Preparation of Robust Metal-Free Magnetic Nanoemulsions Encapsulating Low-Molecular-Weight Nitroxide Radicals and Hydrophobic Drugs Directed Toward MRI-Visible Targeted Delivery

Article abstract of DOI:10.1002/chem.201702785

With a view to developing a theranostic nanomedicine for targeted drug delivery systems visible by magnetic resonance (MR) imaging, robust metal-free magnetic nanoemulsions (mean particle size less than 20 nm) consisting of a biocompatible surfactant and

Full text of DOI:10.1002/chem.201702785

A Journal of
Accepted Article
Title: Preparation of robust metal-free magnetic nanoemulsions
encapsulating low-molecular-weight nitroxide radicals and
hydrophobic drugs directed toward MRI-visible targeted delivery
Authors: Rui Tamura, Kota Nagura, Yusa Takemoto, Satori Moronaga,
Yoshiaki Uchida, Satoshi Shimono, Akihiko Shiino, Kenji
Tanigaki, Tsukuru Amano, Fumi Yoshino, Yohei Noda,
Satoshi Koizumi, Naoki Komatsu, Tatsuhisa Kato, and Jun
Yamauchi
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.201702785
Link to VoR: http://dx.doi.org/10.1002/chem.201702785
Supported by

10.1002/chem.201702785
Chemistry - A European Journal
FULL PAPER
Preparation of robust metal-free magnetic nanoemulsions
encapsulating low-molecular-weight nitroxide radicals and
hydrophobic drugs directed toward MRI-visible targeted delivery
Kota Nagura,[a] Yusa Takemoto,[a] Satori Moronaga, [a] Yoshiaki Uchida,[b,c] Satoshi Shimono,[a] Akihiko
Shiino,[d] Kenji Tanigaki,[e] Tsukuru Amano,[f] Fumi Yoshino,[f] Yohei Noda,[g] Satoshi Koizumi,[g] Naoki
Komatsu,[a] Tatsuhisa Kato,[a] Jun Yamauchi,[a] and Rui Tamura*[a]
Abstract: With a view to developing a theranostic nanomedicine for Introduction
magnetic resonance (MR) imaging-visible targeted drug delivery
Magnetic nanoparticles have attracted tremendous interest,
particularly in biomedical applications, such as drug delivery
system (DDS), hyperthermia therapy, and contrast enhancement
for magnetic resonance imaging (MRI).[1-3] The majority of
nanoparticles used are superparamagnetic iron oxides (magnetite
Fe3O4 or maghemite Fe2O3) or iron-iron oxide core-shell, which
are coated by a variety of polymeric or monomeric materials to
prevent aggregation and sedimentation of the nanoparticles.[4-13]
Practically these iron-containing nanoparticles, which are utilized
as ferrofluids comprising magnetic colloidal particles (particle size
< 100 nm) dispersed in a carrier fluid, can act as T2-enhancing
MRI contrast or hyperthermia therapy agents.[4-13] To employ
these magnetic nanoparticles as drug delivery carriers, various
drug molecules were attached to the biocompatible polymers,
silica layers, or phospholipid bilayers coated on the magnetic
core.[4-8] The use of such magnetic nanoparticles enables the
simultaneous delivery and detection of therapeutic agents in vivo.
However, there are some drawbacks associated with iron oxide-
based contrast agents, because they demonstrated unusual
magnetic susceptibility artifacts which produce dark signals that
may not only be misleading but also result in an incorrect
interpretation of the T2-weighted MR images.[3]
system, robust metal-free magnetic nanoemulsions (mean particle
sizes of less than 20 nm) consisting of the biocompatible surfactant 1
and the incorporated hydrophobic low-molecular-weight 2,2,5-
trimethyl-5-(4-alkoxy)phenylpyrrolidine-N-oxyl radicals [(±)-2] have
been prepared in pH 7.4 phosphate-buffered saline. The structure of
the nanoemulsions has been characterized by electron paramagnetic
resonance spectroscopy, and dynamic light scattering and small
angle neutron scattering measurements. The nanoemulsions showed
high colloidal stability, low cytotoxicity, enough reduction resistance to
excess ascorbic acid, and sufficient contrast enhancement in the
proton longitudinal relaxation time (T1)-weighted MR images in PBS
in vitro and preliminarily in vivo. Furthermore, additional hydrophobic
paclitaxel (7), an anticancer drug, could simultaneously be
encapsulated inside the nanoparticles, and the resulting 7-loaded
nanoemulsions were efficiently incorporated into HeLa cells to
suppress the cell growth.
