Self-Assembled Nanomicelles as MRI Blood-Pool Contrast Agent

Article abstract of DOI:10.1002/chem.201703962

Gadolinium-loaded nanomicelles show promise as future magnetic resonance imaging (MRI) contrast agents (CAs). Their increased size and high gadolinium (Gd) loading gives them an edge in proton relaxivity over smaller molecular Gd-complexes. Their size and stealth properties are fundamental for their long blood residence time, opening the possibility for use as blood-pool contrast agents. Using l-tyrosine as a three-functional scaffold we synthesized a nanostructure building block 8. The double C18 aliphatic chain on one side, Gd-1,4,7,10-tetraazacyclododecane-1-4-7-triacetic acid (Gd-DO3A) with access to bulk water in the center and 2 kDa PEG on the hydrophilic side gave the amphiphilic properties required for the core–shell nanomicellar architecture. The self-assembly into Gd-loaded monodispersed 10–20 nm nanomicelles occurred spontaneously in water. These nanomicelles (Tyr-MRI) display very high relaxivity at 29 mm^?1 s^?1 at low field strength and low cytotoxicity. Good contrast enhancement of the blood vessels and the heart together with prolonged circulation time in vivo, makes Tyr-MRI an excellent candidate for a new supramolecular blood-pool MRI CA.

Full text of DOI:10.1002/chem.201703962

A Journal of
Accepted Article
Title: Self-assembled nanomicelles as MRI blood-pool contrast agent
Authors: Andrej Babic, Vassily Vorobiev, Celine Xayaphoummine,
Gaelle Lapicorey, Anne-Sophie Chauvin, Lothar Helm, and
Eric Allémann
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.201703962
Link to VoR: http://dx.doi.org/10.1002/chem.201703962
Supported by

10.1002/chem.201703962
Chemistry - A European Journal
FULL PAPER
Self-assembled nanomicelles as MRI blood-pool contrast agent
Andrej Babič*[a], Vassily Vorobiev[a], Céline Xayaphoummine[a], Gaëlle Lapicorey[b], Anne-Sophie
Chauvin[b], Lothar Helm[b] and Eric Allémann[a]
Abstract: Gadolinium-loaded nanomicelles show promise as future
magnetic resonance imaging (MRI) contrast agents (CAs). Their
increased size and high gadolinium (Gd) loading gives them an edge
in proton relaxivity over smaller molecular Gd-complexes. Their size
and stealth properties are fundamental for their long blood residence
time opening the possibility for use as blood-pool contrast agents.
Using L-tyrosine as a three-functional scaffold we synthesized a
nanostructure building block 8. The double C18 aliphatic chain on one
side, Gd-DO3A with access to bulk water in the center and 2 kDa PEG
on the hydrophilic side gave the amphiphilic properties required for
the core-shell nanomicellar architecture. The self-assembly into Gd-
loaded monodispersed 10-20 nm nanomicelles occurred
spontaneously in water. Tyr-MRI nanomicelles display very high
relaxivity at 29 mM-1s- 1 at low field strength and low cytotoxicity. Good
contrast enhancement of the blood vessels and the heart together
with prolonged circulation time in vivo, makes Tyr-MRI an excellent
candidate for a new supramolecular blood-pool MRI CA.
The overwhelming majority of CAs used today are chelates of
trivalent gadolinium (Gd) due to the inherently high relaxivity of
this atom.[13] Chelate forms are used for safety reasons since free
Gd is toxic and can lead to nephrogenic systemic fibrosis.[14] The
first chelate approved for human MRI in 1988 was gadopentetate
dimeglumine marketed as Magnevist®. Several other Gd-chelate
CAs have followed and most of them are non-specific.[15] Their
low molecular weight allows them to extravasate and distribute
rapidly through the body making them appropriate for whole body
scans.
On the other hand, blood pool contrast agents (BPCA) are
designed to persist in the blood compartment during MRI scan.[16]
They are primarily used in magnetic resonance angiography
(MRA). Several marketed CA are approved for MRA, however
only gadofosveset (Ablavar®) is a true BPCA. Unfortunately, its
marketing authorization has recently been removed.
Compared to low-molecular weight gadofosveset which displays
prolonged blood residence due to binding to plasma proteins,[17]
macromolecules and supramolecular structures have much
higher molecular weights which prolongs their blood circulation
time. These include proteins,[18] water-soluble fullerenes,[19]
carbohydrates,[20] cyclodextrins,[21] nanoparticles,[22] polymers,[23]
dendrimers,[24] liposomes,[25] and most recently micelles.[26] The
major advantage of macromolecular Gd-chelates is increased
relaxivity as a consequence of slower rotational motion of the
macromolecules.[27]
Introduction
Since its introduction in the 1970s, magnetic resonance imaging
(MRI) is one of the fastest growing techniques in terms of clinical
diagnostics and biomolecular research.[1] Its success comes from
the ability to monitor organs and tissues in real time non-
invasively, and with an excellent spatial resolution.[2] Around
36,000 MRI scanners worldwide perform about 50 million
examinations each year.[3] Improved MRI images are largely
related to the use of paramagnetic contrast agents (CA). Today,
around 50 % of all clinical scans are run with CAs injected before
the MRI acquisition to improve the contrast and resolution of
images.[3-4] The development of improved CAs has lagged behind
the rapid progress of the MRI scanners.[5] Improved CAs could
result in faster and improved diagnosis and clinical outcomes of
diseases that depend on imaging, namely different types of
cancer[6] and related angiogenesis,[7] atherosclerosis,[8] arthritis,[9]
liver diseases,[10] osteoporosis,[11] and brain.[12]
Tyr-MRI
[a]
[b]
Dr. A. Babič, V. Vorobiev, C. Xayaphoummine, Prof. E. Allémann
Pharmaceutical Technology
School of Pharmacy Geneva-Lausanne
University of Geneva
Rue Michel Servet 1, 1211 Geneva, Switzerland
E-mail: andrej.babic@unige.ch
G. Lapicorey, Dr. A.S. Chauvin, Prof. L. Helm
Institut of Chemical Sciences and Engineering
Swiss Federal Institute of Technology of Lausanne
Route Cantonale, 1015 Lausanne, Switzerland
Figure 1. Design elements of Gd-building block for self-assembling Tyr-MRI
nanomicelles. Double C18 lipophilic tail (red), hydrophilic stealth PEG moiety
(blue) and Gd-DO3A as MRI contrast element (green).
Supporting information for this article can be found under:
1
This article is protected by copyright. All rights reserved.

10.1002/chem.201703962
Chemistry - A European Journal
FULL PAPER
Organic micelles could be endowed with excellent pharmaceutical
characteristics on top of the desired relaxivity properties making
them good candidates for new BPCA.[28] The potential benefits of
the micellar CA approach are several-fold: increased contrast
caused by higher local concentration of Gd, inherent higher
relaxivity, long intravascular residence time, and potential stealth
properties yielding increased specificity of blood-pool contrast.
In this work we present the design and synthesis of a new
L-tyrosine based amphiphilic building block 8. In aqueous media
the self-assembly into Tyr-MRI nanomicelles occurs
spontaneously. The nanomicelles have been fully characterized
for structure, size distribution, critical micellar concentration,
stability, magnetic relaxivity, cell toxicity and as MRI contrast
enhancement in vivo.
Results and Discussion
Synthesis of the building block
The new amphiphilic building block was designed to incorporate
three distinct structural elements as depicted in Figure 1. Double
C18 aliphatic chains provide lipophilicity and angled orientation
which favour supramolecular micellar self-assembly. The choice
of the double lipophilic side chain was based on reports that
contrary to single aliphatic chains, double C18 linear chains
provide good blood permanence and no haemolytic effect in vitro
and in vivo.[30] Polyethylene glycol (PEG) is the gold standard and
serves as the hydrophilic polymer which confers prolonged
residence of colloid systems in the blood.[29] Attachment of PEG
also gives stealth properties to the CA system, which are of great
importance for a BPCA.[30] As the third part of the building block
cyclic DO3A was chosen as the Gd chelating moiety with stability
and safety in mind. It is now well established that macrocyclic
systems provide much better kinetic and thermodynamic stability
as compared to linear open-chain chelates.[31] Together the three
structural elements give the desired self-assembly properties to
the building block.
The building block 8 was synthesized in 8 synthetic steps outlined
in Scheme 1. A chiral amino acid L-tyrosine bears three functional
groups and serves as a scaffold for all structural elements
required for core-shell nanomicellar architecture.
Firstly,
L- tyrosine was protected at the amino group to give Z-L-tyrosine
1. Standard HATU activation of the L-tyrosine carboxy group
allowed peptide bond formation with dioctadecyl amine in good
66 % yield.[32] Cleavage of the Z-protecting group was achieved
by catalytic hydrogenation with Pd(OAc)2 at atmospheric pressure
to yield compound 3 after purification. The spacer was attached
by reacting 4-chloromethylbenzoyl chloride with 3 at -15 °C to
minimize the O-acyl side product formation at the phenolic side
chain. The yield was very good and the side product detected by
MS could be efficiently removed by Flash chromatography
purification. On the contrary, the following coupling of 4 with the
macrocycle DO3A(tBu)3 in acetonitrile gave very low yields. The
phenolic hydroxyl group was sufficiently nucleophilic to compete
with DO3A amine group for the reaction with the 4-chloromethyl
group yielding dimers and even trimers as side products
dramatically reducing the yield. The synthesis of 5 was optimized
by using several solvents, temperatures and reagent ratios.
Scheme 1. Synthesis of the L-tyrosine-based building block 8.
2
This article is protected by copyright. All rights reserved.

