, or 20 min with UV light (mercury black lamp, glass filter filtering of
wavelengths under 350 nm, 5 mW/cm2, 37°C). After UV illumination, cells
were incubated for 3 h at 37 °C with 5% CO2 and then harvested using
0.5 mL of trypsine/EDTA solution in PBS (trypsin was inactivated after
cells were detached from the plastic using previously collected FBS containg
medium from the same cell cultures). Cell suspesion was stained with DAPI
distinguishing the living and the dead cells and analyzed by the flow
cytometer LSR II. The experiment was repeated two times in triplicate.
In Vivo Experiment. Primary B and T lymphocytes together with
antigen presenting cells (APC) from the lymph node were prepared
from 6-week-old Bl-6 mouse. The lymph node was dissected, and the
cells were mechanically homogenyzed before being transferred to RPMI
(medium supplemented with 10% fetal inactivated bovine serum (FBS
and RPMI, Gibco BRL, Scotland), 40 mg/mL gentamycin (Lek,
Slovenia), 0.25 mg/mL glutamine (Sevapharma, Czech Republic))
and incubated with the nanoparticles (0.21 mg/mL, 37 °C, 5% CO2,
24 h). After incubation cells were washed with RPMI twice and 3 ꢁ 106
viable lymph node derived leukocytes were transferred in the recipient
Bl-6 mice via caudal vein. After 12 h, the recipient mouse was dissected,
inguinal lymph nodes were extracted and immediately quickly imaged
using a motorized fluorescence binolupe (Zeiss Lumar, V12). The same
lymph nodes were fixed (3.8% PFA in phosphate buffered saline (PBS),
1 h), transferred to 20% sucrose (in PBS solution, 4 °C, 2 h frozen
(ꢀ18 °C) in tissue freezing medium Jung (Leica Microsystems) on metal
specimen disk, and cut at 15 μm with Leica CM 3050S cryocut. Thermo
Scientific Superfrost Plus microscope slides were used (Menzel-Glaser).
The cells on the slides were permeabilized in 0.1% Triton X-100 and
mounted in Mowiol-DAPI.
MRI Visualization of Labeled Stem Cells. About 106 cells were labeled
for 48 h, as is described above. After this time, the cells were harvested
and counted in a B€urker chamber. The cell suspensions were transferred
to 1.5 mL vials. The suspensions were centrifuged at 3000g, and then the
vials were put into wells previously filled with 4% gelatin stabilized by
0.1% phenol and 0.1% NaN3. Magnetic resonance measurements were
performed using a 4.7 T Bruker Biospec spectrometer equipped with a
commercially available resonator coil (Bruker) at 25 °C. A standard T1-
weighted turbo spin echo sequence (repetition time TR = 250 ms, TE =
12 ms, slice thickness = 0.63 mm, turbo factor = 1, number of
acquisitions = 8, FOV = 4 ꢁ 4 cm2, matrix = 256 ꢁ 256) was used.
Experiments with HeLa Cells. Cell Culture Incubation. Stan-
dard Procedure. HeLa cells were grown in Dulbecco’s modified Eagle
medium (DMEM, Gibco BRL, Scotland) up to 80% confluency,
supplemented with 10% fetal inactivated bovine serum (FBS, Gibco
BRL, Scotland), 40 mg/mL gentamycin (Lek, Slovenia), 0.25 mg/mL
glutamine (Sevapharma, Czech Republic) and incubated in the presence
of TiO2@RhdGd nanoparticles (0.21 mg/mL, 37 °C, 5% CO2, 24 h).
Characterization of the Nanoparticle Toxicity. Cells were incubated
according to the standard procedure and consequently were harvested
’ ASSOCIATED CONTENT
S
Supporting Information. Additional details on hydroly-
b
tical stability of TiO2@RhdGd, on derivation of per particle
relaxivity, and on proliferation curves of human mesenchymal
5192
dx.doi.org/10.1021/jm200449y |J. Med. Chem. 2011, 54, 5185–5194

Journal of Medicinal Chemistry
ARTICLE
stem cells. This material is available free of charge via the Internet
at http://pubs.acs.org.
(10) Kꢀríꢀz, J.; Jirꢁak, D.; White, D.; Foster, P. Magnetic resonance
imaging of pancreatic islets transplanted into the right liver lobes of
diabetic mice. Transplant. Proc. 2008, 40, 444–448.
(11) The Chemistry of Contrast Agents in Medical Magnetic Resonance
Imaging; Merbach, A. E., Tꢁoth, E., Eds.; Wiley: Chichester, U.K., 2001.
