MRI-visible poly(ε-caprolactone) with controlled contrast agent ratios for enhanced visualization in temporary imaging applications

Article abstract of DOI:10.1021/bm400978a

Hydrophobic macromolecular contrast agents (MMCAs) are highly desirable to provide safe and efficient magnetic resonance (MR) visibility to implantable medical devices. In this study, we report on the synthesis and evaluation of novel biodegradable poly(ε-caprolactone)-based MMCAs. Poly(α- propargyl-ε-caprolactone-co-ε-caprolactone)s containing 2, 5, and 10 mol % of propargyl groups have been prepared by ring-opening copolymerization of ε-caprolactone and the corresponding propargylated lactone. In parallel, a diazido derivative of the clinically used diethylenetriaminepentaacetic acid (DTPA)/Gd³⁺ complex has been synthesized. Finally, MRI-visible poly(ε-caprolactone)s (PCLs) were obtained by the efficient click ligation of these compounds via a Cu^I-catalyzed [3 + 2] cycloaddition. ICP-MS analyses confirmed the efficient coupling of the complex on the PCL backbone with the MRI-visible PCLs containing 1.0, 2.6, and 3.6 wt % of Gd³⁺. The influence of the Gd³⁺ grafting density on the T₁ relaxation times and on the MRI visibility of the novel biodegradable MMCAs was evaluated. Finally, their stability and cytocompatibility were assessed with regard to their potential as innovative MRI-visible biomaterials for biomedical applications.

Full text of DOI:10.1021/bm400978a

Article
pubs.acs.org/Biomac
MRI-Visible Poly(ε-caprolactone) with Controlled Contrast Agent
Ratios for Enhanced Visualization in Temporary Imaging
Applications
†
Sarah El Habnouni, Benjamin Nottelet,* Vincent Darcos, Barbara Porsio, Laurent Lemaire,
‡
Florence Franconi, Xavier Garric, and Jean Coudane
Institut des Biomolec
ENSCM, Faculty of Pharmacy, 15 Av. C. Flahault, Montpellier, 34093, France
́
ules Max Mousseron (IBMM), UMR CNRS 5247, University of Montpellier 1, University of Montpellier 2 −
†
‡
Micro et Nanomed
PIAM, UFR Sciences, 2bd Lavoisier, 49045 Angers, France
S Supporting Information
́
ecines Biomimet
́
iques (MINT), UMR-S 1066, Universite
́
d’Angers, 4 rue Larrey, 49933 Angers Cedex9, France
*
ABSTRACT: Hydrophobic macromolecular contrast agents (MMCAs) are highly desirable to provide safe and efficient
magnetic resonance (MR) visibility to implantable medical devices. In this study, we report on the synthesis and evaluation of
novel biodegradable poly(ε-caprolactone)-based MMCAs. Poly(α-propargyl-ε-caprolactone-co-ε-caprolactone)s containing 2, 5,
and 10 mol % of propargyl groups have been prepared by ring-opening copolymerization of ε-caprolactone and the
corresponding propargylated lactone. In parallel, a diazido derivative of the clinically used diethylenetriaminepentaacetic acid
+
(
DTPA)/Gd3 complex has been synthesized. Finally, MRI-visible poly(ε-caprolactone)s (PCLs) were obtained by the efficient
I
click ligation of these compounds via a Cu -catalyzed [3 + 2] cycloaddition. ICP-MS analyses confirmed the efficient coupling of
the complex on the PCL backbone with the MRI-visible PCLs containing 1.0, 2.6, and 3.6 wt % of Gd . The influence of the
Gd grafting density on the T relaxation times and on the MRI visibility of the novel biodegradable MMCAs was evaluated.
3+
3+
1
Finally, their stability and cytocompatibility were assessed with regard to their potential as innovative MRI-visible biomaterials for
biomedical applications.
INTRODUCTION
macromolecular contrast agents (MMCAs) able to self-
1
■
3,18−22
assemble into micellar systems.
These water-soluble
Magnetic resonance imaging (MRI) is today one of the
noninvasive technique of choice to provide high spatial and
temporal resolutions in clinical diagnosis and staging of human
MMCAs are rapidly excreted and cannot be used for the MRI
visualization of non water-soluble implanted devices. Another
strategy relies in the development of nonclassical MRI
techniques, as exemplified by the recent use of amide proton
transfer MRI technique to visualize hydrophilic and fast
1
diseases. Unfortunately, in the case of polymeric prostheses
there is an inability to provide a postoperative image of
implanted prostheses, as polymeric biomaterials are intrinsically
23
2
,3
resorbing collagen coatings. In this regard, the originality of
the postmodification approach used by our group lies in the
possibility of providing hydrophobic and multivalent MMCAs.