[a]
K. Nagura, Y. Takemoto, S. Moronaga, Dr. S. Shimono, Prof. N.
Komatsu, Prof. T. Kato, Prof. J. Yamauchi, Prof. R. Tamura
Graduate School of Human and Environmental Studies
Kyoto University, Yoshida Nihonmatsu-cho, Sakyo-ku, Kyoto 606-
8501 (Japan)
Meanwhile, as biocompatible high-performance metal-free
magnetic nanoparticles, a core-shell type of organic radical
nanoparticles have been fabricated to be used as MRI and
electron paramagnetic resonance imaging (EPRI) probes in
vivo.[14-16] They comprised
a
self-assembling amphiphilic
E-mail: tamura.rui.8c@kyoto-u.ac.jp
[b]
[c]
[d]
[e]
[f]
Prof. Y. Uchida
copolymers with stable nitroxide radical groups covalently bonded
to the hydrophobic segment and showed high resistance to
bioreduction. In the case of these polymeric nanoparticles,
specific requirements such as degradability and accurate
molecular weight as well as nontoxicity and biocompatibility must
be satisfied.[7] As another example of metal-free magnetic
nanoparticles, low-toxic, lyotropic mesophase (hexagonal-phase)
nanoparticles (hexosomes) composed of amphiphilic glycerol
monooleate and nitroxide lipids were employed as MRI contrast
agents.[18,19] However, these mean particle sizes are much larger
than those (10 to 100 nm) required for MRI contrast agents which
are optimal for intravenous injection and have the most prolonged
blood circulation time.[6] In this context, branched-bottlebrush
copolymer nanoparticles with hydrodynamic diameters ranged
from 13 to 24 nm and large polypropylenimine-scaffolded
dendrimer molecules, both of which contain spirocyclohexyl
nitroxide polyradicals, were reported to show excellent stability
towards biological reducing agents and substantial in vivo MRI
contrast.[20,21] However, thus far these metal-free nanoparticles
have not possessed the function of DDS carriers.
Graduate School of Engineering Science
Osaka University, Toyonaka, Osaka 560-8531 (Japan)
Prof. Y. Uchida
PRESTO (Japan) Science and Technology Agency
Kawaguchi, Saitama 332-0012, (Japan)
Prof. A. Shiino
Biomedical MR Science Center
Shiga University of Medical Science, Seta, Otsu 520-2192 (Japan)
Dr. K. Tanigaki
Shiga Medical Center Research Institute
Moriyama 5-4-30, Shiga 524-8524 (Japan)
T. Amano, F. Yoshino
Department of Obstetrics and Gynecology
Shiga University of Medical Science, Seta, Otsu 520-2192 (Japan)
Dr. Y. Noda, Prof. S. Koizumi
[g]
Institute of Quantum Beam Science
Ibaraki University, Ibaraki 316-8511 (Japan)
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/chem.201702785
Chemistry - A European Journal
FULL PAPER
With these situation in mind, we have envisaged that if stable
and biocompatible oil-in-water (O/W) nanoemulsions with
appropriate particle sizes of 10 to 100 nm which can encapsulate
large amounts of low-molecular-weight nitroxide radical
molecules and show high resistance to bioreduction are easily
available, these radical nanoparticles are most likely to be used
as T1-enhancing MRI contrast agents and EPRI probes in vivo.
Furthermore, if hydrophobic drugs are incorporated into the same
magnetic nanoemulsions, they may be employed as the drug
carrier for MRI-visible targeted delivery system in vivo.
In this context, we reported that chiral all-organic rod-like
liquid crystalline compounds with a stable five-membered cyclic
nitroxide unit in the central core position exhibited a sort of spin
glass-like inhomogeneous ferromagnetic interactions (the
average spin-spin exchange interaction constant > 0) induced
J
by application of low magnetic fields in the LC phases at high
temperatures (30~150 °C).[22-26] By EPR spectroscopic studies
and DFT calculations, such a unique magnetic phenomenon,
which was referred to as ‘positive magneto-LC effect’, was
correlated with the preferential occurrence of ferromagnetic spin-
spin dipole interactions in magnetic fields by a spin polarization
mechanism due to the inhomogeneous intermolecular contacts
through CH/ and/or CH/O interactions between the neighboring
cyclic nitroxide radical moieties.[27] Furthermore, as the extension
of such a unique magnetic phenomenon to another organic
radical soft material system, we could prepare stable nitroxide
radical liquid microcapsules (300 m ) consisting of
monodispersed core-shell water-in-oil-in-water (W/O/W) double
emulsion droplets.[28] These emulsions were magnetically
transportable and served as a flexible antioxidative magnetic
carrier for on-demand cargo transport systems and droplet-based
sensors.
Here we report that the desired robust nanoemulsions
consisting of equimolar amounts of the biocompatible non-ionic
surfactant polyoxyethylene(20) cetyl ether
C16H33(OCH2CH2)20OH][29-31] and the incorporated low-molecular-
weight 2,2,5-trimethyl-5-(4-alkoxyphenyl)pyrrolidine-N-oxyl
1
[Brij®58,
radical [(±)-2] were easily obtained without adding any stabilizer
and that additional hydrophobic drugs could simultaneously be
incorporated into the nanoemulsions (Chart 1a). Furthermore,
these magnetic nanoemulsions exhibited sufficient contrast
enhancement in the proton longitudinal relaxation time (T1)-
weighted MR images in (–)-PBS in vitro and preliminarily in vivo.