’ AUTHOR INFORMATION
ꢁ
(12) Rieter, W. J.; Kim, J. S.; Taylor, K. M. L.; An, H.; Lin, Wei.;
Tarrant, T.; Lin, W. Hybrid silica nanoparticles for multimodal imaging.
Angew. Chem., Int. Ed. 2007, 46, 3680–3682.
(13) Carniato, F.; Tei, L.; Cossi, M.; Marchese, L.; Botta, M. A
chemical strategy for the relaxivity enhancement of GdIII chelates
anchored on mesoporous silica nanoparticles. Chem.—Eur. J. 2010,
16, 10727–10734.
(14) Tsotsalas, M.; Busby, M.; Gianolio, E.; Aime, S.; De Cola,
L. Functionalized nanocontainers as dual magnetic and optical
probes for molecular imaging applications. Chem. Mater. 2008, 20,
5888–5893.
(15) Prinzen, L.; Miserus, R.-J. J. H. M.; Dirksen, A.; Hackeng, T. M.;
Deckers, N.; Bitsch, N. J.; Megens, R. T. A.; Douma, K.; Heemskerk,
J. W.; Kooi, M. E.; Frederik, P. M.; Slaaf, D. W.; van Zandvoort, M. A. M.
J.; Reutelingsperger, C. P. M. Optical and magnetic resonance imaging of
cell death and platelet activation using annexin A5-functionalized
quantum dots. Nano Lett. 2007, 7, 93–100.
Corresponding Author
*Phone: +420221951259. Fax: +420221951253. E-mail: lukes@
natur.cuni.cz.
’ ACKNOWLEDGMENT
We thank our co-workers from Heyrovsky Institute of Physical
Chemistry Jaromír Jirkovskꢁy and Michal Kolꢁaꢀr for providing
TiO2 nanoparticles. We also thank Magdalena Krulova for work
with mice during in vivo experiments. Support from the Ministry
of Education of the Czech Republic (Grants MSM0021620857
and SVV 261206/2010) is acknowledged. V.K. thanks RP
MSMT 14/63 for support. This work was carried out in the
ꢀ
framework of COST D38 Action (MSMT OC179).
(16) Mulder, W. J. M.; Strijkers, G. J.; Van Tilborg, G. A. F.;
Cormode, D. P.; Fayad, Z. A.; Nicolay, K. Nanoparticulate assemblies
of amphiphiles and diagnostically active materials for multimodality
imaging. Acc. Chem. Res. 2009, 42, 904–914.
’ ABBREVIATIONS USED
CA, contrast agent; TiO2@RhdGd, TiO2 nanoparticles modified
with Gd(III) complex and fluorescent dye; MRI, magnetic
resonance imaging; OI, optical imaging; EPR, enhanced perme-
ability and retention; ICP-AES, inductively coupled plasma
atomic emission spectroscopy; MSC, mesenchymal stem cells
(17) Gerion, D.; Herberg, J.; Bok, R.; Gjersing, E.; Ramon, E.;
Maxwell, R.; Kurhanewicz, J.; Budinger, T. F.; Gray, J. W.; Shuman,
M. A.; Chen, F. F. Paramagnetic silica-coated nanocrystals as an
advanced MRI contrast agent. J. Phys. Chem. C 2007, 111, 12542–12551.
(18) Koole, R.; van Schooneveld, M. M.; Hilhorst, J.; Castermans, K.;
Cormode, D. P.; Strijkers, G. J.; de Mello Donega, C.; Vanmaekelbergh,
D.; Griffioen, A. W.; Nicolay, K.; Fayad, Z. A.; Meijerink, A.; Mulder,
W. J. M. Paramagnetic lipid-coated silica nanoparticles with a fluorescent
quantum dot core: a new contrast agent platform for multimodality
imaging. Bioconjugate Chem. 2008, 19, 2471–2479.
(19) Zhang, Z.; Berns, A. E.; Willbold, S.; Buitenhuis, J. Synthesis of
poly(ethylene glycol) (PEG)-grafted colloidal silica particles with improved
stability in aqueous solvents. J. Colloid Interface Sci. 2007, 310, 446–455.
(20) Endres, P. J.; Paunesku, T.; Vogt, S.; Meade, T. J.; Woloschak,
G. E. DNAꢀTiO2 nanoconjugates labeled with magnetic resonance
contrast agents. J. Am. Chem. Soc. 2007, 129, 15760–15761.
(21) Thurn, K. T.; Paunesku, T.; Wu, A.; Brown, E. M. B.; Lai, B.;
Vogt, S.; Maser, J.; Aslam, M.; Dravid, V.; Bergan, R.; Woloschak, G. E.
Labeling TiO2 nanoparticles with dyes for optical fluorescence micro-
scopy and determination of TiO2-DNA nanoconjugate stability. Small
2009, 5, 1318–1325.