However, one limitation was the lack of fine control over the
final substitution degree of the polymeric backbone and the
necessity to use protection/deprotection steps to covalently
bind the ligand diethylenetriaminepentaacetic acid (DTPA)
before final complexation of the Gd3+ MRI contrast agent. Yet,
control of the overall amount of contrast agent, as well as its
repartition along the polymer chain and its chemical environ-
ment, are known to influence the resulting relaxivity and
transparent to X-rays and are invisible using clinical MRI. In
some cases, the signal void of the prosthetic material can be
useful for diagnoses, however it strongly depends on the size of
the prostheses and on the site of implantation. Therefore this
inability to visualize implanted material restricts the evaluation
of the tissue integration, the postoperative fixation and the
4
material’s fate in the body. This led us to consider in the past
the possibility to produce radio-opaque degradable polyesters.
A first example was provided by the radio-opaque poly(ε-
caprolactone) (PCL) obtained by the anionic chemical
modification of PCL with iodine to generate a poly(α-iodo-ε-
24
5
visualization. The strategy chosen in this work to prepare
hydrophobic and multivalent MMCAs addresses these points.
It is based on the convergent synthesis of an azide-containing
contrast agent, and of a propargyl-functionalized PCL whose
caprolactone-co-ε-caprolactone) copolymer. More recently,
this postmodification strategy was applied to the synthesis of
MRI-visible PCL and MRI-visible poly(methyl acrylate) (PMA)
that were among the first examples of hydrophobic MRI-visible
I
6
,7
late ligation occurs via Cu -catalyzed [3 + 2] cycloaddition
thermoplastics.
Indeed, although MRI polymers have
reaction. Practically, the control of the functionalization of
attracted much attention in the recent past, most efforts are
dedicated to water-soluble polycondensate or amphiphilic
8
−17
structures to be used in drug delivery approaches.
In
Received: July 5, 2013
particular, a preferred strategy relies on the modification of
Revised: September 4, 2013
diblock amphiphilic copolymers with a contrast agent to yield
©
XXXX American Chemical Society
A
dx.doi.org/10.1021/bm400978a | Biomacromolecules XXXX, XXX, XXX−XXX

Biomacromolecules
Article
1
propargylated PCLs is conveniently obtained by ROP of εCL
oil (2.46 g) was obtained (80% yield; Figure S1). RMN H (300 MHz,
25
CDCl ) δ (ppm): 3.33 (t, 2H, CH N ), 2.55 (t, 2H, CH NH ), 2.34
s, 2H, NH ), 1.55 (q, 2H, CH CH CH ). FT-IR (ATR, cm ): 2100
and αPgεCL, as recently described by our group, whereas the
use of a click chemistry reaction for ligation ensures a
quantitative functionalization of the polymer by the contrast
3
2
3
2
2
−1
(
(
2 2 2 2
N3).
In a second step, diN
modified described procedure. Typically, 1-azido-3-aminopropane
0.616 g; 6.2 mmol) was reacted with DTPA dianhydride (1 g; 2.8
3
+
-DTPA was synthesized according to a
3
27
agent. Finally, the direct coupling of the Gd complex on the
polymer avoids the late complexation step between the
(
3
+
macromolecular ligand and Gd , which is generally of lower
mmol) in anhydrous DMF (20 mL) at room temperature and under
nitrogen for 24 h. DMF was distilled under vacuum until 2−3 mL in
volume and precipitated into diethyl ether. The obtained solid was
further dissolved in water and lyophilized to yield a white solid (1.48 g,
efficiency to control the final amount of immobilized contrast
7
agent.
We thus propose in this work a so far nonexplored strategy
3
+
5% yield). RMN 1H (300 MHz, DMSO) δ (ppm): 3.4 (6H,
9
for the efficient and controlled Gd -functionalization of
aliphatic degradable polyesters to provide biocompatible and
slow degrading hydrophobic MRI-visible thermoplastics for
long-term MR visualization. Synthesis of copolyesters MMCAs
with finely controlled compositions is described. The influence
of the copolymers nature is discussed with respect to their MR
visualization. Finally, stability of these MMCAs and their
cytocompatibility are assessed.