Chart 1. (a) Molecular structures used in this work: non-ionic surfactant 1,
nitroxide radicals 2, 2,2,5-trimethyl-5-phenylpyrrolidinyloxy (3), 2,2,5,5-
tetramethylpyrrolidinyloxy (4), di-tert-butyl nitroxide (5), pyrene (6), paclitaxel (7),
tetraphenylporphyrin (8), and hydrocortisone (9). (b) Expected interdigitated
layer structure in the nanoemulsion consisting of equimolar amounts of 1 and
(±)-2. (c) Possible CH/ and CH/O interactions between the neighboring cyclic
nitroxide radical moieties inside the nanoemulsion. See SI for the synthesis of
(±)-2.
Results and Discussion
Preparation
and
characterization
of
magnetic
nanoemulsions
Referring to the smectic liquid crystalline (LC) layer structure
stabilized by interdigitated long alkyl chains, together with
intermolecular CH/ and/or CH/O interactions (Chart 1b and c),
the best combination of surfactant 1 (CMC: 0.007 to 0.077 mM)
bearing a terminal C16 alkyl chain and nitroxides (±)-2 with a long
alkyl chain (n = 14 to 20) or nitroxides (±)-3–5 with no alkyl chain
for comparison was searched to obtain most stable
nanoemulsions with a mean particle size of less than 100 nm.
This article is protected by copyright. All rights reserved.

10.1002/chem.201702785
Chemistry - A European Journal
FULL PAPER
Table 1. The mean particle sizes of the magnetic nanoemulsions composed of
1 and (±)-2 (n = 14, 16, 18 or 20) in PBS after 2 days from preparation at 25 °C.
(±)-2
n=14
55
n=16
26
n=18
17
n=20
18
Particle size (nm)
Figure 1. Characterization of the nanoemulsions prepared from equimolar
amounts of 1 and (±)-2 (n = 18) (final concentration: 10 mM each) in PBS. (a)
EPR spectrum of the sample filled in a capillary with the internal standard Mn2+
and (b) mean particle size (17 nm) determined by DLS measurement. See SI
for the experimental detail.
Figure 3. Mean particle sizes determined by DLS measurement of the
nanoemulsions immediately after preparation from equimolecular amounts of 1
and (±)-2 (final concentration: 10 mM each) in PBS. (±)-2 (n = 14) (red dashed
line), (±)-2 (n = 16) (blue dashed-dotted line), (±)-2 (n = 20) (black solid line),
and (±)-3 (green dotted line).
The nanoemulsions encapsulating hydrophobic (±)-2 (n = 18
or 20) showed outstanding stability; these mean particle sizes of
17 nm (in the range of 9–36 nm) or 18 nm (in the range of 8–30
nm) in PBS, respectively, were constant for more than two months
at 12 °C (Figure 1b and 3, and Table 1). In contrast, the micelle
solution (10 mmol) of 1 alone in PBS was unstable, giving a white
precipitate after 2 days at 12 °C. The mean particle size (17 nm)
of the nanoemulsions of (±)-2 (n = 18) with CMC of less than 0.01
mM was supported by small angle neutron scattering (SANS)
measurement in D2O; the SANS profile observed at iMATERIA
(BL20, J-PARC) was in good agreement with the scattering
function for spherical particles and the particle radius was
estimated to be 6.0 ± 2.0 nm (Figure 4 and for the detail, see SI).
In contrast, the nanoemulsions of (±)-2 (n = 14 or 16) disappeared
after 6 or 10 days to give a white suspension, respectively (Figure
3 and Table 1). Based on the mean diameter (9 nm) of the micelle
of 1 and the molecular length (3.3 nm) of 2 (n = 18) which was
calculated by the Molecular Mechanics using MM2 force field, the
mean particle size (17 nm) for the nanoemulsions of (±)-2 (n = 18)
indicates that both the flexible interdigitated layer structure of long
alkyl chains and the CH/ and/or CH/O interactions between
neighboring cyclic nitroxide radical moieties should greatly
contribute to the remarkable stability of the nanoemulsions (Chart
1b and c). Furthermore, the nanoemulsions of (±)-2 (n = 18 or 20)
were also stable after the 1000 times dilution (to 0.01 mM) with
PBS. Thus, the nanoemulsions of 1 and (±)-2 (n = 18 or 20) turned
out to be the best choice for biomedical application. For further
experiment, (±)-2 (n = 18) with the less molecular weight was used.
Meanwhile, the use of homologous non-ionic surfactants such as
Figure 2. EPR spectra of the nanoemulsions immediately after preparation from
equimolar amounts of 1 and (±)-2 (final concentration: 10 mM each) in PBS with
the internal standard Mn2+. (a) (±)-2 (n = 14), (b) (±)-2 (n = 16), (c) (±)-2 (n = 20),
and (d) (±)-3.