(22) Salih, F. M. Enhancement of solar inactivation of Escherichia coli
by titanium dioxide photocatalytic oxidation. J. Appl. Microbiol. 2002,
92, 920–926.
(23) Farshbaf Dadjour, M.; Ogino, C.; Matsumura, S.; Nakamura, S.;
Shimizu, N. Disinfection of Legionella pneumophila by ultrasonic treat-
ment with TiO2. Water Res. 2006, 40, 1137–1142.
(24) Rozhkova, E. A.; Ulasov, I.; Lai, B.; Dimitrijevic, N. M.; Lesniak,
M. S.; Rajh, T. A high-performance nanobio photocatalyst for targeted
brain cancer therapy. Nano Lett. 2009, 9, 3337–3342.
(25) Cai, R.; Kubota, Y.; Shuin, T.; Sakai, H.; Hashimoto, K.;
Fujishima, A. Induction of cytotoxicity by photoexcited TiO2 particles.
Cancer Res. 1992, 52, 2346–2348.
’ REFERENCES
(1) Prokop, A.; Davidson, J. M. Nanovehicular intracellular delivery
systems. J. Pharm. Sci. 2008, 97, 3518–3590.
ꢀ
ꢀ
(2) Etrych, T.; Chytil, P.; Mrkvan, T.; Sírovꢁa, M.; Ríhovꢁa, B.;
Ulbrich, K. Conjugates of doxorubicin with graft HPMA copolymers
for passive tumor targeting. J. Controlled Release 2008, 132, 184–192.
(3) Lebduꢀskovꢁa, P.; Kotek, J.; Hermann, P.; Vander Elst, L.; Muller,
R. N.; Lukeꢀs, I.; Peters, J. A. A gadolinium(III) complex of a carboxylic-
phosphorus acid derivative of diethylenetriamine covalently bound to
inulin, a potential macromolecular MRI contrast agent. Bioconjugate
Chem. 2004, 15, 881–889.
(4) Rudovskꢁy, J.; Botta, M.; Hermann, P.; Hardcastle, K. I.; Lukeꢀs, I.;
Aime, S. PAMAM dendrimeric conjugates with a GdꢀDOTA phosphi-
nate derivative and their adducts with polyaminoacids: the interplay of
global motion, internal rotation, and fast water exchange. Bioconjugate
Chem. 2006, 17, 975–987.
(5) Hrubꢁy, M.; Koꢀnꢁak, C.; Kuꢀcka, J.; Vꢀetꢀrík, M.; Filippov, S. K.;
Vꢀetviꢀcka, D.; Mackovꢁa, H.; Karlsson, G.; Edwards, K.; Ríhovꢁa, B.;
ꢀ
Ulbrich, K. Thermoresponsive, hydrolytically degradable polymer mi-
celles intended for radionuclide delivery. Macromol. Biosci. 2009,
9, 1016–1027.
(6) Michalet, X.; Pinaud, F. F.; Bentolila, L. A.; Tsay, J. M.; Doose, S.;
Li, J. J.; Sundaresan, G.; Wu, A. M.; Gambhir, S. S.; Weiss, S. Quantum
dots for live cells, in vivo imaging, and diagnostics. Science 2005,
307, 538–544.
(7) Eljamel, M. S. Fluorescence image-guided surgery of brain
tumors: explained step-by-step. Photodiagn. Photodyn. Ther. 2008,
5, 260–263.
(8) Trehin, R.; Figueiredo, J.-L.; Pittet, M. J.; Weissleder, R.;
Josephson, L.; Mahmood, U. Fluorescent nanoparticle uptake for brain
tumor visualization. Neoplasia 2006, 8, 302–311.
(26) Lai, T.-Y.; Lee, W.-C. Killing of cancer cell line by photoexcita-
tion of folic acid-modified titanium dioxide nanoparticles. J. Photochem.
Photobiol., A 2009, 204, 148–153.
(27) Ivankoviꢁc, S.; Gotiꢁc, M.; Jurin, M.; Musiꢁc., S. Photokilling
squamous carcinoma cells SCCVII with ultrafine particles of selected
metal oxides. J. Sol-Gel Sci. Technol. 2003, 27, 225–233.
(28) Huang, N.-P.; Xu, M.-H.; Yuan, C.-W.; Yu, R.-R. The study of
the photokilling effect and mechanism of ultrafine TiO2 particles on
U937 cells. J. Photochem. Photobiol., A 1997, 108, 229–233.
(9) Kotkovꢁa, Z.; Kotek, J.; Jirꢁak, D.; Jendelovꢁa, P.; Herynek, V.;
Berkovꢁa, Z.; Hermann, P.; Lukeꢀs, I. Cyclodextrin-based bimodal fluor-
escence/MRI contrast agents: an efficient approach to cellular imaging.