CH CO H), 3.1 (12H, N(CH ) N and NCH C(O)), 2.8 (8H,
CH NHC(O) and CH N ), 1.7 (4H, CH CH N ). FT-IR (ATR,
cm ): 2100 (N ). LC-MS (ES+, m/z): 558.4 Da [M + H ].
2
2
2
2
2
2
2
3
2
2
3
−
1
+
3
Caution: ω-aminoalkyl azides being potentially explosives all reactions
involving this compound were carried out with the appropriate protection
under a well ventilated hood.
Preparation of Clickable Gd3 Complex 2. Pyridine (1.44 mL; 17.9
+
mmol) was added to an aqueous solution of 1 (1 g; 1.79 mmol). After
1
0 min stirring at room temperature a clear solution was obtained and
EXPERIMENTAL SECTION
GdCl ·6H O (1.33 g; 3.58 mmol) was added. Complexation was let to
3
2
■
run under stirring at 40 °C for 24 h. Water and pyridine were removed
by distillation under vacuum. The resulting solid was dissolved in
water before purification on a Chelex 100 resin. After lyophylization 2
Materials. Isopropanol, ε-caprolactone, and toluene were dried
over calcium hydride for 24 h at room temperature and distilled under
reduced pressure. Tetrahydrofuran was dried by refluxing over a
benzophenone-sodium mixture and distilled. All other materials were
obtained from Aldrich and were used without further purification.
3
+
was obtained as a white solid (90% yield). Absence of free Gd was
2
8
confirmed by a methyl thymol blue (MTB) test (Figure S2), whereas
3+
1
13
the content of complexed Gd in 2 was quantitatively determined by
NMR Spectroscopy. H and C NMR spectra were recorded on a
Bruker spectrometer (AMX300) operating at 300 and 75 MHz,
respectively. Deuterated chloroform or deuterated dimethyl sulfoxide
were used as solvents. Chemical shifts were expressed in ppm with
respect to tetramethylsilane (TMS).
−1
ICP-MS. FT-IR (ATR, cm ): 2100 (N3). MALDI-TOF (dithranol
+
+
m/z): [M + H] m/z 713.16 Da; [2M + H] m/z calcd, 1423.33 Da;
theoretical, 713.17 and 1425.34. Gd3 content (ICP-MS): 15.4 wt %.
Synthesis of Poly(α-propargyl-ε-caprolactone-co-ε-
caprolactone)s P(Pg-CL) 3. α-Propargyl-ε-caprolactone (αPgεCL)
was synthesized from εCL by an anionic modification approach
+
Relaxation times for polymers dissolved in DMSO-d6 were
measured on an AMX400 Bruker spectrometer operating at 400
MHz. T measurements were obtained using T inversion/recovery
25
already described elsewhere. Polymerization was carried out in bulk
1
1
using standard Schlenk technique under an inert atmosphere of argon.
Amounts of εCL and αPgεCL were added in the reaction flask
according to the targeted compositions. In a typical experiment, εCL
sequence. Analyses were carried out at 80 °C to facilitate the molecular
mobility of the polymer. A fixed concentration of 11 mg/mL was used
for all samples.
FT-IR Spectroscopy. Infrared spectra were recorded on a Perkin-
Elmer Spectrum 100 FT-IR spectrometer using the attenuated total
reflectance (ATR) method.
LC-MS Analyses. LC/MS analyses were performed on a Q-TOF
Waters) spectrometer fitted with an electrospray interface. Solvent
used for HPLC and LC/MS were HPLC grade. MALDI analyses were
performed on a Ultra-Flex III (Bruker) spectrometer using a dithranol
matrix.
Size Exclusion Chromatography. Size exclusion chromatography
SEC) was performed at room temperature on a Waters system
equipped with a guard column, a 600 mm PLgel 5 mm Mixed C
column (Polymer Laboratories), and a Waters 410 refractometric
detector. Calibration was established with poly(styrene) standards
from Polymer Laboratories. THF was used as eluent at a flow rate of 1
(
2.66 g; 29.6 mmol), αPgεCL (0.5 g; 3.29 mmol), Sn(Oct) (66.4 mg;
2
0
.164 mol), and isopropanol (25 μL; 0.328 mmol) were placed in an
oven-dried Schlenk tube. The tube was fitted with a rubber septum.
The solution was further degassed by three freeze−pump−thaw cycles.
The resulting mixture was stirred at 140 °C for 3.5 h. To stop the
reaction, polymerization was quenched by addition of an excess of 1 N
HCl. The reaction mixture was poured into cold methanol. The
precipated polymer 3 was collected by filtration and dried in vacuum.