10.1002/chem.201703962
Chemistry - A European Journal
FULL PAPER
A
We found that the best result was obtained when DO3A(tBu)3 and
equimolar amount of 4 were reacted in the presence of K2CO3 in
DMF at room temperature giving satisfactory 60 % yield.
The following O-tyrosine PEGylation also proved very demanding
despite reported examples.[33] In our hands, attempts to O-
PEGylate 5 with tosyl-PEG2000 failed. Replacing tosyl-PEG2k with
Br-PEG2k, allowed the Williamson O-alkylation to proceed in low
yields. We succeeded in attaching the PEG group to 5 under
microwave irradiation at 150 °C in acetonitrile in the presence of
K2CO3 as base yielding
6 in modest 36 % yield. The
macromolecular size of the 2 kDa PEG chain clearly plays a major
role in the poor outcome of this reaction. It is noteworthy that
active targeting can be added to the building block at this step.
This could be achieved by the addition of specific targeting
moieties at the surface of the micelles by using heterobifunctional
PEG and orthogonal protective group strategy for the addition of
targeting ligands.
B
The following total deprotection of tert-butyl esters was carried out
under standard reaction conditions using TFA/CH2Cl2 (1:1)
mixture. Without purification, the deprotected compound 7 was
complexed with gadolinium at pH 5.8 in water, using GdCl3
solution. Free Gd3+ ions were removed by size exclusion
chromatography. The gadolinium loading of the final product 8
was 0.67 and 0.70 determined by bulk magnetic susceptibility
(BMS) and ICP-MS, respectively.
C
C
Tyr-MRI nanomicelles
Compound 8 self-assembled into Tyr-MRI nanomicelles in
aqueous media using either nanoprecipitation[34] or thin-film
hydration method.[35] The self-assembly was surprisingly efficient.
The two different methods gave very similar size distributions
indicating robust self-assembly properties independent of the
method of preparation. Nanoprecipitation being very
straightforward, robust and efficient was chosen as the method of
choice for scale-up experiments.
D
The mean size distribution was around 11 nm with a slightly
negative zeta potential at -13 mV determined by DLS and
electrophoretic mobility, respectively (Figure 2). TEM imaging
confirmed the size of the nanomicelles and also revealed their
spherical shape, further confirming the validity of DLS results. The
stability of the Tyr-MRI nanomicelles was assayed in water for 1
month. No signs of sedimentation or phase separation were
observed demonstrating excellent stability of the micelles. DLS
analysis revealed that size distribution remained unchanged
indicating no chemical or physical degradation. Similar results
were obtained when the nanomicelles were tested at 4, 22, and
37 °C in water, PBS and different concentrations of NaCl (Figures
S2, S3 and S5 respectively). The DLS revealed robust size
stability for 1 week. However, in DMEM buffer Tyr-MRI did show
visual signs of aggregation after several days at 22 and 37 °C.
This was confirmed by DLS analysis indicating the presence of
larger aggregates (Figure S4). The determination of the critical
Figure 2. Characterization of Tyr-MRI nanomicelles. TEM image (A), DLS
volume distribution (B), zeta potential (C), stability in water at 22 °C (D).
NMR relaxometry
The efficiency of MRI contrast agents is commonly addressed by
relaxation enhancement per mM Gd3+, the relaxivity r1. The
relaxivity depends on several parameters describing structure
and dynamics of the compound in solution. The most reliable way
to get information on these parameters is a combined analysis of
1H relaxivity measured as a function of magnetic field (NMRD
profile), yielding mainly information on rotational diffusion of the
compound, and of 17O NMR transverse relaxation 1/T2r and
chemical shift r, depending mainly on the water exchange rate
constant kex.[37]
1
micelle concentration (cmc) using the H NMR relaxivity method
gave a cmc well below 10 µM (Figure S1), ideal for in vivo
applications.[36]
1
Hence, H nuclear magnetic relaxation dispersion (NMRD) and
17O transverse relaxation and chemical shift of Tyr-MRI
nanomicelles were measured.
A simultaneous fit of the
experimental data shown in Figure 3 using common Solomon-
Bloembergen-Morgan theory including the Lipari-Szabo approach
for internal motion[38] allowed to determine correlation times for
3
This article is protected by copyright. All rights reserved.

10.1002/chem.201703962
Chemistry - A European Journal
FULL PAPER
Table 1. 1H relaxivities of selected Gd-micelles
310 K
298 K
relaxivity r1 in mM-1 s-1
@20
MHz
@60
MHz
@20
MHz
@60
MHz
reference
Tyr-MRI
24
3.5
20
18
18
14
19
3.1
-
26
-
19
-
[40]
[23a]
[36]
[36]
[41]
Gd-DOTA
[Gd(DTPA-BA)-(CH2)12
[Gd(DOTA)-C14
[Gd(DOTA)-C18
[Gd(eptpa)-C16
-
-
18
16
15
21
21
21
18
17
21
0
. 0
- 0 . 5
- 1 . 0
- 1 . 5
Figure 4. Cytotoxicity of Tyr-MRI nanomicelles (blue) and Gd-DOTA (red) in
HT1080 fibrosarcoma cells after 24 h.
probably due to the strong hydrophobic interactions coming from
two aromatic rings and the double C18 alkyl chains. The value of
S2 = 0.40 found is close to those measured for a PAMAM G5
dendrimer of similar size (diameter: 8 nm at pH 6.7; S2: 0.36 to
0.43, depending on pH).[42]
3
. 0
3
. 2
3
. 4
3 . 6
Comparison of 1H T1 relaxivities with in vitro results of other
micellar systems with Gd-chelates with one inner-sphere water
molecule shows that the Tyr-MRI nanomicelles has r1 values at
the upper limit of the observed range (Table 1).[3, 43]
- 1
1
0
0
0
/ T [ K
]
Figure 3. 1H NMRD profiles measured by the inversion recovery method[39] at
298 K (diamonds, red) and 310 K (circles, blue) at different frequencies (A).
17O transverse relaxation enhancement ln(1/T2r) as function of inverse
temperature measured at 9.4
differences r as a function of inverse temperature at 9.4 T (circles) (C).
T (squares) (B). Reduced chemical shift
Cell toxicity
Cell toxicity (Figure 4) was assessed by MTT assay on HT1080
fibrosarcoma cells. The nanomicelles displayed no cytotoxicity
comparable to Gd-DOTA up to the concentration of 125 µM. This
is in line with their slightly negative (-13 mV) zeta potential and
PEGylated surface incurring no major toxic effects upon contact
with the cells for 24 h, thus indicating good biocompatibility.
global rotational diffusion of the micelles, g, and for local internal
motion, l, as well as the rate constant for the exchange of the
inner-sphere water molecule directly bound to Gd3+, kex (Table S2).
The water exchange rate constant calculated at 37 °C,
kex310=30106 s-1, is fast enough not to limit relaxivity at magnetic
fields of interest for MRI applications. It is actually three times
faster than water exchange on the DOTA based derivatives
forming micelles studied by Nicolle et al.[36] (Table S1).
Comparing the rotational behavior of the Tyr-MRI micelles with
other systems studied is complicated by the fact that only very few
studies took internal rotation into consideration in analyzing
NMRD profiles. Merbach and co-workers published results on
several micelle-forming systems (Table S1) showing that the
global rotational correlation time is in the order of 2 to 4.4 ns. The
local one is much shorter, mostly in the order of 200 to 400 ps.
The model free order parameter S2 for the Tyr-MRI micelle is
relatively high showing that these micelles are relatively rigid
MRI imaging
Phantom MRI imaging was first performed using both Tyr-MRI
and Gd-DOTA. The results (Figure S6) show that there are no
significant differences in relaxivity between Tyr-MRI and Gd-
DOTA at 3T. This correlates with the NMR spectroscopy relaxivity
results. At high field strengths, the relaxivity gains of
supramolecular CA become negligible compared to small Gd-
complexes such as Gd-DOTA.
4
This article is protected by copyright. All rights reserved.