To a solution of 1 (10 mM) in pH 7.4 phosphate-buffered saline
(PBS) was added an equimolar amount of solid (±)-2 or oily (±)-
332 dissolved in a minimum amount of ether, and the mixture was
subjected to sonication (Branson Model 2510, power 125 W,
frequency 42 KHz) for 3 min at 25 °C to give a white suspension
or an oily mixture, respectively. Then the mixture was heated for
more than 10 min at 90 °C to produce ether-free transparent
nanoemulsions which were passed through a 0.45m membrane
filter (See SI for the experimental detail). The resulting
nanoemulsions showed a beautiful Tyndall scattering upon
irradiation of a visible laser light. In contrast, fairly water-soluble 4
and 5 failed to form emulsions. The stability and mean particle
size of the resulting nanoemulsions encapsulating (±)-2 or (±)-3
were evaluated by electron paramagnetic resonance (EPR)
spectroscopy and dynamic light scattering (DLS) measurement in
PBS. A broad singlet EPR signal due to the self-assembly of
nitroxide radical molecules within the nanoparticles was observed
for the nanoemulsions containing water-insoluble (±)-2 (n= 14, 16,
18 or 20) (Figure 1a and 2a-c). In contrast, the mixed spectrum of
a major three-line shape due to the free nitroxide radical
molecules and the minor broad singlet signal arising from
molecular aggregation was seen for fairly water-soluble (±)-3
(Figure 2d).
Brij®C10
[C16H33(OCH2CH2)10OH]
and
Brij®35
[C12H25(OCH2CH2)23OH] failed to prepare stable nanoemulsions
encapsulating (±)-2 (n = 14, 16, 18 or 20) in the concentration of
This article is protected by copyright. All rights reserved.

10.1002/chem.201702785
Chemistry - A European Journal
FULL PAPER
10 mM each in PBS, due most likely to the too short polyethylene
glycol or alkyl chain length of the surfactants, respectively.
DDS carrier ability
To evaluate the ability of
a
DDS carrier, the magnetic
nanoemulsions incorporating a hydrophobic quasi-drug pyrene
(6) were prepared by adding a solution of one equiv of (±)-2
(n=18) and 10 mol% of 6 (based on the amount of 1) in a small
amount of ether to a PBS solution (10 mM) of 1, followed by
sonication and heating at 90 °C according to the prototypical
procedure mentioned above. The resulting mean particle size was
17 nm, while that of the control nanoemulsions composed of 1
and 6 was 9.4 nm (Figure 6); these mean particle sizes were
comparable to those of the nanoemulsions (17 nm) and the
micelles (9 nm) in the absence of 6. The incorporation of 6 into
the nanoemulsions was confirmed by EPR, UV, and fluorescence
spectroscopies; the fluorescence emission of 6 within the
magnetic nanoemulsions was considerably quenched by the
surrounding nitroxyl groups[20] and revived by adding a large
excess (20 equiv based on the nitroxide molecules) of VC (Figure
7). These results support the interdigitated layer structure
between alkyl chains of 1 and (±)-2 (Chart 1b) in which molecules
of 6 were intercalated.
Figure 4. SANS profile of the nanoemulsion of 1 and (±)-2 (n = 18) in D2O. The
open circles were experimentally obtained. The solid line indicates the
calculated profile assuming spherical particles with radius R = 6.0 ± 2.0 nm. See
SI for the experimental detail.
Reduction resistance to ascorbic acid
It is well known that water-soluble nitroxide radicals are readily
reduced by antioxidants such as ascorbic acid (VC) and
glutathione (GSH) in vivo to give the corresponding
hydroxylamines.[14-16] Therefore, the reduction resistance of the
nanoemulsions of 1 and (±)-2 (n = 18) (each final concentration:
10 mM) was evaluated in the presence of excess VC (final
concentration: 17.5 mM) in PBS at 25 °C and compared with the
experimental result using the same concentration of water-soluble
4-hydroxy-2,2,6,6-tetramethyl-1-piperidinyloxy (TEMPOL).[14-16]
The decay of the spectral intensity for (±)-2 (n = 18) in response
to VC was monitored by EPR spectroscopy (Figure 5). The
nanoemulsions showed increased reduction resistance (half-life:
40 min) compared with TEMPOL (half-life: < 1 min); in fact we
could not track the decay of EPR spectral intensity for TEMPOL
due to the too fast reduction with VC. These results indicate that
the nitroxyl groups are shielded by surfactant 1, but slow water
exchange occurs between the inner and outer spheres of the
nanoemulsion most likely because of the polar circumstances in
the central pore surrounded by nitroxyl groups.
Figure 6. Determination of mean particle sizes (17 and 9.4 nm) of the
nanoemulsions of
1 incorporating 10 mol% of 6 by DLS measurement
immediately after preparation (a) in the presence of an equimolar amount of (±)-
2 (n=18) based on 1 and (b) in the absence of (±)-2, respectively, in PBS.
Figure 7. EPR, UV, and fluorescence spectra of the magnetic nanoemulsions
encapsulating 10 mol% of 6 in PBS. (a) EPR spectrum of the 6-loaded
nanoemulsions of 1 and (±)-2 (n=18) with the internal standard Mn2+, (b) UV
spectra of the 6-loaded nanoemulsions of 1 and (±)-2 (n=18) (solid line) and the
6-loaded nanoemulsions of 1 (broken line), (c) fluorescent spectra of the 6-
loaded nanoemulsions of 1 and (±)-2 (n=18) and the 6-loaded nanoemulsions
of 1 (broken line), and (d) fluorescence spectral change in the 6-loaded
nanoemulsions of 1 and (±)-2 after addition of 20 equiv of VC. See SI for the
experimental detail.
