A design strategy for small molecule-based targeted MRI contrast agents: Their application for detection of atherosclerotic plaques

Article abstract of DOI:10.1039/c4ob01270d

Gadolinium(III) ion (Gd³⁺) complexes are widely used as contrast agents in magnetic resonance imaging (MRI), and many attempts have been made to couple them to sensor moieties in order to visualize biological phenomena of interest inside the bo

Full text of DOI:10.1039/c4ob01270d

Organic &
Biomolecular Chemistry
View Article Online
View Journal
PAPER
A design strategy for small molecule-based
targeted MRI contrast agents: their application
Cite this: DOI: 10.1039/c4ob01270d
for detection of atherosclerotic plaques†
Shimpei Iwaki,a Kazuya Hokamura,b Mikako Ogawa,c Yasuo Takehara,d
Yasuaki Muramatsu,a Takehiro Yamane,a Kazuhisa Hirabayashi,a Yuji Morimoto,e
Kohsuke Hagisawa,f Kazuhide Nakahara,g Tomoko Mineno,g Takuya Terai,a
Toru Komatsu,a,h Tasuku Ueno,a Keita Tamura,i Yusuke Adachi,i Yasunobu Hirata,j
Makoto Arita,a Hiroyuki Arai,a Kazuo Umemura,b Tetsuo Naganok and
Kenjiro Hanaoka*a
Gadolinium(III) ion (Gd3+) complexes are widely used as contrast agents in magnetic resonance imaging
(MRI), and many attempts have been made to couple them to sensor moieties in order to visualize bio-
logical phenomena of interest inside the body. However, the low sensitivity of MRI has made it difficult to
develop practical MRI contrast agents for in vivo imaging. We hypothesized that practical MRI contrast
agents could be designed by targeting a specific biological environment, rather than a specific protein
such as a receptor. To test this idea, we designed and synthesized a Gd3+-based MRI contrast agent,
2BDP3Gd, for visualizing atherosclerotic plaques by linking the Gd3+-complex to the lipophilic fluoro-
phore BODIPY to stain lipid-rich environments. We found that 2BDP3Gd was selectively accumulated into
lipid droplets of adipocytes at the cellular level. Atherosclerotic plaques in the aorta of Watanabe heritable
hyperlipidemic (WHHL) rabbits were clearly visualized in T1-weighted MR images after intravenous injec-
tion of 2BDP3Gd in vivo.
Received 19th June 2014,
Accepted 13th August 2014
DOI: 10.1039/c4ob01270d
www.rsc.org/obc
Introduction
Magnetic resonance imaging (MRI) is widely used for noninva-
aGraduate School of Pharmaceutical Sciences, The University of Tokyo, 7-3-1 Hongo,
Bunkyo-ku, Tokyo 113-0033, Japan. E-mail: khanaoka@mol.f.u-tokyo.ac.jp;
Fax: +81 3 5841 4855; Tel: +81 3 5841 4852
sive imaging in both clinical medicine and life sciences. MR
images are constructed from 1H NMR signals mainly derived
from water protons, so the signal intensity depends upon the
concentration and relaxation times (T1 and T2) of water mole-
cules. Paramagnetic ions such as the gadolinium(III) ion (Gd3+)
effectively shorten the T1 (spin–lattice) relaxation time of water
protons by promoting rapid exchange of inner-sphere water
molecules with bulk water, and enhance the signal intensity in
T1-weighted MR images. Gd3+ complexes are widely used as
MRI contrast agents to enhance tissue contrast inside the
body.1
Compared with other imaging modalities such as fluo-
rescence imaging, ultrasound and PET, MRI provides high
spatial resolution deep within tissues.2 Therefore, its appli-
cation for in vivo imaging is important, and many Gd3+
complex-based MRI contrast agents have been developed for
this purpose.1,3 In general, there are two design strategies for
functional MRI contrast agents. One is to employ “smart” MRI
contrast agents that induce a change of the water proton relax-
ation time (T1 or T2) when they are structurally modified by
bDepartment of Pharmacology, Hamamatsu University School of Medicine,
1-20-1 Handayama, Hamamatsu, Shizuoka 431-3192, Japan
cPhoton Medical Research Center, Hamamatsu University School of Medicine,
1-20-1 Handayama, Hamamatsu, Shizuoka 431-3192, Japan
dDepartment of Radiology, Hamamatsu University School of Medicine,
1-20-1 Handayama, Hamamatsu, Shizuoka 431-3192, Japan
eDepartment of Integrative Physiology and Bio-Nano Medicine, National Defense
Medical College, Namiki 3-2, Tokorozawa, Saitama 359-8513, Japan
fDepartment of Physiology, National Defense Medical College, Namiki 3-2,
Tokorozawa, Saitama 359-8513, Japan
gFaculty of Pharmacy, Takasaki University of Health and Welfare, 60 Nakaorui,
Takasaki-shi, Gunma 370-0033, Japan
hJST, PRESTO, 5 Sanbancho, Chiyoda-ku, Tokyo 102-0075, Japan
iDepartment of Physiology, Graduate School of Medicine, The University of Tokyo,
7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan
jDepartment of Cardiovascular Medicine, Graduate School of Medicine, The
University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan
kOpen Innovation Center for Drug Discovery, The University of Tokyo, 7-3-1 Hongo,
Bunkyo-ku, Tokyo 113-0033, Japan
†Electronic supplementary information (ESI) available: Supplementary methods,
figures, tables and video. See DOI: 10.1039/c4ob01270d
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
exposure to a target, such as a specific enzyme activity.4–6 The atherosclerotic plaques.17,18 So, we anticipated that the gadoli-
other is to conjugate Gd3+ complexes to specific ligands or nium ion complex conjugated with these dyes would accumu-
antibodies directed to the targeted biomolecules.5,7,8 However, late in the lipid cores of atherosclerotic plaques, irrespective of
there are few examples of application of functional MRI con- the concentrations of specific target biomolecules. However,
trast agents based on the former strategy for in vivo whole- general protocols for staining with Sudan IV and Oil red O
body imaging because it is very difficult to control the bio- include incubation with organic solvents such as isopro-
distribution within the body. In the case of the latter strategy, panol,19 so that these scaffolds would not be suitable for
the low sensitivity of the MRI technique makes it difficult to in vivo imaging. Instead, we focused on the lipophilic fluoro-
develop practical MRI contrast agents for in vivo imaging.3 In phore BODIPY (boron dipyrromethene) as a candidate athero-
general, the detection limit of the Gd3+ complex in MRI sclerotic targeting moiety, since BODIPY is also used for
measurements is around several μM,9 whereas target bio- staining lipid droplets in adipocytes, but does not require
molecules are often present only at the nanomolar order of incubation in an organic solvent.16,19 Thus, we expected that
concentration.10,11 To achieve sensitive detection, a new strat- the BODIPY-Gd3+ complex conjugate would be accumulated in
egy to achieve sufficient accumulation of MRI contrast agents atherosclerotic plaques in vivo.