10.1002/chem.201703962
Chemistry - A European Journal
FULL PAPER
However, as proof of concept the phantom MRI images gave us
an insight into the sensitivity of the new CA at 3T. The
experiments demonstrated that Tyr-MRI contrast could be
obtained at 0.05 mM gadolinium concentration and above. After
promising results of MRI contrast enhancement in vitro we
proceeded with in vivo experiments.
First, clinically used Gd-DOTA was injected intravenously in male
C57BL/6J mice. Gd-DOTA is eliminated rapidly in humans during
the alpha-phase with half-life in minutes as a consequence of its
small molecular weight resulting in glomerular filtration in the
kidneys. This is followed by slower elimination kinetics in the beta-
phase with terminal half-life of 1.5 h.[44] In our hands Gd-DOTA
demonstrated similar behaviour in mice. The CA gave good
contrast enhancement immediately after intravenous injection
however it was rapidly eliminated and distributed into the
extracellular compartment (Figure 5, B-D). At 30 min no contrast
enhancement could be observed in the target vascular
compartment. The contrast shifts to the kidneys at this time point
confirming the predominantly renal elimination route. (Figure 5C
and 5D).
On the other hand, Tyr-MRI nanomicelles enhanced the
vasculature contrast immediately after IV administration but more
importantly also at longer time points (Figure 5, bottom).
Furthermore, 1 h after injection the heart coronal sections shows
a high signal in the heart and the surrounding blood vessels as a
result of high blood perfusion of this organ (Figure 6). The size
and stealth properties of Tyr-MRI nanomicelles reduces their
uptake by reticuloendothelial system of the liver and spleen
compared to polymeric nanoparticles.
In the case of Gd-DOTA, the same MRI sections fail to provide
heart contrast and show only strong contrast from the bladder
indicating CA eliminated via the kidneys. Tyr-MRI displays much
lower bladder contrast at the same time point corroborating the
desired prolonged blood circulation time but at the same time
indicating a partial renal elimination. As a consequence Tyr-MRI
could allow cardio-vascular imaging and perfusion monitoring by
MRI. The presence of a good blood-pool MRI CA is a prerequisite
for these imaging modalities since long MRI sequences are
required to obtain higher resolution images.
Figure 5. Tyr-MRI (bottom, E-H) displays prolonged blood-pool contrast
enhancement. T1-weighted 3T MRI scans were performed before IV injection
(A, E), at t=0 (B, F), at 30 min (C, G) and at 1h (D, H) after injection. Gd-DOTA
was injected as control (top, A-D).
Conclusions
We have presented the 8-step synthesis of L-tyrosine-based
building block 8. The amphiphilic product self-assembled in water
to give Gd-based Tyr-MRI nanomicelles as supramolecular MRI
contrast agent. The nanomicelles showed low dispersity,
excellent MRI relaxivity, good stability and low cytotoxicity in vitro.
High MRI contrast was achieved by virtue of their size and Gd-
content compared to small molecular weight Gd-based CA. In vivo
experiments gave very encouraging results of blood-pool contrast
enhancement. Further preclinical studies are ongoing to
understand the exact pharmacokinetics of the nanomicelles and
to evaluate the performance of this CA for high-resolution 4D MRI.
Figure 6. Coronal MRI scans of mice injected with Gd-DOTA (A) and Tyr-MRI
(B) at 3 T. 1h after injection Tyr-MRI shows high contrast enhancement of the
heart.
5
This article is protected by copyright. All rights reserved.