(
1
H NMR (300 MHz, CDCl ) δ (ppm): 4.96 (m, (CH ) CH), 4.02 (t,
3
3 2
CH O), 3.60 (m, CH OH), 2.50 (m, COCHCH ), 2.26 (t, COCH ),
(
2
2
2
2
1
1
.96 (m, CCH), 1.49−1.70 (m, CH -CH -CH -CH -O-), 1.29−
2
2
2
2
.42 (m, m, CH -CH -CH -CH -O-), 1.18 (d, (CH ) CH). IR (ATR,
2 2 2 2 3 2
−1
cm ): 3280 (CCH), 1720 (CO). Mn
= 18000 g/mol, Đ =
SEC
1
.7.
Synthesis of MRI-Visible Poly(ε-caprolactone)s (MRI-PCLs) 4.
mL/min.
ICP-MS Analyses. Gd3+ was quantified using an Element XR sector
Copolymer 3, complex 2 (3 equiv/αPgεCL units) and CuBr (2 equiv/
αPgεCL units) were solubilized in DMF. The solution was degassed
by three freeze−pump−thaw cycles. N,N,N′,N″,N″-Pentamethyldie-
thylenetriamine (PMDETA) (2 equiv/ αPgεCL units) degassed by
argon bubbling was added to the reaction medium. The reaction was
carried out for 48 h at room temperature under stirring. The crude
medium was dissolved in THF for dialysis (MWCO 3500 g/mol)
against distilled water. Copolymer 4 was recovered after removal of the
solvents and drying in vacuum. Copolymers with 1.0, 2.6, and 3.6 wt %
field ICP-MS (inductively coupled plasma-mass spectrometry) at
́
Geosciences in Montpellier (University Montpellier II). Internal
standardization used an ultrapure solution enriched with indium.
Synthesis and Characterization of MRI-Visible PCL. Synthesis
of Diazido Functionalized Diethylenetriaminepentaacetic Acid
(
diN -DTPA) 1. In a first step, 1-azido-3-aminopropane was prepared
3
26
according to a previously described procedure. A solution of 3-
chloropropylamine hydrochloride (4 g; 30.8 mmol) and sodium azide
3+
(
6 g; 92.3 mmol) in water (1 mL per mmol) was heated at 80 °C
overnight. After removing most of the water by distillation under
vacuum, the reaction mixture was cooled in an ice bath. Diethyl ether
of Gd , as determined by ICP-MS, were noted 4 , 4 , and 4 ,
1 2 3
respectively.
MRI Visualization. MR Imaging Protocol. MR imaging experi-
ments were performed on a Bruker Avance DRX system (Bruker
Biospin SA, Wissembourg, France). The system was equipped with a
150 mm vertical superwide-bore magnet operating at 7 T, a 84 mm
inner diameter shielded gradient set capable of 144 mT·m− maximum
gradient strength and a 30 mm diameter birdcage resonator. Modified
(
50 mL) and KOH pellets (4 g) were added, keeping the temperature
below 10 °C in an ice bath. After separation of the organic phase, the
aqueous layer was further extracted with diethyl ether (2 × 20 mL).
The combined organic layers were dried over MgSO4 and
concentrated to give oil which was purified by distillation. Colorless
1
B
dx.doi.org/10.1021/bm400978a | Biomacromolecules XXXX, XXX, XXX−XXX

Biomacromolecules
Article
a
Scheme 1. Reaction Scheme for the Synthesis of MRI-PCLs 4
a
Reagents and conditions: (a) NaN , water, 80 °C, overnight; (b) DTPA dianhydride, DMF , RT, 24 h; (c) pyridine, RT, 10 min/GdCl ·6H O, 40
3
an
3
2
°
C, 24 h; (d) LDA, THF , −78 °C, 1 h/propargyl bromide, THF , −30 °C, 3 h; (e) εCL, Sn(Oct) , iPrOH, 140 °C, 3.5 h; (f) CuBr, PMDETA,
an
an
2
DMF, RT, 48 h.
2
or native PCL films (∼1 cm ) were embedded in a degassed 1% (w/
In Vitro Cytocompatibility. Murine fibroblasts cells (designated
L929) were used to assess the in vitro cytocompatibility of the
materials as recommended by the International and European
Standards (ISO 10993−5:2009). L929 cells were cultured in
DMEMα supplemented with 10% fetal bovine serum (FBS), penicillin
(100 U/mL), streptomycin (100 μg/mL), and glutamine (2 mM).