in the target region is required. Toward this end, we hypo-
thesized that MRI contrast agents targeted to specific biological
environments of interest, rather than to specific proteins such
Results
as receptors, could be efficiently accumulated in the target
Molecular design of MRI contrast agents for atherosclerotic
region at a sufficient concentration to allow clear MRI-based
plaques
visualization of biological phenomena, irrespective of the con-
To validate the concept, we designed and synthesized a Gd3+
centrations of specific target biomolecules.
complex conjugated with a BODIPY moiety, BDP-Gd,20 as a
To validate this concept, we tried to develop MRI contrast
MRI contrast agent candidate for atherosclerotic plaques
(Fig. 1a). BODIPY is a widely used fluorophore which has a
high extinction coefficient (ε) and a high quantum yield of
fluorescence (Φfl).21 An additional advantage of BODIPY is that
the contrast agent would be detectable not only by MRI, but
also by fluorescence measurements, because the BODIPY
moiety is a fluorophore. With such MRI-fluorescence dual-
modal probes,22,23 the contrast agent distribution inside cells
or tissues can be precisely evaluated by means of fluorescence
microscopy observation.
agents for atherosclerosis. Atherosclerosis is a progressive
disease characterized by accumulation of lipids and fibrous
elements in the form of atherosclerotic plaques in large
arteries.12 Its most important clinical complication is acute
occlusion due to the formation of a thrombus or blood clot,
which is usually associated with plaque rupture, and results
in myocardial infarction or stroke.13,14 Atherosclerosis is a
primary cause of heart disease and stroke, and is the under-
lying cause of about 30% of all deaths in westernized
societies.15 Therefore, the detection of atherosclerotic plaques
is very important both in clinical medicine and life science. To
develop suitable MRI contrast agents, we focused on staining
dyes for atherosclerotic plaques. Sudan IV and Oil red O are
widely used in histology for staining not only lipid droplets,
which are lipid storage compartments surrounded by a phos-
pholipid monolayer, in adipocytes,16 but also the lipid cores of
Fluorescence imaging of contrast agent distribution in live cells
First, we applied BDP-Gd to live adipocytes and observed its intra-
cellular localization with a fluorescence microscope (Fig. 1b).
Strong fluorescence of the BODIPY moiety was observed selec-
tively in lipid droplets in adipocytes, but not in preadipocytes, as
Fig. 1 (a) Chemical structure of BDP-Gd. (b) Confocal microscopy images of adipocytes (left) and preadipocytes (right) loaded with BDP-Gd (10 μM
in DMEM containing 0.1% DMSO as a cosolvent). Upper and lower panels show DIC and fluorescence images, respectively. (c) Fluorescence (left)
and Sudan IV staining (right) images of the aorta isolated from ApoE−/− or wild-type mice injected with BDP-Gd (1st injection; 5 mM, 100 μL. 2nd
injection; 5 mM, 150 μL, 2 h after 1st injection). Fluorescence images were captured 1 h after the 2nd injection.
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
expected (Fig. 1b). This result indicates that BDP-Gd is indeed
selectively accumulated in lipid-rich environments such as lipid
droplets inside live cells.
Ex vivo fluorescence imaging of aorta of ApoE−/− mice
We expected that BDP-Gd would also be accumulated in athero-
sclerotic plaques in vivo. To test this idea, we first performed
fluorescence imaging of atherosclerotic plaques of ApoE−/−
mice, which are widely used mouse models of hyperlipidemia
and atherosclerosis.17 Apolipoprotein E (ApoE) is the major com-
ponent of lipoprotein and facilitates hepatic uptake of lipopro-
teins.17 Anesthetized mice were injected with BDP-Gd into the
orbital vein and sacrificed 3 h later. Strong fluorescence of the
BODIPY moiety was observed from atherosclerotic plaques in the
isolated aorta of ApoE−/− mice, and was colocalized with Sudan
IV staining, while no marked fluorescence was seen in the case
of wild-type mice (Fig. 1c). Sudan IV staining is often used to
identify atherosclerotic plaques, so this result indicates that
BDP-Gd accumulates preferentially in atherosclerotic plaques of
ApoE−/− mice.