10.1002/chem.201703962
Chemistry - A European Journal
FULL PAPER
Chemical shifts (δ) are quoted in parts per million (ppm) and coupling
constants (J) are in hertz (Hz), s stands for singlet, d for doublet, dd for
doublet of doublets, t for triplet, q for quartet, m for multiplet. Residual
solvent peaks were used as the internal reference for the proton and
carbon chemical shifts. NMR spectra were processed with Mnova version
8.1.2 software package. Low resolution mass spectrometry (LRMS) was
carried out on a HTS PAL-LC10A – API 150Ex instrument in ESI (positive
mode). High resolution mass spectrometry (HRMS) was carried out on a
QSTAR Pulsar (AB/MDS Sciex) instrument in ESI (positive mode). Matrix-
assisted laser desorption/ionisation-Time of flight (MALDI-TOF) mass
spectra were obtained using an Axima CFR+ (Shimadzu) with a cyano-4-
hydroxycinnamic acid (CHCA) matrix in linear or reflectron mode. MALDI
spectra were processed using Shimatzu Biotech LaunchPad version 2.8
software. Chemical structures were drawn and named according IUPAC
nomenclature using ChemBioDraw Ultra version 14.0.0.117 software
package. The pH was measured on a Metrohm 691 pH-meter using a
spearhead electrode (Zofingue, Switzerland), calibrated with Metrohm
buffers.
Experimental Section
Chemicals
All reagents were obtained from commercial suppliers and used without
further purification. (Benzyloxy)carbonyl-L-tyrosine ((L)-Z-Tyr-OH) was
purchased from Bachem (Bubendorf, Switzerland). The N,N’-
dioctadecylamine (HN(C18H37)2) was supplied by Sigma-Aldrich (Buchs,
Switzerland).O-(7-azabenzotriazole-1-yl)-N,N,N',N'-tetramethyluronium
hexafluoro phosphate (HATU) was provided by GenScript Corporation
(Piscataway, USA). Tri-tert-butyl 2,2',2''-(1,4,7,10-tetraazacyclododecane-
1,4,7-triyl)triacetate (DO3A(tBu)3) was purchased from CheMatech (Dijon,
France). -methoxy-ω-bromo poly(ethylene glycol) (mPEG-Br, MW 2kDa)
was obtained from Iris Biotech Gmbh (Marktredwitz, Germany). The 4-
(chloromethyl)benzoyl chloride, palladium acetate (Pd(OAc)2), potassium
iodide (KI), potassium carbonate (K2CO3), cesium carbonate (Cs2CO3),
triethylamine (TEA), sodium hydroxide (NaOH), gadolinium (III) chloride
99.99 % (GdCl3) and pyrene were purchased from Sigma-Aldrich (Bucks,
Switzerland). All other chemicals were purchased from Acros Organics
(Basel, Switzerland). Phosphate Buffer Saline (PBS) pH 7.4 was
purchased from Invitrogen (Zug, Switzerland). Gd-DOTA (DOTAREM)
used as control was supplied by Guerbet AG (Zürich, Switzerland) as
sterile injectable solution at 0.5 M Gd concentration. Deuterated NMR
solvents were obtained from Cambridge Isotope Laboratories (Tewksbury,
USA). Where anhydrous conditions were necessary, standard syringe-
Synthesis
Synthesis of Tyr-MRI building block
Synthesis of (benzyloxy)carbonyl-L-tyrosine 1. Compound
1 was
synthesized according to published procedure.[45] Yield 96%. 1H NMR (600
MHz, DMSO-d6) δ 12.67 (s, 1H, 1H, COOH), 9.22 (s, 1H, 1H, OH), 7.58
(d, J = 8.4 Hz, 1H, ), 7.38 – 7.24 (m, 5H, 5H, Ar), 7.08 – 7.01 (m, 2H, 2H,
Ar), 6.69 – 6.63 (m, 2H, 2H, Ar), 4.98 (s, 2H, 2H, CH2-O), 4.10 (ddd, J =
10.4, 8.3, 4.5 Hz, 1H, 1H, CH), 2.94 (dd, J = 13.9, 4.5 Hz, 1H, 1H, CHa),
2.72 (dd, J = 13.9, 10.4 Hz, 1H, 1H, CHb). 13C NMR (151 MHz, DMSO-d6)
δ 173.93, 156.45, 156.33, 137.49, 130.47, 128.76, 128.34, 128.16, 127.94,
115.41, 65.67, 56.35, 36.22. LRMS (ESI) m/z calculated for
C17H17NO5Na [M + Na]+ 338.33 found 338.55.
septa techniques were used under
a positive pressure of argon.
Compressed hydrogen was supplied from Pangas (Carouge, Switzerland)
with quality H2 4.5 UN 1049. Tetrahydrofuran (THF) and dichloromethane
(CH2Cl2) were obtained from an Anhydrous Engineering alumina column
based drying system. All other solvents used were of HPLC grade.
Chloroform (CHCl3) was purchased from Fisher Chemical (Basel,
Switzerland). N,N-dimethylformamide (DMF), methanol (CH3OH), diethyl
ether (Et2O) and acetone were purchased from Sigma-Aldrich (Buchs,
Switzerland). Ethyl acetate (AcOEt) was purchased from Biosolve (Dieuze,
France), acetonitrile (CH3CN) was supplied by Carlo Erba Reagents
(Balerna, Switzerland). Hexane ≥ 95 % of n-hexane was purchased from
Fisher Chemical (Basel, Switzerland). The water used for the preparations
was deionised by a Milli-Q lab water system (Millipore, Molsheim, France).
17O-enriched water, normalized 19.2 at percentage enrichment (purity >
99.99 % by mass fraction) was from Enritech Enrichment Technologies Ltd
(Israel).
Synthesis of (S)-benzyl (1-(dioctadecylamino)-3-(4-hydroxyphenyl)-1-
oxopropan-2-yl)carbamate 2. Z-(L)-Tyr-OH 1 (2.52 g, 7.98 mmol) and
DIPEA (2.78 mL, 16.0 mmol) were dissolved in CHCl3 (380 mL). HATU
(3.64 g, 9.58 mmol) was added and the reaction stirred at ambient
temperature for 10 minutes prior to the addition of dioctadecylamine (5.00
g, 9.58 mmol) portionwise. After stirring for 15 h at room temperature, the
urea side product was removed by filtration. The solvent was evaporated
under reduced pressure and the crude product purified by automated flash
column chromatography using CH2Cl2/MeOH gradient. The fractions
containing the compound were pooled and the solvents evaporated. The
product was dried in high vacuum to remove the remaining volatile urea
derivative to give compound 2 as white solid (3.50 g, 4.27 mmol, 66 %).
TLC (CH2Cl2/MeOH 15:1): Rf = 0.57. 1H NMR (500 MHz, CDCl3) δ 7.39 –
7.28 (m, 5H, Ar), 7.01 (d, J = 8.0 Hz, 2H, Ar), 6.68 (d, J = 8.5 Hz, 2H, Ar),
5.95 (br, 1H, OH), 5.62 (d, J = 8.8 Hz, 1H, NH), 5.09 (dd, 2H, CH2-O), 4.78
(td, J = 8.2, 6.0 Hz, 1H, CH-NH), 3.48 – 3.41 (m, 1H, CH2-CH), 3.09 – 3.00
(m, 2H, CH2), 3.00 – 2.91 (m, 2H, CH2), 2.86 (dd, J = 13.5, 6.0 Hz, 1H,
CH2-CH), 1.55 – 1.39 (m, 4H, CH2), 1.36 – 1.12 (m, 60H, aliphatic chain),
0.89 (t, J = 6.9 Hz, 6H, CH3). 13C NMR (500 MHz, CDCl3) δ 171.10, 155.78,
155.18, 136.51, 130.74, 128.62, 128.18, 128.04, 127.98, 115.49, 77.16,
66.93, 52.29, 47.96, 46.69, 40.40, 32.08, 29.88, 29.86, 29.78, 29.75, 29.71,
29.55, 29.52, 29.44, 29.19, 27.71, 27.21, 26.96, 22.84, 14.27. HRMS (ESI)
m/z calculated for C53H91N2O4+ [M + H]+ 819.6973, found 819.6959.
Methods
Thin layer chromatography (TLC) was performed with aluminium backed
silica plates (Merck-Keiselgel 60 F254) with a suitable solvent and was
visualized using a UV fluorescence lamp (254 and 366 nm) and/or
developed with nihydrine, sulfuric or phosphomolybdic acid (PMA).
Microwave reactions were performed in a Biotage® Initiator+ Robot Sixty
(Uppsala, Switzerland). Flash chromatography was performed on an
automated PuriFlash® 4100 machine from Interchim (Montlucon, France)
equipped with a PDA detector (200 – 800 nm) and automated fraction
collector using Interchim silica columns puriFlash® HP 30 μm. The system
was monitored using Flash Interchim software version 5.0x. Aluminium
oxide columns were prepared with Al2O3 neutral Brockmann I for
chromatography, 50-200 μm, 60
Switzerland). Semi-preparative HPLC column was conducted on a Waters
Symmetry 300TM - 5 μm (19 x 150 mm), C4 column (Baden-Dättwil,
Å from Acros Organics (Basel,
Switzerland). Size exclusion chromatography was performed on
a
Synthesis
of
(S)-2-amino-3-(4-hydroxyphenyl)-N,N-dioctadecyl-
Sephadex® G-10 from Sigma-Aldrich (Buchs, Switzerland) with Milli-Q
water as eluent. Analytical UPLC was conducted using Macherey-Nagel
EC50/2 Nucleodur Gravity 1.8 µm column (50 x 2.1 mm) fitted on a Waters
system equipped with a Waters PDA detector (Baden-Dättwil, Switzerland).
Buffer A = CH3CN + 0.1 % Formic acid), buffer B = H2O + 0.1 % Formic
acid. Flow rate = 400.0 µL/min at 25 °C. The retention times (TR)
propanamide 3. Pd(OAc)2 (35 mg, 0.155 mmol) was added to a solution of
2 (2.54 g, 3.10 mmol) in a mixture of dry THF/MeOH (1:1, 60.0 mL) under
argon atmosphere. Argon was replaced by hydrogen and the mixture was
stirred at ambient temperature and 1 bar for 20 h. The reaction mixture
was filtered through a short plug of Celite® and washed with THF (2×10
mL). After evaporation of the solvents, the product was purified by
automated flash chromatography using a gradient CH2Cl2/MeOH for 20
column volumes. The fractions were pooled and evaporated to dryness to
give compound 3 as colorless solid (1.21 g, 1.77 mmol, 57 %). TLC TLC
determined by the Empower2 software are quoted in minutes. 1H and 13
NMR spectra were recorded on Varian Gemini 300 MHz, Varian Innova
500 MHz or Bruker Avance III Cryo 600 MHz spectrometers at 298 K.
C
6
This article is protected by copyright. All rights reserved.