Sample disks (ø 15 mm) were cut from copolymer films and
disinfected in ethanol for 30 min before immersion in a solution of
sterile PBS containing penicillin and streptomycin (1 mg/mL) and
incubation for 48 h at 37 °C. Films were then rinsed 2 times with
sterile PBS before soaking for 12 h in sterile PBS. After disinfection,
disks were placed in TCPS 12-well plates and Viton O-rings were used
to maintain the samples on the bottom of the wells and avoid cells
growing on TCPS underneath the samples. Disks were finally seeded
w) agarose gel prior to imaging. Gadolinium-free samples
corresponded to a PCL film. To test signal enhancement, T weighting
was introduced into the MR images using an inversion pulse in rapid
three-dimensional (3-D) acquisition with relaxation enhancement
RARE) sequence (TR = 3000 ms; mean echo time (TEm) = 8 ms;
RARE factor = 8; FOV = 3 × 3 × 1.5 cm; matrix 128 × 128 × 64).
Inversion time was set at 1100 ms, sufficient to allow canceling of the
embedding gel.
Stability Studies. MRI-visible films were prepared with high (0.4 wt
) and low (0.1 wt %) contents of Gd . Films were prepared by
1
2
9
(
3+
%
dissolution of the appropriate amounts of PCL and 4 in dichloro-
2
methane followed by solvent evaporation. Film samples (10 mg) were
cut out for stability studies. Released gadolinium was quantified by
ICP-MS after sample immersion in 10 mL of PBS saline buffer and
stirring at 130 rpm and at 37 °C. At scheduled time points (1, 3, 7, 15,
0, and 90 days), 1 mL aliquots of solution were taken and replaced by
fresh buffer. Gd was quantified by ICP-MS.
4
with 1 × 10 cells and viability was evaluated after 1, 2, and 3 days
using PrestoBlue assay, which reflects the number of living cells
present on a surface at a given time point. At scheduled time points,
culture medium was removed and replaced by 1 mL of fresh medium
3
3+
C
dx.doi.org/10.1021/bm400978a | Biomacromolecules XXXX, XXX, XXX−XXX

Biomacromolecules
Article
Table 1. Properties of P(Pg-CL)s 3 and MRI-PCLs 4
P(Pg-CL)s
MRI-PCLs
Gd wt%
Gd wt%
single ligation
(calculated)
a
b
c
M
̅
SEC
cross-linking
(calculated)
Gd wt%
Gd mol%
n
FαPrεCL
(g/mol)
Đ
(experimental)
(experimental)
P(Pg-CL) 3
2
34000
25000
18000
2.3
1.9
1.7
P(DPTA[Gd]-CL) 41
2.4
5.4
1.2
2.7
4.3
1.0
2.6
3.6
0.8
2.1
3.0
2
1
2
P(Pg-CL) 3
5
P(DPTA[Gd]-CL) 42
5
2
5
P(Pg-CL)10 33
10
P(DPTA[Gd]-CL)10 43
8.6
a
b
c
SEC analyses with THF as solvent and PS standards. Determined by ICP-MS. Calculated from the weight percentages.
containing 10% of PrestoBlue. After 2 h of incubation at 37 °C, 200 μL
of supernatant were taken from each well and analyzed for
fluorescence at 530 nm (ex.) and 615 nm (em.) with a Victor X3
should be noted that complexation conditions vary in literature
depending on the chelate. Classically, temperature and reaction
time are used to optimize complexation. In this work, higher
temperatures could not be used to guarantee the conservation
of the thermo-labile azide groups, and complexation time was
doubled without benefice. However, it is well-known that
commercial contrast agents like Magnevist, which chelating
agent is DTPA, contain non complexed ligands (ca. 2%) to
ensure the chelation of potentially released free Gd . One
may wonder the potential biological effect of the 30%
remaining free chelate, especially with regard to a potential
complexation of calcium in vivo. This point will be addressed in
the last paragraph regarding biocompatibility.
(
Perkin-Elmer).