Figure 5. Resistance of (±)-2 (n = 18) in nanoemulsions to VC reduction. The
reduction in the spectral intensity in PBS was monitored by EPR spectroscopy
using a double integration method. See SI for the experimental detail.
Similarly, 10 mol% of hydrophobic drugs or quasi-drugs
such as paclitaxel (Taxol®) (7), tetraphenylporphyrin (8) and
This article is protected by copyright. All rights reserved.

10.1002/chem.201702785
Chemistry - A European Journal
FULL PAPER
hydrocortisone (9) were successfully incorporated into the
magnetic nanoparticles by an analogous procedure. The mean
particle size of the nanoemulsions encapsulating 7 was 12 nm by
DLS measurement, while that incorporating 8 or 9 was 16 nm
(Figure 8 d-f). Furthermore, the EPR spectrum of 7-loaded
nanoemulsions was a broad three-line shape, whereas 8 or 9-
loaded ones showed a broad singlet signal (Figure 8a-c). It was
noticeable that the former exhibited the spectrum of a dispersed
nitroxide radical in solution in contrast to the latter ones of self-
assembled radicals. These results suggest that the bulky 7
bearing hydrogen-bond donating NH and OH groups should
reside in the central position of nanoemulsions and interact with
neighboring nitroxyl groups by hydrogen bond to form a quasi-
solution, resulting in i) the formation of a deeply interdigitated
layer structure between alkyl chains so as to decrease the mean
diameter of the nanoemulsions and ii) a considerably slow rotation
of the nitroxyl groups, both of which are advantageous to enlarge
the MRI contrast ability as described later. In contrast, the
nanoemulsions incorporating 8 or 9 with a fairly planar molecular
structure would take an intercalated molecular assembly in the
interdigitated layer, similar to the case of 6-loaded ones.
low as 10M of 7 was diluted with PBS by 1/8 [final concentration
of 1 and (±)-2 (n=18): 1.25 mM each)], 1/16 (0.625 mM), 1/32
(0.313 mM), 1/64 (0.156 mM), 1/128 (0.078 mM) and 1/256 (0.039
mM) before incubation. After incubation for 24 h at 37 °C under
5% CO2, cells were washed with PBS twice and the cytotoxicity
was evaluated by Cell Counting Kit-8 (CCK-8) assay.
Consequently, both the micelles and 7-unloaded nanoemulsions
showed high toxicity in the concentrations of 0.625 mM (1/16
dilution) or higher, while they exhibited much lower or little toxicity
in the concentration of 0.313 mM (1/32) or lower (Figure 9a). In
contrast, both the 7-loaded nanoemulsions and free 7 displayed
significant toxicity against Hela cells in the concentration of
0.156M (1/64 dilution) or higher of 7. Thus, the fact that the drug
7 released from the nanoemulsions is pharmaceutically active
suggests that 7-loaded nanoemulsions were incorporated into the
cancer cells and are useful as a DDS drug carrier.
Figure 9. Evaluation of in vitro cytotoxicity of 7-loaded magnetic nanoemulsions.
HeLa cells in a culture medium were treated with a PBS solution of (a) the
micelles of 1 and the 7-unloaded nanoemulsions composed of 1 and (±)-2
(n=18), and (b) the 7-loaded nanoemulsions and free 7. The initial solution of
the nanoemulsions containing 10 mM of 1 and (±)-2 (n=18) and 10 M of 7 was
diluted with PBS by 1/8, 1/16, 1/32, 1/64, 1/128 and 1/256 before incubation;
the final concentrations of 1 and (±)-2 (n=18) were 1.25, 0.625, 0.313, 0.156,
0.078 and 0.039 mM, while those of 7 were 1.25, 0.625, 0.313, 0.156, 0.078
and 0.039 M, respectively. The cell viability was assayed using CCK-8 kit after
incubation for 24 h at 37 °C. See SI for the experimental detail.
Figure 8. (a-c) EPR spectra with the internal standard Mn2+ and (d-f) mean
particle sizes determined by DLS measurement immediately after preparation
of the nanoemulsions of 1 and (±)-2 (n=18) encapsulating 10 mol% of (a, d) 7,
(b, e) 8, and (c, f) 9 in PBS.
MRI measurement
The water proton relaxivity of the magnetic nanoemulsions
containing 1.25 to 10 mM of 1 and (±)-2 (n=18) in PBS was
evaluated on an MRI machine at 4.7 T (Figure 10a, A-D). The
nanoemulsions prepared from the 10 mM (final concentration)
solution of 1 and (±)-2 (n=18) significantly brightened the T1-
weighted MR phantom images compared with PBS alone (Figure
10a, L). Furthermore, we could observe the bright MR phantom
images of the magnetic nanoemulsions incorporating 10 mol% of
quasi-drugs 6 and 8 (Figure 10a, I and J) or drugs 7 and 9 (Figure
10a, E and K). Particularly, the 7-loaded nanoemulsions exhibited
In vitro cytotoxicity
To verify the uptake of the nanoemulsions into cancer cells, HeLa
cells (9.0 x 103 cells in 100 L per well) cultured in a 96-well dish
with Dulbecco’s modified Eagle’s medium (DMEM) containing
10% fetal bovine serum and 1% penicillin/streptomycin were
treated with a PBS solution (10 L) of the micelles of 1, the 7-
unloaded or loaded nanoemulsions composed of 1 and (±)-2
(n=18), or free
7 (Figure 9). The initial solution of the
nanoemulsions containing 10 mM of 1 and (±)-2 (n=18) and as
This article is protected by copyright. All rights reserved.