10.1002/chem.201703962
Chemistry - A European Journal
FULL PAPER
(CH2Cl2/MeOH 15:1): Rf = 0.48. 1H NMR (300 MHz, CDCl3) δ 6.99 (d, J =
8.4 Hz, 2H, Ar), 6.69 (d, J = 8.4 Hz, 2H, Ar), 3.84 (t, J = 6.8 Hz, 1H, CH-
NH2), 3.55 – 3.41 (m, 2H, CH2), 3.19 – 2.96 (m, 4H, CH2), 2.91 (dd, J =
13.6, 6.1 Hz, 1H, CH2-CH), 2.67 (dd, J = 13.6, 7.5 Hz, 1H, CH2-CH), 1.56
eluting 3.5 % MeOH in CH2Cl2 to give 6 as slightly orange oil that slowly
solidified upon standing (126 mg, 0.04 mmol, 36% yield). TLC on Al2O3
plate (CH2Cl2/CH3OH 20:1): Rf = 0.35. 1H NMR (300 MHz, CD3OD) δ 1H
NMR (300 MHz, CDCl3) δ 7.73 (br s, 2H), 7.48 (br s, 2H), 7.13 (br s, 1H),
6.94 (d, 2H), 6.73 (d, 2H), 5.17 (br s, 1H), 4.28 (m, 2H), 3.62 (br s, 164H),
3.35 (s, 3H), 3.50-2.81 (m, 16H), 1.46-1.42 (m, 18H), 1.23 (br s, 64 H),
0.85 (br t, 6H). 13C NMR (300 MHz, DMSO-d6) δ 170.69, 169.74, 165.06,
155.99, 130.04, 127.97, 127.80, 114.93, 99.57, 71.30, 70.32, 69.80, 69.61,
69.23, 68.66, 68.14, 58.06, 51.52, 47.17, 45.62, 36.91, 31.34, 29.08, 28.96,
28.90, 28.76, 28.62, 27.77, 27.15, 26.25, 26.21, 22.14, 13.95. MALDI-TOF
m/z calculated for C158H296N6O54 [M] 3144.05, found a polydisperse profile
with 44 Da distribution around 3186.92 Da.
– 1.38 (m, 4H, CH2), 1.34 – 1.16 (m, 60H, aliphatic chain), 0.86 (t, 6H). 13
C
NMR (500 MHz, CDCl3) δ 174.14, 155.49, 130.48, 128.68, 115.78, 52.99,
47.69, 46.76, 41.75, 32.08, 29.86, 29.84, 29.81, 29.80, 29.79, 29.77, 29.75,
29.70, 29.58, 29.52, 29.49, 29.46, 27.79, 27.23, 27.06, 22.84, 14.28.
+
HRMS (ESI) m/z calculated for C45H85N2O2 [M + H]+ 685.6606, found
685.6591.
Synthesis
of
(S)-4-(chloromethyl)-N-(1-(dioctadecylamino)-3-(4-
hydroxyphenyl)-1-oxopropan-2-yl)benzamide 4. 4-chloro-methylbenzoyl
chloride (386 mg, 2.04 mmol) was added to a solution of 3 (1.40 g, 2.04
mmol) and KI (140 mg, 0.83 mmol) in dry CH2Cl2 (70 mL) at -15°C. The
reaction mixture was left for 10 minutes under argon atmosphere at this
temperature. A solution of TEA (285 µL in 12 mL dry CH2Cl2, 2.043 mmol)
was then added dropwise over 15 minutes and the reaction mixture
maintained at -15°C. After 2h the temperature was allowed to rise to room
temperature overnight. Potassium iodide were filtered off and the
supernatant was evaporated to dryness under vacuum. The residue was
purified by automated flash column chromatography using n-
hexane/ethylacetate gradient. 4 was obtained as colorless solid after
solvent evaporation and drying in vacuo (1.19 g, 1.42 mmol, 70 %). TLC
(Hex/AcOEt 10:3): Rf = 0.30. 1H NMR (300 MHz, CDCl3) δ 7.76 (d, J = 8.2
Hz, 2H, Ar), 7.41 (d, J = 8.2 Hz, 2H, Ar), 7.19 (d, J = 8.2 Hz, 1H, NH), 7.01
(d, J = 8.4 Hz, 2H, Ar), 6.68 (d, J = 8.4 Hz, 2H, Ar), 5.34 – 5.23 (m, 1H,
CH-NH), 4.58 (s, 2H, CH2-Cl), 3.49 (dd, J = 7.0 Hz, 1H, CH2), 3.25 – 3.01
(m, 4H, CH2), 2.96 (dd, J = 13.7, 5.7 Hz, 1H, CH2), 1.54 – 1.44 (m, 4H,
CH2), 1.37 – 1.15 (m, 60H), 0.86 (t, 6H, CH3). 13C (300 MHz, CDCl3):
171.22, 166.12, 155.88, 141.13, 133.74, 130.64, 128.77, 127.71, 127.04,
115.61, 51.02, 48.15, 46.83, 45.43, 38.96, 32.05, 29.85, 29.80, 29.75,
29.70, 29.50, 29.43, 27.69, 27.20, 26.95, 22.82, 14.25. HRMS (ESI) m/z
calculated for C53H90ClN2O3+ [M + H]+ 837.6635, found 837.6634.
Synthesis
of
2,2',2''-(10-(4-((1-(dioctadecylamino)-3-(4-(2-(2-
methoxy(ethoxy)44ethoxy)phenyl)-1-oxopropan-2-yl) carbamoyl) benzyl)-
1,4,7,10-tetraazacyclododecane-1,4,7-triyl)triacetate 7. Compound 6 (705
mg, 0.21 mmol) was treated in a mixture of TFA/CH2Cl2 (50/50, 1.5 mL) at
room temperature. After 15 h, the solvents were removed under reduced
pressure. The crude product was washed repeatedly with diethyl ether
before further drying under high vacuum overnight to afford the compound
7 as a TFA salt (669 mg, 0.21 mmol, 99% yield). 1H NMR (500 MHz,
CD3OD) δ 7.95 – 7.79 (m, 2H, Ar), 7.79 – 7.61 (m, 2H, Ar), 7.19 (d, J = 8.6
Hz, 2H, Ar), 7.09 (d, J = 8.1, 4.0 Hz, IMP), 6.87 (d, J = 8.7 Hz, 2H, Ar),
6.71 (d, J = 8.4 Hz, IMP), 5.17 (t, J = 7.4 Hz, 1H, CH-NH), 4.12 – 4.05 (m,
2H, CH2), 3.87 – 2.95 (m, 211H, PEG chain, CH2-Ar, CH2-cyclen, CH2-
alkyles), 2.99 (dd, 2H, CH2-CH), 1.64 – 1.43 (m, 4H, CH2), 1.43 – 1.21 (m,
56H, aliphatic chain), 0.90 (t, J = 6.9 Hz, 6H, CH3). 13C NMR (300 MHz,
CDCl3) δ 186.15 (IMP), 171.77, 161.00 - 159.44, 158.07, 130.51, 128.50,
127.99, 121.27, 117.45, 114.80 (CF3COOH), 113.64, 109.82 (IMP), 80.87,
71.84, 70.73, 70.50, 70.40, 69.83, 59.18, 52.20, 49.16, 32.04, 29.88, 29.70,
29.48, 29.36, 27.12, 26.98, 22.81, 14.25. MALDI-TOF m/z calculated for
C158H296N6O54 [M] 3144.05, found a polydisperse profile with 44 Da gap
around 3186.92 Da.
Synthesis
of
2,2',2''-(10-(4-((1-(dioctadecylamino)-3-(4-(2-(2-
Synthesis of (S)-tri-tert-butyl 2,2',2''-(10-(4-((1-(dioctadecylamino)-3-(4-
methoxy(ethoxy)44ethoxy)phenyl)-1-oxopropan-2-yl) carbamoyl) benzyl)-
1,4,7,10-tetraazacyclododecane-1,4,7-triyl) gadolinium triacetate 8. GdCl3
solution (38 mM, 414.0 µL, 0.016 mmol, 0.90 eq) previously adjusted to
pH 5.8 with NaOH (0.10 M) to a stirred aqueous solution of TFA salt of 7
(55 mg, 17 µmol) in 1.00 mL H2O previously adjusted to pH 5.6 with NaOH
(0.10 M). pH was kept constant to 5.8 for 3 days using NaOH (0.10 M and
20 mM). After 3 days at room temperature, the clear solution was warmed
up at 60 °C for 15 h. pH was readjusted to 5.75. The inorganic salts were
removed by size-exclusion chromatography on Sephadex G-10 column
eluted with water. The fractions of interest were freeze-dried to afford
colourless solid. The absence of free gadolinium was verified by the
xylenol orange test.[46] Analytical NMR spectra not available due to the
paramagnetic properties of the sample. MALDI-TOF m/z calculated for
C158H296GdN6O54 [M] 3296.96, found a polydisperse profile with 44 Da gap
around 3186.92 Da. BMS [Gd3+] in GdCl3 stock solution: theoretical =
40.22 mM, found 37.94 mM. BMS [Gd3+] in 8: theoretical = 15.12 mM,
found 10.13 mM, corresponding to 0.67 loading. ICP-MS: theoretical =
5.00 ppb, found 3.65 ppb, corresponding to 0.70 Gd3+ loading.
hydroxyphenyl)-1-oxopropan-2-yl)
carbamoyl)
benzyl)-1,4,7,10-
tetraazacyclododecane-1,4,7-triyl)triacetate 5. 4 (400 mg, 0.78 mmol) was
dissolved in DMF (20 mL) under argon atmosphere. Oven dried K2CO3
(100 mg, 0.72 mmol) and KI (40 mg, 0.24 mmol) were added to the
DO3A(tBu)3 (295 mg, 0.57 mmol) was transferred into the reaction mixture
left stirring at room temperature for 72 hours. After filtration, the solvent
was removed in vacuo. The product was isolated by the general procedure
used in both methods of synthesis. The crude product was isolated by
automated flash chromatography column acetone/CH2Cl2 gradient. The
fractions containing the compound were evaporated to afford compound 5
as an orange solid (615 mg, 0.47 mmol, 60 %). TLC (CH2Cl2/MeOH 15:1):
Rf = 0.21. 1H NMR (300 MHz, CDCl3) δ 7.74 (d, J = 8.2 Hz, 2H, Ar), 7.52
(d, J = 8.2 Hz, 2H, Ar), 7.18 (s, 1H, OH), 7.07 (d, J = 8.0 Hz, 1H, NH), 6.99
(d, J = 8.4 Hz, 2H, Ar), 6.86 (d, J = 8.4 Hz, 2H, Ar), 5.17 (dd, J = 6.6 Hz,
1H, CH-NH), 3.57 – 1.91 (m, 30H, CH2-CH, CH2-N, CH2-CH2), 1.56 – 1.34
(m, 31H, CH3, CH2-CH2), 1.33 – 1.09 (m, 60H, CH2), 0.86 (t, J = 6.6 Hz,
6H, CH3). 13C NMR (300 MHz, CDCl3) δ 173.64, 172.62, 170.92, 169.28,
165.46, 155.97, 141.60, 133.20, 130.59, 130.49, 127.56, 126.93, 115.81,
83.04, 82.51, 59.45, 56.06, 55.79, 51.00, 47.79, 46.74, 39.40, 32.02, 29.82,
29.76, 29.68, 29.46, 29.39, 29.13, 28.38, 28.33, 28.29, 28.28, 28.21, 28.10,
27.99, 27.67, 27.25, 26.89, 22.79, 14.24. HRMS (ESI) m/z calculated for
C79H139N6O9+ [M + H]+ 1316.0598, found 1316.0595.
Gadolinium determination
Inductively coupled plasma mass spectrometry (ICP-MS). The gadolinium
ion metal content was determined by ICP-MS performed on an Agilent
7700x spectrometer (Agilent Technologies, Basel, Switzerland) at the
University of Geneva. Rhenium (185Re) was used as internal standard. The
measurements were performed on isotope 155Gd, 156Gd, 157Gd, 158Gd,
159Gd and 160Gd. The Gd3+ concentration was determined with respect to
standards within the same concentration range (0.1 ppb to 500.0 ppb). The
carrier gas flow rate of the spectrometer amounted to 1.05 L/min. The ICP
RF power was set at 1550 W. The torch depth was set to be 10 mm. The
concentration of gadolinium ion assessed by ICP-MS was performed in the
blank matrix provided (gadolinium free compound 7) and in 2 % aqueous
solution of HNO3. All the solutions (blank, standards, and samples) were
Synthesis of (R)-tri-tert-butyl 2,2',2''-(10-(4-((1-(dioctadecylamino)-3-(4-(2-
(2-methoxy(ethoxy)44ethoxy)phenyl)-1-oxopropan
-2-yl)carbamoyl)
benzyl) -1,4,7,10-tetraazacyclododecane-1,4,7-triyl)triacetate 6. 5 (150
mg, 0.11 mmol) and oven dried Cs2CO3 (186 mg, 0.57 mmol) in anhydrous
DMF (3.9 mL) were stirred together for 5 minutes, prior to the addition of
the Br-PEG2000-OMe (1.5 eq) under argon atmosphere. The reaction
mixture was refluxed up to 100 °C for 4 hours before removing the heating
plate and left stirring continued for 15 hours at room temperature. The
volatiles from the filtrate were removed in vacuo. The crude was
redissolved in CH2Cl2 and purified on alumina chromatography column
7
This article is protected by copyright. All rights reserved.