RESULTS AND DISCUSSION
■
Synthesis of MRI-Visible Poly(ε-caprolactone)s (MRI-
3
+
PCLs). Preparation of Clickable Gd Complex 2. The present
work aims at providing MRI-PCLs with controlled content of
gadolinium by coupling of an azide-containing contrast agent to
a propargyl-functionalized PCL obtained by ROP of εCL and
αPgεCL. In this regard, the first step consisted in the
preparation of diazido functionalized diethylenetriaminepenta-
3+
34
Synthesis of MRI-Visible Poly(ε-caprolactone)s (MRI-PCLs)
acetic acid (diN -DTPA; Scheme 1). Our choice to use a
3
4
. In parallel to the synthesis of 2, propargylated PCLs (P(Pg-
bifunctional ligand was motivated by scale-up considerations
with regard to the final context of biomaterials where multigram
scale MRI-PCLs would be required. Indeed, although described
in the literature, the synthesis of monofunctional DOTA-,
DTTA-, or DTPA-based ligands is a multistep process that
CL)) 3 were prepared. This was achieved by the copoly-
merization of αPgεCL and εCL, as already described by our
25
group. ROP allowed generating a family of copolymers with a
controlled ratio of propargylated units whose properties are
summarized in Table 1. Quantitative incorporation of the
propargylated units was obtained. Polymers were characterized
by NMR, SEC, and FT-IR. Increased intensities of the signals
characteristic of the propargyl group at 1.96 ppm on the NMR
spectra and at 3300 cm− on the FT-IR spectra were observed
30−33
requires careful purifications.
Commercially monofunc-
tional ligands are also available but, due to the mentioned
synthetic problems encountered, they still are specialty
chemicals whose expensive costs are limiting their use to
applications where very low functionalizations are required, like
1
19
(
Figures S3−S5). Our aim being to study the influence of the
for example in the case of polymer monosubstitution. We
contrast agent grafting density on the polymer visualization,
three copolymers were prepared with 2, 5, and 10% of
thus preferred to base our approach on a straightforward and
cost-effective preparation of diN -DTPA. As will be discussed
3
propargyl units (3 , 3 , and 3 , respectively). We limited our
in In Vitro Stability of MRI-PCLs, although difunctional ligands
possess less coordinating groups, the use of such a ligand was
not detrimental to the overall complex stability. A similar
1
2
3
study to low ratios because our previous work on MRI-visible
PCLs proved the efficient visualization of polymers incorporat-
6
27
ing low amounts of contrast agents.
approach was already used by Perez-Baena et al. who
prepared dipropargyl-DTPA for cross-linking reactions of
polyacrylic derivatives. In the present work, compound 1 was
readily synthesized from commercially available DTPA
dianhydride and an excess of 1-azido-3-aminopropane, by an
amidification reaction in almost quantitative yield. No
monofunctional derivative was produced under the chosen
I
MRI-PCLs 4 were finally obtained by a Cu -catalyzed [3 + 2]
cycloaddition between the azide groups of 2 and the alkyne
groups of 3. Typically, a 3 equiv excess of 2 relative to the
alkyne moles present in copolymers 3 was used and reaction
was performed in a dilute regime to avoid nondesired
intermolecular cross-linking and to favor the reaction of 2
with a single propargyl group (Scheme 2). Under these
conditions, complex 2 was efficiently coupled to P(Pg-CL)
copolymers 3 via 1,2,3-triazole groups without precipitation of
the resulting polymer that remained soluble in the reaction
medium. Compounds 4 were first characterized by NMR and
FT-IR. FT-IR spectra showed bands at 1690 cm− correspond-
ing to the carboxylate groups engaged in the complexation of
+
conditions as proved by the single peak at 558.4 Da [M + H ]
obtained in LC-MS analysis (data not shown) in accordance
with the theoretical value of 558.3 Da. The formation of the
1
desired diazide was further corroborated by H NMR and FT-
IR analyses (Figure S1). The subsequent complexation of 1
1
with GdCl ·6H O yielded chelate 2 in good yield. Purification
3
2
on Chelex 100 resin was used to efficiently remove the
3+
3+
3+
−1
noncomplexed Gd . The absence of free Gd was confirmed
by a methyl thymol blue (MTB) test (λABS max = 425 nm;
Figure S2). MALDI-TOF analysis showed three peaks
corresponding to residual free chelator 1 at 558.2 Da and the
protonated complex 2 under a monomeric form (713.2 Da)
and a dimeric form (1423.3 Da) resulting from shared
electrostatic interactions between two complexes 2. ICP-MS
Gd . Bands at 2100 cm corresponding to residual free azide
groups were also visible (Figure 1), thus, confirming that part of
ligands 2 reacted with a single propargyl group (Scheme 2a).