10.1002/chem.201702785
Chemistry - A European Journal
FULL PAPER
the brightest contrast among them including the 7-unloaded
nanoemulsions (Figure 10a, E). This is likely associated with the
decrease in the mean particle size (12 nm) and the unique EPR
spectrum as described in the previous section (Figure 8a and d),
which allows relatively fast water exchange between the inner and
outer spheres of the nanoemulsions and the increase in rotational
correlation constant of nitroxyl groups.[18,19]
injected into an anesthetized mouse (weight around 30 g) lying
supinely in an animal holder. Subsequently, the cradle was
inserted into the MRI machine (7.0 T) to acquire T1-weighted MR
images by the spin-echo method. Since it was anticipated that the
injected nanoemulsions would be transported by the blood stream
and accumulated to the pituitary gland in the brain, T1 weighted
MR images of the pituitary gland were taken after the injection of
To evaluate the relaxation rate enhancement by the nitroxide-
based magnetic nanoemulsions, the longitudinal relaxation rate
(r1) was determined from the relaxation times as a function of
concentration by using spin-echo method at 4.7 T with respect to
the 7-unloaded and loaded nanoemulsions; linear regression
analysis yielded the r1 of 0.12 and 0.16 mM–1s–1, respectively
(Figure 10 b and c). These r1 values were less than that (4.0 mM–
1s–1 at 4.7 T) of Magnevist® [one of Gd(III) complex agents] in
deionized water,[33] but comparable or larger than those (in the
order of 0.1 mM–1 s–1 at 4.7 T) for ordinary water-soluble small
mononitroxide molecules.[34] However, the fact that there is a good
linear relation between the T1 contrast enhancement and the
concentration of nitroxyl groups would increase the significance
of our nanoemulsions for biomedical application.[21,35]
Furthermore, it is a big advantage that nitroxide radicals can act
as T1-enhancing MRI contrast agents with much lower toxicity in
comparison with the commonly used molecular Gd(III) complex
agents.[36-40]
the nanoemulsions. As a result, a distinct MRI contrast
enhancement was observed in the pituitary gland after injection
with high reproducibility (Figure 11 and S1). This result
demonstrates that the nanoemulsions are effective as an in vivo
T1-shortening MRI contrast agent. Further detailed experiment to
clarify where the magnetic nanoemulsions accumulate in healthy
and tumor-bearing mice will be reported in due course.
Figure 11. T1-weighted MR images of brain tissues of two mice: (a, c) before
and (b, d) 13 min after the intravenous injection of 300 L of the nanoemuslions
composed of 1 and (±)-2 (n=18) (10 mM each) in PBS. The images in (a) and
(b) correspond to one mouse, and those in (c) and (d) refer to the other mouse.
Distinct contrast enhancement was observed in the pituitary glands of both mice
(indicated by black arrows). See SI for the experimental detail.
Conclusions
We have successfully prepared the ideal and robust, metal-
free magnetic nanoemulsions with a desired mean particle size
(17 nm) by simply heating a mixture of equimolar amounts of
biocompatible non-ionic surfactant 1 and low-molecular weight
nitroxide radical (±)-2 (n=18) at 90 °C in PBS, both of which
possess long alkyl chains with similar length. The magnetic
nanoemulsions exhibited high reduction resistance to ascorbic
acid and sufficiently bright contrast in the proton T1-weighted MR
images in vitro and preliminarily in vivo, and incorporated as much
as 10 mol % of hydrophobic drugs or quasi-drugs into the
nanoparticles. From the cell viability assay, it has been concluded
that the magnetic nanoemulsions show low toxicity in the
concentrations (0.313 mM or lower) of 1 and (±)-2 (n=18) and that
the anticancer drug 7-loaded nanoemulsions effectively suppress
the HeLa cell growth in the concentrations of 0.078 M or higher
of 7. Thus, it is highly expected that these nanoemulsions will
serve as a theranostic nanomedicine for MRI-visible targeted drug
delivery system in vivo, if they have an additional function to be
Figure 10. T1-weighted MR phantom images of various magnetic
nanoemulsions in PBS at 4.7 T and determination of the longitudinal relaxation
rates (r1). (a) Panels (A-D) are the images of the magnetic nanoemulsions
composed of equimolar amounts of 1 and (±)-2 (n=18) (final concentrations: 10,
5, 2.5 and 1.25 mM, respectively). Panels (E-H) are the images of the magnetic
nanoemulsions composed of 1 and (±)-2 (n=18) (10, 5, 2.5 and 1.25 mM,
respectively) encapsulating 10 mol% of 7. Panels (I-K) are the images of the
magnetic nanoemulsions composed of 1 and (±)-2 (n=18) (10 mM each)
encapsulating 10 mol% of (I) 6, (J) 8, or (K) 9. The control image of PBS alone
is shown in panel (L). Plots of T1–1 vs the concentration of (±)-2 (n=18) in (b) the
7-unloaded nanoemulsions and (c) the 7-loaded nanoemulsions; the r1 values
were determined to be 0.12 and 0.16 mM–1s–1, respectively. See SI for the
experimental detail.