10.1002/chem.201703962
Chemistry - A European Journal
FULL PAPER
prepared with a final concentration of 50 ppm in 2 % HNO3 Milli-Q aqueous
solution (30 mL) and filtered through a 0.45 μm PVDF filter. The prepared
samples were stored at 4 °C prior to further dilution down to 50 ppb in 2 %
HNO3, and analysis. None of the samples neither the blank contained
sulfur, phosphorous, fluoride, nor chloride atoms.
grids and observed at 120 kV in the electron microscope. The carbon grids
were pretreated under plasma for 20 seconds at 0.3 Torr and 400 V.
Stability. Nanomicelles formed by nanoprecipitation in Milli-Q water at 1.0
mM of concentration were analyzed during 1 month by DLS at 22 °C.
Additionally the stability was evaluated for 1 week at 4 °C, 22 °C and 37 °C.
Influence of temperature was evaluated on micelles diluted 20/80 (V/V) in
Mili-Q water, Phosphate Buffered Saline (PBS, pH=7.4), and Dulbecco’s
Modified Eagle Medium (DMEM, pH=7.4). The stability was also evaluated
on micelles diluted 20/80 (V/V) in 0.09 %, 0.9 % and 1.8 %.NaCl solutions
at 22 °C.
Bulk Magnetic Suceptibility (BMS). BMS of gadolinium loaded building
block was performed on
a Bruker DRX-400 NMR spectrometer.
Concentration of the gadolinium was assessed following the Evans
method,[47] the shift of the internal standard tert-butanol alkyl protons in the
paramagnetic environment compared to the diamagnetic reference (Milli-Q
water) contained in a coaxial NMR tube.
Cell cytotoxicity. The cytotoxicity of the micelles was evaluated with a
standard MTT cell viability assay on the HT1080 cell line. The cell line was
purchased from ATCC. Cells were routinely maintained in high-glucose
(4.5 g/L) with L-glutamine and pyruvate DMEM (Gibco by Life
technologies) supplemented with 10 % (v/v) heat inactivated fetal calf
serum (FCS) (Eurobio, France) and penicillin-streptomycin (10,000 U/mL)
(Gibco by life technologies). They were incubated at 37 °C in humidified
5 % CO2 atmosphere. They were split every two days with trypsin/EDTA
solution if confluent. Cells (100 µL) were seeded in a 96-well plate at initial
density of 1x104 cells/well. After 24 h of incubation, cells reached 80 % of
confluency and the media was replaced by a solution of micelles or Gd-
DOTA (Dotarem) ( 100 µL) for comparison at different concentrations (1.9,
3.9, 7.8, 15.6, 31.25, 62.5, and 125 µM). Then after 24 h or 48 h of
incubation the media was replaced by MTT solution (100µL, 0.65mg/mL),
after one washing with PBS. After further incubation for 3 h in incubator,
MTT solution was replaced with DMSO (50 µL) and incubated for 30 min
more. The absorbance of the formazan dissolved in DMSO was read at
570 nm by the BioTek Synergy Mx microplate reader and the results
recorded with the Gen5 2.00 Software. The cell viability was calculated as
follows: Cell viability (%) = [(Absorbance of treated well)-(Absorbance of
non-treated well)] / [(Absorbance of non-treated well)-(Absorbance of
negative control)]*100. The viability for each concentration were performed
in triplicates.
NMR measurements
Nuclear Magnetic Relaxation Dispersion profiles (1H NMRD) were
measured with the inversion recovery method[39] at 25.0 °C and 37.0 °C on
Bruker Minispecs mq20 20 MHz (0.47 T), mq30 30 MHz (0.71 T), mq40
40 MHz (0.93 T), mq60 60 MHz (1.41 T), on a Bruker Avance-200 console
connected to 100 MHz (2.35 T) and 200 MHz (4.7 T) cryomagnets, and on
a Bruker Avance-II 400 MHz (9.4 T). For variable-temperature 17O NMR
spectroscopy, measurements were performed on a Bruker Avance-II
spectrometer at 54.3 MHz (9.4 T). Sealed spherical NMR samples
adapted for 10 mm NMR tubes were used to avoid susceptibility
corrections of the chemical shift.[48] Relaxation rates R1 (1/T1) and R2 (1/T2)
were obtained by the inversion-recovery method and by the Carr-Purcell-
Meiboom-Gill method.[49] For all experiments, the temperature was
measured by the substitution technique.[50]
Tyr-MRI nanomicelles
Preparation. Thin film method. 3.2 mg of compound 8 (1.0 µmol) was
dissolved in chloroform (2.0 mL) in a round-bottom flask and mixed for 1 h.
The organic solvent was removed using a rotary vacuum evaporation and
was further dried under a constant flow of nitrogen for 30 min to form a thin
film. Ultra-pure water (1.0 mL) was added to the thin film to re-hydrate and
vortexed for 1 min. The solution was heated at 40°C and stirred to get a
preliminary nanomicelles. The nanomicelles were then ultrasonicated for
20 s using 2+2 s pulse-pause sequence. The final nanomicelle solution
was obtained by filtering through 0.22 mm filter.
MRI phantoms. Standard Echo Gradient 2D MRI on phantom solutions
were performed in a 96 well plate using Magnet 3.0 T RS2D Mediso with
8 cm core and equipped with a volume coil (excitation/reception). The MRI
parameters were used for coronal section, 12 slices x 1 mm thickness,
FOV = 80 x 40 mm, TE/TR = 3.78 / 101 ms, matrix = 256 x 128, Flip Angle
= 75°, 20 averages, duration = 4 min 35 sec. The reference of Gd-DOTA
was placed on top left corner for plate orientation.
Nanoprecipitation method.[34] 3.2 mg (1.0 µmol) of compound 8 was
dissolved in a mixture of acetone (200 µL) and ethanol (200 µL). The
organic phase was added dropwise into ultrapure water (1.40 mL) using a
microsyringe in 5 min under magnetic stirring. The organic solvents and
the excess of water was evaporated under reduced pressure in 30 min.
The final mass of the sample was 1.00 g, yielding the 1.00 mM
concentration of nanomicelles containing 0.70 mM concentration of Gd3+
ions.
Animal experiments. The authorization for animal experiments
(GE/50/17) was obtained from the veterinary authorities of the canton of
Geneva, Switzerland. Male mice C57BL/6J were injected in the tail vein
either with Gd-DOTA or with Tyr-MRI. The CAs (200 µL at 3.0 mM) diluted
in PBS were injected via a tail catheter (30 G). MRI scans were performed
before the injection, immediately after injection, at 30 min, 1 h and 2 h after
injection. The mice were sacrificed after the last scan.
In vivo MRI imaging. The images were obtained with a standard 3D
Gradient Echo sequence with a 3.0 T MRI RS2D (Mediso, Hungary) with
8 cm bore and equipped with a volume coil (excitation/reception). The
parameters used for coronal section MRI: 45 slices x 0.5 mm thickness,
FOV = 80 x 40 mm, TE/TR = 4.4 / 10 ms, matrix = 256 x 128, Flip Angle =
70°, 20 averages, total scan duration: 12 min 40 s.
Characterization. Dynamic light scattering (DLS). Hydrodynamic
diameters of the nanomicelles were measured by DLS using a NANO ZS
ZEN 3600 Zeta-potential/Particle system instrument from Malvern
(Lausanne, Switzerland) running the ZetaSizer 7.01. software. The
analyses were performed with 4 mW He–Ne Laser (633 nm) at scattering
angle of 173° at 25°C in Polystyrene (PS) semi-micro cuvette from VWR
(Nyon, Switzerland). Parameters were set to be 1.590 (polystyrene latex)
for the refractive index material with a material absorption of 0.010. Zeta
potential (ZP) was determined using the same Nano ZS 3600 instrument
from Malvern in folded capillary cells DTS 1070 from Malvern.
Acknowledgements
Transmission electron microscopy (TEM). The morphological examination
of micelles was performed using a TEM microscope FEI TECNAI® G2
Sphera equipped with a soft camera TCL (Gräfelfing, Germany). Samples
were negatively stained with uranyl 2 % on a carbon film 200 Mesh copper
The authors are grateful to the Sciences Mass Spectrometry
service platform at the University of Geneva for the HRMS and
MALDI-TOF analysis. We also thank Dr. Laurence Marcourt for
8
This article is protected by copyright. All rights reserved.