1
H NMR experiments showed the expected peaks of PCL with
a strong perturbation of the proton signal that increased with
the content of immobilized contrast agent on the polymer
backbone (Figure 2). As expected, peak broadening was
observed as a consequence of the perturbation induced by
3
+
analysis showed a 70% complexation ratio of Gd with 1. It
D
dx.doi.org/10.1021/bm400978a | Biomacromolecules XXXX, XXX, XXX−XXX

Biomacromolecules
Article
3
+
Scheme 2. Possible Structures of MRI-PCLs 4 as a Function
of the Type of Ligation
Gd , and no quantification could be done by NMR. A MTB
a
3+
test was performed after decomplexation of Gd by treatment
of MRI-PCLs in nitric acid to obtain a qualitative visualization
of the amount of Gd3+ present in the macromolecular contrast
agents (Figure S6). To quantitatively characterize the MRI-
PCLs, ICP-MS analyses were performed. Results are listed in
Table 1 and can be used to determine the type of ligation, that
is, single ligation, intra-, or intermolecular cross-linking. Indeed,
two ideal cases can be considered with either only the single
ligation occurring (Scheme 2a) or only double ligations, that is,
intra- or intermolecular cross-linking taking place (Scheme
2
b,c). The expected weight percentages of Gd3+ should be
twice higher in the first case compared to the second one.
Calculated values of these two ideal scenarios are provided in
Table 1. Results show a good agreement between the
experimental amounts of Gd3+ found and the values calculated
when considering only intra- or intermolecular cross-linking.
Starting from copolymers 3 , 3 , and 3 , containing 2, 5, and
a
(
a) Single ligation, (b) intramolecular crosslinking, and (c)
1
2
3
intermolecular crosslinking (in brown: PCL backbone, in blue: ligand
, in green: free azide groups).
1
0% of alkyne groups, respectively, the weight percentages of
2
complexed gadolinium obtained were 1.0, 2.6, and 3.6 wt %, to
be compared with theoretical values of 1.2, 2.7, and 4.3 wt %,
respectively. These results, combined to the free azido groups
shown by FT-IR analysis and the maintained solubility of the
copolymers 4, tend to demonstrate that the intramolecular
cross-linking scenario shown in Scheme 2b is predominant in
our system with only a limited amount of 2 reacting via a single
ligation. The concordance between the calculated and the
experimental values also proves the soundness of the chosen
strategy based on the use of complex rather than on the
3+
postcomplexation of Gd with a PCL macroligand. Indeed, this
strategy enabled the efficient coupling of contrast agents on the
degradable PCL backbones without decomplexation of
gadolinium from 2 during the coupling reaction, which yields
3
+
MRI-visible PLCs with predetermined amounts of Gd .
MR Characterizations. MRI-PCLS T1 Relaxation Time
Measurements. The effect of grafted gadolinium was measured
by determining the T relaxation time for hydrogen atoms on
1
the MRI-PCLs backbone. As expected, the longitudinal
relaxation time T of hydrogen atoms on MRI-PCLs was
1
significantly lower than in genuine PCL (Table 2). As observed
Figure 1. FT-IR spectra of MRI-PCLs copolymers 41−3. The plain line
−1
highlights the band at 1690 cm corresponding to the carboxylate
Table 2. Longitudinal Relaxation Times T of PCL and
3+
1
groups engaged in the complexation of Gd .
a
Compounds 3 and 4
2
1−3
T1 relaxation time (ms)
proton
δ (ppm)
PCL
32
41
42
43
A
B
4.0
2.3
1.3
1.4
1554
1587
1514
1512
1558
1694
1505
1502
536
562
554
552
273
289
291
299
252
346
196
289
C
D
a
4
00 MHz, DMSO-d , 80 °C, 11 mg/mL.
6
6
3+
in our previous study, T decreased with Gd substitution
1
ratio, and all protons of the MRI-PCLs were similarly impacted.