Finally, to examine whether the magnetic nanoemulsions
can actually serve as an in vivo MRI contrast agent, a preliminary
MRI experiment was performed using living mouse brain tissues.
Three hundred microliter (L) of the nanoemulsions composed of
1 and (±)-2 (n=18) (10 mM each) in PBS were intravenously
This article is protected by copyright. All rights reserved.

10.1002/chem.201702785
Chemistry - A European Journal
FULL PAPER
selectively incorporated and accumulated into the targeted cancer
cells.
[19] B. W. Muir, D. P. Acharya, D. F. Kennedy, X. Mulet, R. A. Evans, S. M.
Pereira, K. L. Wark , B. J. Boyd, T. Nguyen, T. M. Hinton, L. J.
Waddington, N. Kirby, D. K. Wright, H. X. Wang, G. F. Egan, B. A.
Moffat, Biomaterials 2012, 33, 2723-2733.
[20] A. Sowers, J. R. McCombs, Y. Wang, J. T. Palette, S. W. Morton, E. C.
Dreaden, M. D. Boska, M. F. Ottaviani, P. T. Hammond, A. Rajca, J. A.
Johnson, Nat. Commun. 2014, 5, 5460.
Acknowledgements
The present work was supported by JSPS KAKENHI (Grant
number 26248024). The neutron experiment at the Materials and
Life Science Experimental Facility of the J-PARC was performed
under a user program (Proposal No. 2016PM0005).
[21] A. Rajca, Y. Wang, M. Boska, J. T. Paletta, A. Olankitwanit, M. A.
Swanson, D. G. Mitchell, S. S. Eaton, G. R. Eaton, S. Rajca, J. Am.
Chem. Soc. 2012, 134, 15724-15727.
[22] Y. Uchida, K. Suzuki, R. Tamura, N. Ikuma, S. Shimono, Y. Noda, J.
Yamauchi, J. Am. Chem. Soc. 2010, 132, 9746-9752.
[23] K. Suzuki, Y. Uchida, R. Tamura, S. Shimono, J. Yamauchi, J. Mater.
Chem. 2012, 22, 6799-6806.
Keywords: Radicals• EPR spectroscopy • magnetic
nanoparticles • imaging agents• drug delivery
[24] Y. Uchida, K. Suzuki, R. Tamura, J. Phys. Chem. B 2012, 116, 9791-
9795.
[1] W. J. Stark, Angew. Chem. Int. Ed. 2011, 50, 1242-1258.
[2] H. Goesmann, C. Feldmann, Angew. Chem. Int. Ed. 2010, 49, 1362-
1395.
[25] R. Tamura, Y. Uchida, K. Suzuki in Handbook of Liquid Crystals:
Second Edition (Eds: J. W. Goodby, P. J. Collings, T. Kato. C.
Tschierske, H. F. Gleeson, P. Raynes), Wiley-VCH, Weinheim, 2014
,
[3] Magnetic Nanomaterials: Nanomaterials for the Life Sciences Vol. 4
vol. 8, pp. 837-864.
(Ed. C. Kumar), Wiley-VCH, Weinheim, 2009
.
[26] K. Suzuki, Y. Takemoto, S. Takaoka, K. Taguchi, Y. Uchida, D. G.
Mazhukin, I. G. Grigor’ev, R. Tamura, Chem. Commun. 2016, 52, 3935-
3938.
[4] J. Huang, Y. li, Y.A. Orza, Q. Lu, P. Guo, L Wang, L. Yang, H. Mao,
Adv. Funct. Mater. 2016, 26, 3818-3836.
[5] Z. Al-Ahmady, K. Kostarelos, Chem. Rev. 2016, 116, 3883-3918
[6] S. Laurent, D. Forge, M. Port, A. Roch. C. Robic, L. V. Elst, R. N. Muller,
Chem. Rev. 2008, 108, 2064-2110.
[27] A. Kh. Vorobiev, N. A. Chumakova, D. A. Pomogailo, Y. Uchida, K.
Suzuki, Y. Noda, R. Tamura, J. Phys. Chem. B 2014, 118, 1932-1042.
[28] Y. Uchida, Y. Iwai, T. Akita, T. Mitome, K. Suzuki, R. Tamura, N.
Nishiyama, J. Mater. Chem. B 2014, 2, 4130-4133.
[29] J. Chamieh, F. Davanier, V. Jannin, F. Demarne, H. Cottet, Inter. J.
Pharm. 2015, 492, 46-54.
[7] A-H. Lu, E. L. Salabas, F. Shüth, Angew. Chem. Int. Ed. 2007, 46, 1222-
1244.
[8] C. Corot, P. Robert, J-M. Idée, M. Port, Adv. Drug Deliv. Rev. 2006, 58,
1471-1504.
[30] L. Dupont-Leclercq, S. Giroux, B. Henry, P. Rubini, Langmuir 2007, 23,
10463-10470.