Chem.—Eur. J. 2010, 16, 10094–10102.
5193
dx.doi.org/10.1021/jm200449y |J. Med. Chem. 2011, 54, 5185–5194

Journal of Medicinal Chemistry
ARTICLE
(29) Kalbꢁaꢀcovꢁa, M.; Macak, J. M.; Schmidt-Stein, F.; Mierke, C. T.;
Schmuki, P. TiO2 nanotubes: photocatalyst for cancer cell killing. Phys.
Status Solidi RRL 2008, 2, 194–196.
ꢀ
(30) Rehoꢀr, I.; Kubíꢀcek, V.; Kotek, J.; Hermann, P.; Lukeꢀs, I.;
1
Szꢁakovꢁa, J.; Elst, L. V.; Muller, R. N.; Peters, J. A. H NMR relaxivity
of aqueous suspensions of titanium dioxide nanoparticles coated with a
gadolinium(III) chelate of a DOTA-monoamide with a phenylhospho-
nate pendant arm. J. Mater Chem. 2009, 19, 1494–1500.
ꢀ
(31) Rehoꢀr, I.; Kubíꢀcek, V.; Kotek, J.; Hermann, P.; Szꢁakovꢁa, J.;
Lukeꢀs, I. Modification of nanocrystalline TiO2 with phosphonate- and
bis(phosphonate)-bearing macrocyclic complexes: sorption and stability
studies. Eur. J. Inorg. Chem. 2011, 1981–1989.
(32) Kubíꢀcek, V.; Rudovskꢁy, J.; Kotek, J.; Hermann, P.; Elst, L. V.;
Muller, R. N.; Kolar, Z. I.; Wolterbeek, H. T.; Peters, J. A.; Lukeꢀs, I. A
bisphosphonate monoamide analogue of DOTA: a potential agent for
bone targeting. J. Am. Chem. Soc. 2005, 127, 16477–16485.
(33) Vitha, T.; Kubíꢀcek, V.; Hermann, P.; Kolar, Z. I.; Wolterbeek,
H. T.; Peters, J. A.; Lukeꢀs, I. Complexes of DOTA-bisphosphonate
conjugates: probes for determination of adsorption capacity and affinity
constants of hydroxyapatite. Langmuir 2008, 24, 1952–1958.
(34) Kubíꢀcek, V.; Vitha, T.; Kotek, J.; Hermann, P.; Vander Elst, L.;
Muller, R. N.; Lukeꢀs, I.; Peters, J. A. Towards MRI contrast agents
responsive to Ca(II) and Mg(II) ions: metal-induced oligomerization of
dota-bisphosphonate conjugates. Contrast Media Mol. Imaging 2010,
5, 294–296.
(35) Powell, D. H.; Dhubhghaill, O. M. N.; Pubanz, D.; Helm, L.;
Lebedev, Y. S.; Schlaepfer, W.; Merbach, A. E. Structural and dynamic
parameters obtained from 17O NMR, EPR, and NMRD studies of
monomeric and dimeric Gd3+ complexes of interest in magnetic
resonance imaging: an integrated and theoretically self-consistent ap-
proach. J. Am. Chem. Soc. 1996, 118, 9333–9346.
ꢁ
(36) Livramento, J. B.; Tꢁoth, E.; Sour, A.; Borel, A.; Merbach, A. E.;
Ruloff, R. High relaxivity confined to a small molecular space: a
metallostar-based, potential MRI contrast agent. Angew.Chem., Int. Ed.
2005, 44, 1480–1484.
(37) Kraitchman, D. L.; Gilson, W. D.; Lorenz, C. H. Stem cell
therapy: MRI guidance and monitoring. J. Magn. Reson. Imaging 2008,
27, 299–310.
(38) Arbab, A. S.; Frank, J. A. Cellular MRI and its role in stem cell
therapy. Regener. Med. 2008, 3, 199–215.
(39) Hirao, T.; Masunaga, T.; Ohshiro, Y.; Agawa, T. A novel
synthesis of dialkyl arenephosphonates. Synthesis 1981, 1, 56–57.
ꢀ
(40) Kolꢁaꢀr, M.; Mꢀeꢀst'ꢁankovꢁa, H.; Jirkovskꢁy, J.; Heyrovskꢁy, M.; Subrt,
J. Some aspects of physico-chemical properties of TiO2 nanocolloids
with respect to their age, size, and structure. Langmuir 2006, 22,
598–604.
5194
dx.doi.org/10.1021/jm200449y |J. Med. Chem. 2011, 54, 5185–5194