Derivatization of BDP-Gd for in vivo application
Next, we applied BDP-Gd to MR imaging of atherosclerotic
plaques of Watanabe heritable hyperlipidemic (WHHL)
rabbits. The WHHL rabbit is an animal model of atherosclero-
sis caused by deficiency of the low density lipoprotein (LDL)
receptor and is considered to be a better animal model of
human atherosclerosis than ApoE−/− mice because WHHL
Fig. 2 Fluorescence images of adipocytes loaded with BDP-Gd deriva-
tives. Confocal microscopy images of adipocytes loaded with BDP-Gd
rabbits have similar lipoprotein metabolism to humans.24 derivatives (10 μM in DMEM containing 0.1% DMSO as a cosolvent). (Top
left) BDP-Gd, (top right) BDP-DO3A-Gd, (middle left) BDP-thio-
DO3A-Gd, (middle right) BDP-thioPh-DO3A-Gd, (bottom left) BDP-
thio-BnDOTA-Gd, (bottom right) BDP-thioPh-BnDOTA-Gd.
However, no clear increment of MR signal in atherosclerotic
plaques was seen in WHHL rabbits injected with BDP-Gd (data
not shown). The reasons for this may be that the sensitivity of
BDP-Gd was insufficient and also that BDP-Gd was rapidly
cleared from the body. Therefore, we next derivatized BDP-Gd
in order to obtain a more sensitive MRI contrast agent.
First, to identify the essential chemical structure for
accumulation in a lipid-rich environment, BDP-DO3A-Gd,
BDP-thio-DO3A-Gd, BDP-thioPh-DO3A-Gd, BDP-thio-BnDO-
TA-Gd and BDP-thioPh-BnDOTA-Gd were designed and syn-
thesized (Fig. 2). Adipocytes were loaded with these contrast
agents, and the localization of the contrast agents inside the
cells was observed under a fluorescence microscope. Only
Fig. 3 Chemical structures of MRI contrast agents for atherosclerotic
plaques.
some of these contrast agents were specifically localized to
lipid droplets of adipocytes, and it appeared that a common
BODIPY-thiourea-phenyl structure is important for accumu-
lation in lipid droplets. Furthermore, in order to obtain a
sufficient Gd3+ concentration for MR imaging at the targeted
region, we selected the reported multi (three) Gd3+ complex-
bearing scaffold, which can show very high r1 relaxivity,25 as a
backbone, and designed and synthesized contrast agents
containing two (2BDP3Gd), one (1BDP3Gd) or no (0BDP3Gd)
BODIPY-thiourea-phenyl moieties on this backbone as MRI
contrast agent candidates (Fig. 3). Their relaxivities and photo-
physical properties are summarized in Table 1. 2BDP3Gd and
1BDP3Gd showed three to four times higher r1 relaxivity per
Table 1 r1 Relaxivity and photophysical properties
a
b
b
r1
λabs
λem
(mM−1 s−1
)
(nm)
(nm)
Φfl
b,c
Probe
BDP-Gd
22
50
42
31
497
507
499
—
509
512
509
—
0.160
0.010
0.110
—
2BDP3Gd
1BDP3Gd
0BDP3Gd
a All data were measured at 37 °C, 20 MHz in PBS. r1 Relaxivity was
calculated per molecule. b All data were measured in 100 mM HEPES
buffer (pH 7.4). c Φfl is the relative fluorescence quantum yield determined
Gd3+ than a widely used MRI contrast agent, Magnevist® (Gd- using fluorescein in 0.1 M NaOH aq. (0.85) as a fluorescence standard.
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
DTPA; gadolinium(III) diethylenetriamine-N,N,N′,N″,N″-penta- atherosclerotic plaques of ApoE−/− mice in ex vivo fluorescence
acetic acid; the r1 relaxivity of Magnevist® is 4.3 mM−1 s−1 in imaging of the aorta (Fig. S4, ESI†). These results confirmed
water (25 °C, pH 7.4)).1 These high relaxivities may be due to that we had successfully developed more sensitive MRI con-
the high hydrophobicity of BODIPY, i.e. 2BDP3Gd and trast agents that retained the ability to accumulate in lipid-rich
1BDP3Gd appear to show intermolecular stacking in aqueous environments at both the cellular level and intact mouse level.
solution, as judged from their low fluorescence quantum
In vivo MR imaging of atherosclerotic plaques of WHHL
rabbits
yields and their absorption spectra, and the resulting incre-
ment of the apparent molecular weight may lead to enhance-
ment of the r1 relaxivity due to slower molecular rotation of Finally, we applied our improved MRI contrast agents to in vivo
the Gd3+ complex.1,26 On the other hand, in organic solvents MR imaging of atherosclerotic plaques in WHHL rabbits.
such as MeOH, 2BDP3Gd and 1BDP3Gd showed strong fluo- WHHL or wild-type rabbits were anesthetized, and T1-weighted
rescence without fluorescence quenching due to intermolecu- MR images were obtained before contrast agent injection.