10.1002/chem.201703962
Chemistry - A European Journal
FULL PAPER
[19] E. Toth, R. D. Bolskar, A. Borel, G. Gonzalez, L. Helm, A. E.
Merbach, B. Sitharaman and L. J. Wilson, Journal of the American
Chemical Society 2005, 127, 799-805.
[20] J. P. Andre, C. F. G. C. Geraldes, J. A. Martins, A. E. Merbach,
M. I. M. Prata, A. C. Santos, J. J. P. de Lima and E. Toth, Chemistry-
a European Journal 2004, 10, 5804-5816.
the high-field NMR experiments, Dr. Christoph Bauer for his help
in TEM imaging. This work was performed within the framework
of the EU COST Action TD1004 “Theranostics imaging and
therapy: an action to develop novel nanosized systems for
imaging-guided drug delivery”.
[21] I. M. Mavridis and K. Yannakopoulou, International Journal of
Pharmaceutics 2015, 492, 275-290.
[22] a) S. Santra, S. D. Jativa, C. Kaittanis, G. Normand, J. Grimm
and J. M. Perez, ACS Nano 2012, 6, 7281-7294; b) B. B. Zhang, H.
T. Jin, Y. Li, B. D. Chen, S. Y. Liu and D. L. Shi, Journal of Materials
Chemistry 2012, 22, 14494-14501; c) F. Yazdani, B. Fattahi and N.
Azizi, Journal of Magnetism and Magnetic Materials 2016, 406, 207-
211.
Conflict of Interest
The authors declare no conflict of interest.
[23] a) E. Toth, L. Helm, K. E. Kellar and A. E. Merbach, Chemistry-a
European Journal 1999, 5, 1202-1211; b) S. Aime, M. Botta, E.
Garino, S. G. Crich, G. Giovenzana, R. Pagliarin, G. Palmisano and
M. Sisti, Chemistry-a European Journal 2000, 6, 2609-2617.
[24] a) J. Lim, B. Turkbey, M. Bernardo, L. H. Bryant, Jr., M. Garzoni,
G. M. Pavan, T. Nakajima, P. L. Choyke, E. E. Simanek and H.
Kobayashi, Bioconjug Chem 2012, 23, 2291-2299; b) M. R.
Longmire, M. Ogawa, P. L. Choyke and H. Kobayashi, Wiley
Interdisciplinary Reviews-Nanomedicine and Nanobiotechnology
2014, 6, 155-162.
[25] a) V. Weissig, J. Babich and V. Torchilin, Colloids and Surfaces
B-Biointerfaces 2000, 18, 293-299; b) J. Zou, R. Sood, S. Ranjan, D.
Poe, U. A. Ramadan, P. K. Kinnunen and I. Pyykko, J
Nanobiotechnology 2010, 8, 32; c) Y. Xiao, Y. Liu, S. Yang, B.
Zhang, T. Wang, D. Jiang, J. Zhang, D. Yu and N. Zhang, Colloids
Surf B Biointerfaces 2016, 141, 83-92.
[26] a) J. P. Andre, E. Toth, H. Fischer, A. Seelig, H. R. Macke and
A. E. Merbach, Chemistry-a European Journal 1999, 5, 2977-2983;
b) A. Accardo, D. Tesauro, P. Roscigno, E. Gianolio, L. Paduano, G.
D'Errico, C. Pedone and G. Morelli, Journal of the American
Chemical Society 2004, 126, 3097-3107; c) A. Accardo, D. Tesauro,
A. Morisco, G. Mangiapia, M. Vaccaro, E. Gianolio, R. K. Heenan, L.
Paduano and G. Morelli, Journal of Biological Inorganic Chemistry
2009, 14, 587-599; d) M. Grogna, R. Cloots, A. Luxen, C. Jerome, C.
Passirani, N. Lautram, J. F. Desreux and C. Detrembleur, Polymer
Chemistry 2010, 1, 1485-1490; e) S. Dixit, M. Das, S. Alwarappan,
N. L. Goicochea, M. Howell, S. Mohapatra and S. Mohapatra, Rsc
Advances 2013, 3, 2727-2735.
[27] a) K. J. Chen, S. M. Wolahan, H. Wang, C. H. Hsu, H. W.
Chang, A. Durazo, L. P. Hwang, M. A. Garcia, Z. K. Jiang, L. Wu, Y.
Y. Lin and H. R. Tseng, Biomaterials 2011, 32, 2160-2165; b) E.
Poselt, H. Kloust, U. Tromsdorf, M. Janschel, C. Hahn, C. Masslo
and H. Weller, Acs Nano 2012, 6, 1619-1624.
Keywords: magnetic resonance imaging • contrast agent •
gadolinium • nanomicelles • self-assembly
[1] R. Weissleder, Molecular Imaging: Principles and Practice,
People's Medical Publishing House, 2010, p.
[2] A. R. Kherlopian, T. Song, Q. Duan, M. A. Neimark, M. J. Po, J.
K. Gohagan and A. F. Laine, BMC Syst Biol 2008, 2, 74.
[3] L. Helm, A. E. Merbach and E. v. Tꢀth, The chemistry of contrast
agents in medical magnetic resonance imaging, John Wiley & Sons
Inc., Hoboken, NJ, 2013, p. xvi, 496 pages.
[4] L. Faucher, M. Tremblay, J. Lagueux, Y. Gossuin and M. A.
Fortin, ACS Appl Mater Interfaces 2012, 4, 4506-4515.
[5] L. Helm, Future Med Chem 2010, 2, 385-396.
[6] a) P. Chandran, A. Sasidharan, A. Ashokan, D. Menon, S. Nair
and M. Koyakutty, Nanoscale 2011, 3, 4150-4161; b) Z. X. Zhou, M.
Qutaish, Z. Han, R. M. Schur, Y. Q. Liu, D. L. Wilson and Z. R. Lu,
Nature Communications 2015, 6; c) Z. X. Zhou, Z. Han and Z. R. Lu,
Biomaterials 2016, 85, 168-179.
[7] a) A. Dirksen, S. Langereis, B. F. M. de Waal, M. H. P. van
Genderen, T. M. Hackeng and E. W. Meijer, Chemical
Communications 2005, 2811-2813; b) I. M. Debergh, P. Smeets, W.
P. Ceelen and M. Peeters, Annals of Oncology 2008, 19, 253-253.
[8] a) M. J. Lipinski, V. Amirbekian, J. C. Frias, J. G. 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; b) J. Sanz and Z. A.
Fayad, Nature 2008, 451, 953-957.
[9] A. H. H. Leung, J. F. Jin, S. X. Wang, H. Lei and W. T. Wong,
Bioconjugate Chemistry 2014, 25, 1112-1123.
[10] a) Y. S. Lim, S. Y. Kim, J. H. Byun, H. J. Won, S. J. Lee, S. Han,
J. An, H. C. Lee and Y. S. Lee, Hepatology 2014, 60, 843a-843a; b)
A. M. Chow, M. Q. Tan, D. S. Gao, S. J. Fan, J. S. Cheung, Z. W.
Qiao, K. Man, Z. R. Lu and E. X. Wu, Investigative Radiology 2013,
48, 46-54.
[11] a) B. McMahon, P. Mauer, C. P. Mccoy, T. C. Lee and T.
Gunnlaugsson, Journal of the American Chemical Society 2009,
131, 17542-17543; b) J. E. Adams, Nature Reviews Endocrinology
2013, 9, 28-42.
[12] a) Z. J. Liu, X. X. Song and Q. Tang, Nanoscale 2013, 5, 5073-
5079; b) N. Chen, C. Shao, Y. M. Qu, S. Li, W. Gu, T. T. Zheng, L.
Ye and C. J. Yu, Acs Applied Materials & Interfaces 2014, 6, 19850-
19857.
[13] S. P. Lin and J. J. Brown, Journal of Magnetic Resonance
Imaging 2007, 25, 884-899.
[14] L. Yang, I. Krefting, A. Gorovets, L. Marzella, J. Kaiser, R.
Boucher and D. Rieves, Radiology 2012, 265, 248-253.
[15] V. M. Runge, T. Ai, D. P. Hao and X. M. Hu, Investigative
Radiology 2011, 46, 807-816.
[16] a) M. R. Prince, D. L. Narasimham, J. C. Stanley, T. L.
Chenevert, D. M. Williams, V. Marx and K. J. Cho, Radiology 1995,
197, 785-792; b) G. D. Zhang, R. Zhang, M. P. Melancon, K. Wong,
J. You, Q. Huang, J. Bankson, D. Liang and C. Li, Biomaterials
2012, 33, 5376-5383.
[17] M. Lewis, S. Yanny and P. N. Malcolm, Journal of Medical
Imaging and Radiation Oncology 2012, 56, 187-191.
[18] K. W. Y. Chan and W. T. Wong, Coordination Chemistry
Reviews 2007, 251, 2428-2451.
[28] V. P. Torchilin, Advanced Drug Delivery Reviews 2002, 54, 235-
252.
[29] K. Knop, R. Hoogenboom, D. Fischer and U. S. Schubert,
Angew Chem Int Ed Engl 2010, 49, 6288-6308.
[30] a) A. S. Karakoti, S. Das, S. Thevuthasan and S. Seal,
Angewandte Chemie-International Edition 2011, 50, 1980-1994; b)
H. J. Gao, J. J. Liu, C. H. Yang, T. J. Cheng, L. P. Chu, H. Y. Xu, A.
M. Meng, S. J. Fan, L. Q. Shi and J. F. Liu, International Journal of
Nanomedicine 2013, 8, 4229-4246.
[31] M. Port, J. M. Idee, C. Medina, C. Robic, M. Sabatou and C.
Corot, Biometals 2008, 21, 469-490.
[32] N. Kamaly, T. Kalber, A. Ahmad, M. H. Oliver, P. W. So, A. H.
Herlihy, J. D. Bell, M. R. Jorgensen and A. D. Miller, Bioconjugate
Chemistry 2008, 19, 118-129.
[33] D. R. Lu, Y. Wang, H. Wang and R. K. Bai, European Polymer
Journal 2010, 46, 1417-1424.
[34] A. Babič, S. Pascal, R. Duwald, D. Moreau, J. Lacour and E.
Allémann, Advanced Functional Materials 2017.
[35] G. Wang, J. J. Wang, X. L. Chen, L. Du and F. Li, Journal of
Controlled Release 2016, 235, 276-290.
[36] G. M. Nicolle, E. Toth, K. P. Eisenwiener, H. R. Macke and A. E.
Merbach, Journal of Biological Inorganic Chemistry 2002, 7, 757-
769.
[37] D. H. Powell, O. M. NiDhubhghaill, D. Pubanz, L. Helm, Y. S.
Lebedev, W. Schlaepfer and A. E. Merbach, Journal of the American
Chemical Society 1996, 118, 9333-9346.
[38] F. A. Dunand, E. Toth, R. Hollister and A. E. Merbach, Journal
of Biological Inorganic Chemistry 2001, 6, 247-255.
9
This article is protected by copyright. All rights reserved.