[9] M. Piazza, M. Colombo, I. Zanoni, F. Granucci, P. Tortora, J. Weiss, T.
Gioannini, D. Prosperi, F Peri, Angew. Chem. Int. Ed. 2011, 50, 622-
626.
[31] J. Schefer, R. McDaniel, B. Schoenborn, J. Phys. Chem. 1988, 92, 729-
732.
[10] H. Chen, C. Deng, X. Zhang, Angew. Chem. Int. Ed. 2010, 49, 607-611.
[11] L. Zao, T. Chano, S. Morikawa, Y. Saito, A. Shiino, S. Shimizu, T.
Maeda, T. Irie, S. Aonuma, H. Okabe, T. Kimura, T. Inubushi, N.
Komatsu, Adv. Funct. Mater. 2012, 22, 5107-5117.
[32] K. Hideg, H. O. Hankovszky, H. A. Halasz, J. Chem. Soc., Perkin Trans.
1
1988, 2905-2911.
[33] L. Zhao, A. Shiino, H. Qin, T. Kimura, N. Komatsu, J. Nanosci.
Nanotechnol. 2015, 15, 1076-1082.
[12] G. Mikhaylov, U. Mikac, A. A. Magaeva, V. I. Itin, E. P. Naiden, I.
Psakhye, L. Babes, T. Reinheckel, C. Peters, R. Zeiser, M. Bogyo, V.
[34] K. Matsumoto, Biol. Pharm. Bull. 2009, 32, 711-716.
[35] K. Matsumoto, H. Yakumaru, M. Narazaki, H. Nakagawa, K. Anzai, H.
Ikehara, N. Ikota, Magn. Reson. Imag. 2008, 26, 117-121.
[36] D. Bardelang, M. Hardy, O. Ouari, P. Tordo in Encyclopedia of Radicals
in Chemistry, Biology and Materials, Vol.4 Polymers and Materials (Eds:
C. Chatgilialoglu, A. Studer), John Wiley & Sons, Chichester, 2012, pp.
1965-2015.
Turk, S. G. Psakhye, B. Turk, O. Vasiljeva, O. Nature Nanotech. 2011
6, 594-602.
,
[13 S. Cheong, P. Ferguson, K. W. Feindel, I. F. Hermans, P. T. Callaghan,
C. Meyer, A. Slocombe, C-H. Su, F-Y. Cheng, C-S. Yeh, B. Ingham, M.
F. Toney, R. D. Tilley, Angew. Chem. Int. Ed. 2011, 50, 4206-4209.
[14] X. Zhuang, C. Xiao, K. Oyaizu, N. Chikushi, X. Chen, H. Nishide, J.
Poly. Sci. Part A: Polym. Chem. 2010, 48, 5404-5410.
[37] Dynamic Contrast-Enhanced Magnetic Resonance Imaging in
Oncology (Eds.: A. Jackson, D. L. Buckley, G. J. M. Parker), Springer-
[15] T. Yoshitomi, R. Suzuki, T. Mamiya, H. Matsui, A. Hirayama, Y.
Nagasaki, Bioconjugate Chem. 2009, 20, 1792-1798.
Verlag, Berlin Heidelberg, 2005
.
[38] Z. Zhelev, R. Bakalova, I. Aoki, K. Matsumoto, V. Gadjeva, K. Anzai, I.
Kanno, Chem. Commun. 2009, 45, 53-55.
[16] T. Yoshitomi, D. Mikyamoto, Y. Nagasaki, Biomacromolecules 2009
10, 596-601.
,
[39] X. Wan, T. C. Fu, A. Funk, A.; R. E. London, Brain Res. Bull. 1995, 36,
91-96.
[17] E. Soussan, S. Cassel, M. Blanzat, I. Rico-Lattes, Angew. Chem. Int.
Ed. 2009, 48, 274-288.
[40] K. Matsumoto, M. Narazaki, H. Ikehira, K. Anzai, N. Ikota, J. Magn.
Reson. 2007, 187, 155-162.
[18] N. Bye, O. E. Hutt, T. M. Hinton, D. P. Acharya, L. J. Waddington, B. A.
Moffat, D. K. Wright, H. X. Wang, X. Mulet, B. W. Muir, Langmuir 2014
30, 8898-8906.
,
This article is protected by copyright. All rights reserved.

10.1002/chem.201702785
Chemistry - A European Journal
FULL PAPER
Entry for the Table of Contents (Please choose one layout)
FULL PAPER
Metal-free magnetic nanoemulsions
encapsulating hydrophobic drugs were
prepared with a view to developing a
theranostic nanomedicine for MRI-
visible targeted drug delivery system.
K. Nagura, Y. Takemoto, S. Moronaga, Y.
Uchida, S. Shimono, A. Shiino, K.
Tanigaki, T. Amano, F. Yoshino, Y. Noda,
S. Koizumi, N. Komatsu, T. Kato, J.
Yamauchi, R. Tamura*
Page No. – Page No.
Preparation of robust metal-free
magnetic nanoemulsions
encapsulating low-molecular-weight
nitroxide radicals and hydrophobic
drugs directed toward MRI-visible
targeted delivery
This article is protected by copyright. All rights reserved.