lar stacking (Fig. S1 and Table S1, ESI†). When solutions of Then, the rabbits were injected with 2BDP3Gd, 1BDP3Gd or
2BDP3Gd, 1BDP3Gd, 0BDP3Gd and Magnevist® were each dia- 0BDP3Gd into an ear vein, and T1-weighted MR images were
lyzed (MWCO: 25 000) for 24 h and the concentrations of Gd3+ obtained at 1, 2 and 24 h after the injection (Fig. 4). The MR
complexes were determined by ICP-AES, we found that signal intensity in the arterial wall of WHHL rabbits, but not
2BDP3Gd and 1BDP3Gd were highly retained inside the dia- wild-type rabbits, increased time-dependently when 2BDP3Gd
lysis membranes compared to 0BDP3Gd and Magnevist® was injected, while no significant increment of MR signal in
(Fig. S2, ESI†). This result supports the idea that 2BDP3Gd and the arterial wall was seen in WHHL or wild-type rabbits
1BDP3Gd exhibit intermolecular stacking in aqueous solution injected with 0BDP3Gd or 1BDP3Gd (Fig. 4a and c). After the
because of their hydrophobic BODIPY moiety. The r1 relaxivity MR imaging, the rabbits were euthanized and the whole aorta
values are also higher than would be expected for monomeric was removed together with vertebrae for ex vivo MR imaging
complexes such as Magnevist®, and the increment of the (Fig. 4b and d). When rabbits were injected with 2BDP3Gd, the
apparent molecular weight due to intermolecular stacking may signal intensity of the isolated aorta of WHHL rabbits was
contribute to this effect.27
higher than that of wild-type rabbits, while no clear difference
We examined 2BDP3Gd and 1BDP3Gd in cellular experi- was seen when 1BDP3Gd was injected. Furthermore, in the
ments, as well as BDP-Gd. These contrast agents were efficien- fluorescence imaging of isolated aortas of rabbits, the stron-
tly accumulated into lipid droplets of adipocytes (Fig. S3, gest fluorescence was observed from the aorta of WHHL
ESI†). Furthermore, like BDP-Gd, they were accumulated in rabbits injected with 2BDP3Gd (Fig. S5, ESI†).
Fig. 4 (a) T1-Weighted MR images (in vivo imaging) of WHHL (upper) or normal (lower) rabbits before and after administration of 2BDP3Gd,
1BDP3Gd or 0BDP3Gd (7.4 μmol kg−1). White arrowheads indicate the aorta. (b) T1-Weighted MR images (ex vivo imaging) of the isolated aorta of
WHHL (upper) or normal (lower) rabbits. White arrowheads indicate the aorta. (c) Time-dependent changes of the normalized MR signal intensity of
the aorta in in vivo MR imaging of atherosclerotic plaques. The signal intensity was normalized by the MR signal intensity of a Gd3+ phantom
(0.1 mM Magnevist®). Error bars show the standard deviation (3 ROIs). (d) MR signal intensity in ex vivo MR images of the isolated aorta. The signal
intensity was normalized by the MR signal intensity of a Gd3+ phantom (0.1 mM Magnevist®). Error bars show the standard deviation (3 ROIs).
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
Histological staining of atherosclerotic plaques of WHHL
rabbits
thought to be derived from macrophages, foam cells derived
from smooth muscle cells are also seen in human athero-
sclerotic plaques,29 so we expected that 2BDP3Gd would also
be potentially useful for diagnosis of atherosclerosis in
humans. On the other hand, only weak fluorescence of
BODIPY was observed from the aorta of WHHL rabbits
injected with 1BDP3Gd (Fig. S6, ESI†), probably because a
single BODIPY-thiourea-phenyl moiety is not enough to
support sufficient accumulation of the contrast agent in athero-
sclerotic plaques for MR imaging. The reproducibility of the
MR imaging of atherosclerotic plaques with 2BDP3Gd and
histological staining was confirmed using another WHHL
rabbit (Fig. S7 and S8, ESI†). Thus, our strategy to target a
lipid-rich environment for visualizing atherosclerotic plaques
was proved to work at a practical level.
To confirm that 2BDP3Gd was accumulated in atherosclerotic
plaques, the aorta isolated from WHHL rabbits was histologi-
cally stained (Fig. 5). As shown by HE staining (Fig. 5e),
thickening of the aorta and infiltration of leucocytes were
observed, and the accumulation of lipids was also observed by
means of Oil red O staining (Fig. 5b). Furthermore, the
accumulation of fibrous elements was observed in the azan-
stained image (Fig. 5d). These results indicate the presence of
atherosclerotic plaques in the sample. The localization of the
BODIPY fluorescence (Fig. 5a; green) was well matched with
that of Oil red O staining (Fig. 5b), so the increment of the
signal intensity in the arterial wall of WHHL rabbits appeared
to be due to the accumulation of 2BDP3Gd in the lipid-rich
environment, such as lipid droplets, of atherosclerotic
plaques. Interestingly, the BODIPY fluorescence was also well
colocalized with α-actin staining (Fig. 5c), but not with the
immunostaining of macrophages (Fig. 5a; red). This result
indicates that 2BDP3Gd accumulated mainly in foam cells
derived from smooth muscle cells; indeed, many foam cells in
atherosclerotic plaques of WHHL rabbits have been reported
to be derived from smooth muscle cells.28 Although the
majority of foam cells in human atherosclerotic plaques are
In vivo fluorescence imaging of atherosclerotic plaques of LDL
receptor-deficient mice
2BDP3Gd is a lipophilic Gd3+ complex, so it is expected that
2BDP3Gd would also accumulate in the adipose tissue or
liver,20,30,31 as well as atherosclerotic plaques. However, athero-
sclerotic plaques are formed in the intima of arterial walls, so
it should be straightforward to distinguish MRI signal changes
in atherosclerotic plaques from those in adipose tissues or
liver. Indeed, we successfully visualized atherosclerotic
plaques of WHHL rabbits by MRI as shown in Fig. 4. To
support this, we carried out in vivo fluorescence imaging of
atherosclerotic plaques of LDL receptor-deficient mice.32
2BDP3Gd was intravenously injected, and fluorescence
imaging of the carotid artery of LDL receptor-deficient mice
was performed in vivo (Fig. S9, ESI†). The plaque to fatty tissue
ratio of the fluorescence signal intensity did not exceed 1,
whereas the plaque to carotid artery ratio was in the range of
1.5 to 15. It is important to note that specificity against the
normal arterial wall is the critical issue, not that against the
liver or fatty tissue. Indeed, although strong fluorescence of
2BDP3Gd was observed from adipose tissue, we successfully
identified atherosclerotic plaques in the carotid artery in vivo.