10.1002/chem.201703962
Chemistry - A European Journal
FULL PAPER
[39] R. L. Vold, J. S. Waugh, M. P. Klein and D. E. Phelps, Journal of
Chemical Physics 1968, 48, 3831-&.
[40] S. Laurent, L. V. Elst and R. N. Muller, Contrast Media Mol
Imaging 2006, 1, 128-137.
[46] a) G. Brunisholz and M. Randin, Helvetica Chimica Acta 1959,
42, 1927-1938; b) A. Barge, G. Cravotto, E. Gianolio and F. Fedeli,
Contrast Media & Molecular Imaging 2006, 1, 184-188.
[47] D. F. Evans, J. Chem. Soc. 1959, 2003-2005.
[41] S. Torres, J. A. Martins, J. P. Andre, C. F. Geraldes, A. E.
Merbach and E. Toth, Chemistry 2006, 12, 940-948.
[42] S. Laus, A. Sour, R. Ruloff, E. Toth and A. E. Merbach,
Chemistry 2005, 11, 3064-3076.
[43] W. J. Mulder, G. J. Strijkers, G. A. van Tilborg, A. W. Griffioen
and K. Nicolay, NMR Biomed 2006, 19, 142-164.
[44] S. Aime and P. Caravan, Journal of Magnetic Resonance
Imaging 2009, 30, 1259-1267.
[48] A. D. Hugi, L. Helm and A. E. Merbach, Helvetica Chimica Acta
1985, 68, 508-521.
[49] a) S. Aime, M. Botta, S. G. Crich, G. B. Giovenzana, R.
Pagliarin, M. Piccinini, M. Sisti and E. Terreno, Journal of Biological
Inorganic Chemistry 1997, 2, 470-479; b) S. Meiboom and D. Gill,
Review of Scientific Instruments 1958, 29, 688-691.
[50] C. Ammann, P. Meier and A. E. Merbach, Journal of Magnetic
Resonance 1982, 46, 319-321.
[45] M. Li, B. L. Luo, Q. Liu, Y. M. Hu, A. Ganesan, P. Huang and S.
J. Wen, Organic Letters 2014, 16, 10-13.
10
This article is protected by copyright. All rights reserved.

10.1002/chem.201703962
Chemistry - A European Journal
FULL PAPER
Entry for the Table of Contents
FULL PAPER
Andrej Babič*, Vassily Vorobiev, Céline
Xayaphoummine, Gaëlle Lapicorey,
Anne-Sophie Chauvin, Lothar Helm and
Eric Allémann
Self-assembled nanomicelles as MRI
blood-pool contrast agent
11
This article is protected by copyright. All rights reserved.