Furthermore, we could also identify atherosclerotic plaques in
the aorta by fluorescence endoscopy in real-time in vivo
(Fig. S10 and video S1, ESI†). These data indicate that accumu-
lation of 2BDP3Gd in adipose tissue does not impair visualiza-
tion of atherosclerotic plaques.
Discussion
Because MRI can provide high-spatial-resolution images of
sites deep inside tissues, it is suitable for imaging of athero-
sclerosis.33 Some MRI contrast agents for atherosclerosis have
been reported,34,35 and most of them are Gd3+ complexes or
magnetic nanoparticles conjugated to specific ligands or anti-
bodies that are directed towards atherosclerotic markers such
Fig. 5 Frozen section of the aorta of WHHL rabbits after administration
of 2BDP3Gd. Frozen sections were histologically examined by (a) fluor-
escence imaging (red: anti-RAM antibody, green: 2BDP3Gd, blue: DAPI),
(b) Oil red O staining, (c) α-actin (smooth muscle cell marker) staining,
(d) azan staining, and (e) haematoxylin and eosin (HE) staining. Black
arrowheads in (e) indicate leucocytes. Scale bar: 50 μm.
as adhesion molecules (VCAM-1,36 integrin αVβ3 and selec-
37
tin38,39), a scavenger receptor of macrophages40 and fibrin.41,42
Furthermore, to obtain higher sensitivity, liposomes,43 nano-
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
particles44 and micelles40,45 containing 103–104 Gd3+ and tar- plaques of LDL receptor-deficient mice was also performed by
geting moieties have been developed. However, the target bio- the use of an endoscope, showing that this agent enables real-
molecules for atherosclerosis are present in plaques only at the time visualization of atherosclerotic plaques in vivo. Thus, our
nanomolar concentration.33 On the other hand, our alternative environment-targeting design strategy is effective for the
strategy to target a specific environment enables efficient molecular design of practical MRI contrast agents for in vivo
accumulation of the MRI contrast agent in atherosclerotic whole-body imaging of lipid-rich environments such as athero-
plaques, irrespective of the concentrations of specific target sclerotic plaques. It should also be possible to extend this
biomolecules. Therefore, in principle, 2BDP3Gd should be design strategy to other pathological environments, such
effective for the sensitive detection of atherosclerotic plaques. as hypoxia50,51 and the relatively low pH environment of
Very recently, an elastic layer-specific MRI contrast agent tumors.52,53
based on a similar design strategy was reported,34 and this
contrast agent could also successfully visualize atherosclerotic
plaques. However, although this elastic layer-specific MRI con-
trast agent can evaluate the morphology of arterial walls, our
Methods
Supplementary methods
contrast agent can visualize atherosclerotic plaques with large
lipid cores, which tend to rupture easily.42 Thus, our contrast
Experimental details of general procedures, materials, instru-
ments, synthesis, fluorometric analysis, relaxivity measure-
agent should be more suitable to predict risk. In addition, an
MR angiography experiment showed that 2BDP3Gd has a
ments, cell culture, fluorescence microscopic imaging of
adipocytes, in vivo fluorescence imaging of LDLR−/− mice and
longer half-life in the circulation than 0BDP3Gd (Fig. S11,
ESI†). It is known that lipophilic Gd3+ complexes bind to
ultrathin endoscopic fluorescence imaging can be found in
the ESI.†
serum albumin,1,46,47 and albumin binding prevents the Gd3+
complexes from being excreted through the kidneys and
Ex vivo imaging of mouse aorta
extends their circulation time.46 Indeed, the binding of
Mice were fixed on a custom-made cradle and anesthetized
continuously with mixture of oxygen and isoflurane
2BDP3Gd to serum albumin was confirmed by in vitro fluo-
rescence measurement (Fig. S12 and S13, ESI†). 2BDP3Gd also
binds to LDL in vitro and in cellulo (Fig. S12–S14, ESI†). The
binding of 2BDP3Gd to serum albumin or LDL may contribute
to the long-circulation property of 2BDP3Gd. The fortuitously
prolonged circulation time of 2BDP3Gd may also contribute to
its efficient accumulation into atherosclerotic plaques.48 We
also examined the toxicity of 2BDP3Gd after i.v. injection into
JCl:ICR mice (n = 5) at the same dose as used in the MRI exper-
iment (7.4 μmol kg−1). No mice died within 10 days, and thus
2BDP3Gd did not show serious acute toxicity at the dose used,
although we cannot rule out the risk of nephrogenic systemic
fibrosis (NSF), a rare but serious side effect of Gd3+ contrast
agents.23,49
a
(1.5–2.0%). Then, 5 mM probe solution in saline (100 μL) was
administered to the mouse via the orbital vein. An additional
probe solution (150 μL) was administered to the same mouse
via the orbital vein at 120 min after the first injection. The
mouse was sacrificed by perfusion fixation with 4% form-
aldehyde at 60 min after the second injection. The full-length
aorta was carefully isolated and longitudinally opened. Fluo-
rescence images of the aorta were captured with a Maestro™
In-Vivo Imaging System (Perkin Elmer). The excitation wave-
length was 445–490 nm, and the fluorescence emission wave-
length was 520–800 nm. Next, atherosclerotic plaques were
visualized by Sudan IV staining, i.e. the aorta was fixed with
10% formalin solution for 30 min, and then washed with 70%
EtOH for 3 min. After that, 0.5% (w/v) Sudan IV staining solu-
tion (in acetone–70% EtOH, 30% H2O = 50/50) was added and
the preparation was shaken for 15 min. To remove excess
Sudan IV, the aorta was washed with 70% EtOH for 3 min and
with water for 5 min. Then, bright-field images of the aorta
were captured with a compact digital camera (FinePix F40fd,
FUJIFILM).
Conclusions
We have developed a new design strategy focusing on a lipid-
rich environment in order to visualize atherosclerotic plaques
by means of MRI. Among the contrast agent candidates exam-
ined, the lipophilic BODIPY-thiourea-phenyl moiety was
identified as an essential structure for recognition of lipid-rich
environments, such as lipid droplets of atherosclerotic
MR imaging of atherosclerotic plaques of rabbits
plaques, in vitro. We established that 2BDP3Gd containing this Watanabe heritable hyperlipidemic (WHHL) rabbits and
structure was accumulated preferentially in atherosclerotic normal Japanese white (JW) rabbits were purchased from
−/−
plaques in the ApoE
mouse model of atherosclerosis. Kitayama Labes (Nagano, Japan). Animals under general
2BDP3Gd was also able to visualize atherosclerotic plaques in anesthesia with continuous inhalation of oxygen containing
the aorta of WHHL rabbits in T1-weighted MR images in vivo. isoflurane (2.0–2.5%) (Escaine; Mylan Pharma Co., Tokyo,
Moreover, the BODIPY moiety enabled precise observation of Japan) were subjected to MR imaging in the supine position
the localization of 2BDP3Gd inside the artery ex vivo by means with a superconducting MR scanner (Signa HDxt 3.0 T, GE
of fluorescence microscopy, providing the capability for multi- Healthcare, WI, USA), using an 8-channel phased array knee
modal imaging. In vivo fluorescence imaging of atherosclerotic coil. T1-Weighted images were acquired before and at 1 h, 2 h
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
and 24 h after intravenous injection of the probe solution in at Kyushu University for helpful discussions. S.I. was sup-
saline (5 mM, 7.4 μmol kg−1) via the marginal pinna vein. The ported by a Grant-in-Aid for JSPS Fellows.
parameters used for the ECG gated double IR prepared fast
spin echo sequence were median TR (ms)/ef TE (ms)/FA
(degree) of 261/8.6/90, a matrix size of 320 × 256; a FOV (cm) of
12 × 9 for axial and 18 × 18 for sagittal, a slice thickness (mm)
Notes and references
of 3.0, an echo train of 2, section intervals (mm) of 5 for axial
and 0 for sagittal, and the average number of signals (NEX)
of 4.
For MR angiography, the coronal IR prepared 3-dimen-
sional Fast Spoiled Gradient Echo sequence was used with the
following parameters; TR (ms)/TE (ms)/FA (degree) of 4.3–4.7/
1.4–1.6/15, a matrix size of 320 × 160; a FOV (cm) of 17.8 ×
12.5, a slice thickness (mm) of 6 and the NEX of 1.
1 P. Caravan, J. J. Ellison, T. J. McMurry and R. B. Lauffer,
Chem. Rev., 1999, 99, 2293–2352.
2 R. Weissleder and M. J. Pittet, Nature, 2008, 452, 580–589.
3 E. Terreno, D. D. Castelli, A. Viale and S. Aime, Chem. Rev.,
2010, 110, 3019–3042.
4 A. Y. Louie, M. M. Hüber, E. T. Ahrens, U. Rothbächer,
R. Moats, R. E. Jacobs, S. E. Fraser and T. J. Meade, Nat.
Biotechnol., 2000, 18, 321–325.
After the imaging, animals were euthanized and the whole
aorta was removed with vertebrae for ex vivo MR imaging. In
this case, the pulse sequence and the parameters used were
axial conventional spin echo, TR (ms)/TE (ms) of 300/11, a
matrix size of 384 × 160; a FOV (cm) of 9 × 6.8, a slice thickness
(mm) of 1, section intervals (mm) of 2.5, and the NEX of 12.
The average intensities of the vessel wall and the phantom
(0.1 mM) were measured by placing three regions of interest
(ROIs) on the vessel wall and the largest possible ROI on the
phantom. Then, the three values for the vessel wall were aver-
aged, and divided by the value for the phantom. Error bars
5 J. L. Major and T. J. Meade, Acc. Chem. Res., 2009, 42, 893–
903.
6 K. Hanaoka, Chem. Pharm. Bull., 2010, 58, 1283–1294.
7 L. Frullano, C. Wang, R. H. Miller and Y. Wang, J. Am.
Chem. Soc., 2011, 133, 1611–1613.
8 L. M. De León-Rodríguez, A. Lubag, D. G. Udugamasooriya,
B. Proneth, R. A. Brekken, X. Sun, T. Kodadek and A. Dean
Sherry, J. Am. Chem. Soc., 2010, 132, 12829–12831.
9 K. Hanaoka, A. J. M. Lubag, A. Castillo-Muzquiz,
T. Kodadek and A. D. Sherry, Magn. Reson. Imaging, 2008,
26, 608–617.
show the standard deviation of the signal intensities of the 10 G. J. Kelloff, K. A. Krohn, S. M. Larson, R. Weissleder,
three ROIs.
D. A. Mankoff, J. M. Hoffman, J. M. Link, K. Z. Guyton,
W. C. Eckelman, H. I. Scher, J. O’Shaughnessy,
B. D. Cheson, C. C. Sigman, J. L. Tatum, G. Q. Mills,
D. C. Sullivan and J. Woodcock, Clin. Cancer Res., 2005, 11,
7967–7985.
Fluorescence imaging of the isolated WHHL rabbit aorta
Fluorescence images of the aorta (ESI Fig. S4†) were captured
with a Maestro™ 2 (Perkin Elmer). The excitation wavelength
was 457 22.5 nm, and fluorescence emission (>490 nm) was
observed.
11 W. S. El-Deiry, C. C. Sigman and G. J. Kelloff, J. Clin. Oncol.,
2006, 24, 3261–3273.
12 A. J. Lusis, Nature, 2000, 407, 233–241.
13 R. Ross, N. Engl. J. Med., 1999, 340, 115–126.
14 C. K. Glass and J. L. Witztum, Cell, 2001, 104, 503–
516.
Histological staining of the isolated WHHL rabbit aorta
Aortic sections were subjected to HE, Azan-Mallory or Oil red
O staining. Immunohistochemistry was performed according
to reported methods,54 using anti-rabbit macrophage mono-
clonal antibody (RAM-11, Dako Corp., Santa Barbara, CA,
USA)55 or anti-alpha-actin monoclonal antibody (M0851, Dako
Corp., Santa Barbara, CA, USA). The second antibody for
macrophages was Fluoromount (goat anti mouse Alexa 594,
Diagnostic BioSystems, USA).
15 V. L. Roger, A. S. Go, D. M. Lloyd-Jones, E. J. Benjamin,
J. D. Berry, W. B. Borden, D. M. Bravata, S. Dai, E. S. Ford,
C. S. Fox, H. J. Fullerton, C. Gillespie, S. M. Hailpern,
J. A. Heit, V. J. Howard, B. M. Kissela, S. J. Kittner,
D. T. Lackland, J. H. Lichtman, L. D. Lisabeth,
D. M. Makuc, G. M. Marcus, A. Marelli, D. B. Matchar,
C. S. Moy, D. Mozaffarian, M. E. Mussolino, G. Nichol,
N. P. Paynter, E. Z. Soliman, P. D. Sorlie, N. Sotoodehnia,
T. N. Turan, S. S. Virani, N. D. Wong, D. Woo and
M. B. Turner, Circulation, 2012, 125, e2–e220.
Acknowledgements
16 T. C. Walther and R. V. Farese Jr., Annu. Rev. Biochem.,
2012, 81, 687–714.
This work was supported in part by MEXT (Specially Promoted
Research Grant no. 22000006 to T.N., and Grants 24689003 17 R. L. Reddick, S. H. Zhang and N. Maeda, Arterioscler.,
and 24659042 to K.H.). K.H. was also supported by Grants-in- Thromb., Vasc. Biol., 1994, 14, 141–147.
Aid from the Takeda Science Foundation, the Uehara Memor- 18 S. Steffens, N. R. Veillard, C. Arnaud, G. Pelli, F. Burger,
ial Foundation, the Hattori Hokokai Foundation, the Inoue
Foundation for Science, The Research Foundation for Pharma-
C. Staub, M. Karsak, A. Zimmer, J.-L. Frossard and
F. Mach, Nature, 2005, 434, 782–786.
ceutical Sciences and the Astellas Foundation for Research on 19 L. L. Listenberger and D. A. Brown, Curr. Protoc. Cell Biol.,
Metabolic Disorders. The authors thank Dr Michiaki Kumagai
2007, 24, 24.2.1–24.2.11.
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
20 T. Yamane, K. Hanaoka, Y. Muramatsu, K. Tamura, 39 H. Alsaid, G. De Souza, M.-C. Bourdillon, F. Chaubet,
Y. Adachi, Y. Miyashita, Y. Hirata and T. Nagano, Bioconju-
gate Chem., 2011, 22, 2227–2236.
21 A. Loudet and K. Burgess, Chem. Rev., 2007, 107, 4891–4932.
22 L. E. Jennings and N. J. Long, Chem. Commun., 2009, 3511–
3524.
23 A. Louie, Chem. Rev., 2010, 110, 3146–3195.
24 M. Shiomi and T. Ito, Atherosclerosis, 2009, 207, 1–7.
25 Z. Zhang, M. T. Greenfield, M. Spiller, T. J. McMurry,
A. Sulaiman, C. Desbleds-Mansard, L. Chaabane, C. Zahir,
E. Lancelot, O. Rousseaux, C. Corot, P. Douek, A. Briguet,
D. Letourneur and E. Canet-Soulas, Investig. Radiol., 2009,
44, 151–158.
40 M. J. Lipinski, V. Amirbekian, J. C. Frias, J. G. S. Aguinaldo,
V. Mani, K. C. Briley-Saebo, V. Fuster, J. T. Fallon,
E. A. Fisher and Z. A. Fayad, Magn. Reson. Med., 2006, 56,
601–610.
R. B. Lauffer and P. Caravan, Angew. Chem., Int. Ed., 2005, 41 M. Sirol, V. Fuster, J. J. Badimon, J. T. Fallon, P. R. Moreno,
44, 6766–6769.
J.-F. Toussaint and Z. A. Fayad, Circulation, 2005, 112,
1594–1600.
26 K. Hanaoka, K. Kikuchi, T. Terai, T. Komatsu and
T. Nagano, Chem. – Eur. J., 2008, 14, 987–995.
27 G. Liang, J. Ronald, Y. Chen, D. Ye, P. Pandit, M. L. Ma,
B. Rutt and J. Rao, Angew. Chem., Int. Ed., 2011, 50, 6283–
6286.
42 K. Overoye-Chan, S. Koerner, R. J. Looby, A. F. Kolodziej,
S. G. Zech, Q. Deng, J. M. Chasse, T. J. McMurry and
P. Caravan, J. Am. Chem. Soc., 2008, 130, 6025–6039.
43 A. Maiseyeu, G. Mihai, T. Kampfrath, O. P. Simonetti,
C. K. Sen, S. Roy, S. Rajagopalan and S. Parthasarathy,
J. Lipid Res., 2009, 50, 2157–2163.
28 G. S. Getz and C. A. Reardon, Arterioscler., Thromb., Vasc.
Biol., 2012, 32, 1104–1115.
29 A. C. Doran, N. Meller and C. A. McNamara, Arterioscler., 44 P. M. Winter, A. M. Morawski, S. D. Caruthers,
Thromb., Vasc. Biol., 2008, 28, 812–819.
R. W. Fuhrhop, H. Zhang, T. A. Williams, J. S. Allen,
E. K. Lacy, J. D. Robertson, G. M. Lanza and S. A. Wickline,
Circulation, 2003, 108, 2270–2274.
30 H.-J. Weinmann, G. Schuhmann-Giampieri, H. Schmitt-
Willich, H. Vogler, T. Frenzel and H. Gries, Magn. Reson.
Med., 1991, 22, 233–237.
31 H.-J. Weinmann, W. Ebert, B. Misselwitz and H. Schmitt-
Willich, Eur. J. Radiol., 2003, 46, 33–44.
32 S. Ishibashi, J. L. Goldstein, M. S. Brown, J. Herz and
D. K. Bums, J. Clin. Invest., 1994, 93, 1885–1893.
33 J. Sanz and Z. A. Fayad, Nature, 2008, 451, 953–957.
34 M. R. Makowski, A. J. Wiethoff, U. Blume, F. Cuello,
45 V. Amirbekian, M. J. Lipinski, K. C. Briley-Saebo,
S. Amirbekian, J. G. S. Aguinaldo, D. B. Weinreb, E. Vucic,
J. C. Frias, F. Hyafil, V. Mani, E. A. Fisher and Z. A. Fayad,
Proc. Natl. Acad. Sci. U. S. A., 2007, 104, 961–966.
46 P. Caravan, Acc. Chem. Res., 2009, 42, 851–862.
47 S. Aime, M. Botta, M. Fasano and E. Terreno, Chem. Soc.
Rev., 1998, 27, 19–29.
A. Warley, C. H. P. Jansen, E. Nagel, R. Razavi, 48 A. K. Gupta and M. Gupta, Biomaterials, 2005, 26, 3995–
D. C. Onthank, R. R. Cesati, M. S. Marber, T. Schaeffter, 4021.
A. Smith, S. P. Robinson and R. M. Botnar, Nat. Med., 2011, 49 L. Daftari Besheli, S. Aran, K. Shaqdan, J. Kay and
17, 383–388. H. Abujudeh, Clin. Radiol., 2014, 69, 661–668.
35 S. Yasuda, K. Ikuta, T. Uwatoku, K. Oi, K. Abe, F. Hyodo, 50 A. L. Harris, Nat. Rev. Cancer, 2002, 2, 38–47.
K. Yoshimitsu, K. Sugimura, H. Utsumi, Y. Katayama and 51 J. M. Brown and W. R. Wilson, Nat. Rev. Cancer, 2004, 4,
H. Shimokawa, J. Vasc. Res., 2008, 45, 123–128.
437–447.
36 M. Nahrendorf, F. A. Jaffer, K. A. Kelly, D. E. Sosnovik, 52 P. Montcourrier, P. H. Mangeat, C. Valembois, G. Salazar,
E. Aikawa, P. Libby and R. Weissleder, Circulation, 2006,
114, 1504–1511.
A. Sahuquet, C. Duperray and H. Rochefort, J. Cell Sci.,
1994, 107, 2381–2391.
37 C. Burtea, S. Laurent, O. Murariu, D. Rattat, G. Toubeau, 53 H. Izumi, T. Torigoe, H. Ishiguchi, H. Uramoto, Y. Yoshida,
A. Verbruggen, D. Vansthertem, L. Vander Elst and
M. Tanabe, T. Ise, T. Murakami, T. Yoshida, M. Nomoto
R. N. Muller, Cardiovasc. Res., 2008, 78, 148–157.
and K. Kohno, Cancer Treat. Rev., 2003, 29, 541–549.
38 F. Chaubet, I. Bertholon, J.-M. Serfaty, R. Bazeli, H. Alsaid, 54 T. Tsukada, M. Rosenfeld, R. Ross and A. M. Gown,
M. Jandrot-Perrus, C. Zahir, P. Even, L. Bachelet, Z. Touat, Arterioscrelosis, 1986, 6, 601–613.
E. Lancelot, C. Corot, E. Canet-Soulas and D. Letourneur, 55 M. Shiomi, S. Yamada and T. Ito, Atherosclerosis, 2005, 178,
Contrast Media Mol. Imaging, 2007, 2, 178–188.
287–294.
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2014