At this point, one can note that, although not obtained by the
same chemical strategies, T relaxation times were comparable
1
6
to the ones obtained by Blanquer et al. Typically, for the
reported PCL containing 0.8 wt % of Gd3+ a relaxation time of
about 600 ms was found. This matches the 550 ms found for 4
1
3+
1
containing 1.0 wt % of Gd . Advantageously, the chemical
Figure 2. H NMR (DMSO-d , 300 MHz) spectra of PCL, P(Pg-CL)
32
6
3
+
, and MRI-PCLs copolymers 41−3 (marks correspond to DMSO-d6
strategy reported here allows the targeting of defined Gd
at 2.5 ppm and residual water at 3.3 ppm).
contents. As a consequence, it is possible to precisely evaluate
the relative decrease of T relaxation time for the protons of
1
E
dx.doi.org/10.1021/bm400978a | Biomacromolecules XXXX, XXX, XXX−XXX

Biomacromolecules
Article
2
compounds 4 compared to genuine PCL as a function of their
gadolinium content. This is shown in Figure 3. To clarify the
this purpose, circular films (10 mg, 0.8 cm ) of mixed PCL and
3
+
MRI-PLCs (4 , 4 , and 4 ) were prepared with increasing Gd
1 2 3
3
+
concentrations in the range 2−40 μg of Gd per film. T signal
1
enhancement was compared to a PCL control film. One should
note that films morphologies were not uniform, as a result of
the film formation process, and PCLs mixtures presented
different filming-properties. This nonuniformity may have an
3+
effect on T enhancement as Gd -based T enhancement is
1
1
mainly dependent on the inner sphere relaxation phenomenon
3
+
which implies water molecules exchange at the Gd site.
However, in the present study this effect, if existing, was not
preponderant as shown by the following results. First, it is
3+
noticeable that all Gd -containing films presented a T signal
1
enhancement, whereas PCL films did not (Figures 4 and S7).
At constant Gd3+ grafting density (rows in Figure 4), the signal
3
+
enhancement increased with the overall content of Gd for all
copolymers, with an evident and clear signal observed above 2.5
μg of Gd3+ per mg of PCL. Indeed, irrespective of the grafting
density, the signal enhancement was weak for lower doses, as
Figure 3. Influence of the grafting density of gadolinium on the T1
relaxation time of MRI-PCLs protons (T ′ and T are the T relaxation
1
1
1
times of compounds 4 and genuine PCL, respectively).
3
+
seen for films 1−9 and strongly enhanced above 2.5 μg Gd
discussion, grafting densities, that is, molar percentages of CL
per mg of PCL as seen for films 10−15. Second, assessment of
3
+
units substituted by Gd , have been calculated based on the
grafting density effect on MR conspicuity was done by
3
+
3+
experimental values of Gd weight percentages (Table 1) and
used in Figure 3. Two main observations can be drawn from
considering a constant overall Gd content (columns in
3
+
Figures 4 and S7). Two behaviors were found. For low Gd
contents, below 2.5 μg of Gd3+ per mg of PCL, signal
enhancement was improved with the copolymers 4 and 4 that
presented high grafting densities (film 4 vs films 5 and 6 and
this nonlinear evolution. First, it is noticeable that T relaxation
1
times are strongly influenced even when low grafting densities
2
3
of Gd3 are found on the polymers backbone, as shown by the
+
3
+
7
%
0% decrease of T in copolymer 4 that contains only 0.8 mol
of Gd . Second, it is clearly visible that a limit is rapidly
film 7 vs films 8 and 9). This was not the case at higher Gd
1
1
3+
3+
contents, above 2.5 μg of Gd per mg of PCL, with even a
decrease of T signal enhancement for the copolymers 4 and
4 compared to 4 , as clearly shown by comparison between
reached with almost no further decrease of T above 2.1 mol %
1
2
1
3+
of Gd . These results are of importance in the frame of MRI
applications as they show that limited amounts of Gd3+ may be
used, which is of benefice when considering the toxicity of free
Gd .
MR-Imaging of MRI-PCLs. Having in hand MRI-PCLs with
3
1
film 13 and films 14 and 15. This signal decrease is likely due to
3
+
susceptibility effects resulting from high local Gd concen-
3+
35
trations. These combined observations tend to demonstrate
that a compromise between the two parameters, namely, the
overall Gd3+ content and the Gd grafting density on the
polymer backbone, seems to be necessary for optimal MR
3+
defined contents of gadolinium, we were interested in
evaluating the influence of (i) the overall amount of gadolinium
present in the samples and (ii) the grafting density of Gd3+
complexes immobilized on the PCL backbone on the MR
visualization. The final aim was to evaluate the optimum
visualization. This corroborates the evolution of T relaxation
1
times that showed that above 2.1 mol % no further T decrease
1
was observed. It also confirms the fact that a fine control of
3+
3+
compromise between MRI-visibility and low dose of Gd . For
Gd content and grafting density is of first importance for
Figure 4. MR visualization of PCL/MRI-PCLs 4 films as a function of the total amount of Gd3+ per film and of the Gd3+ grafting density on the
MRI-PCLs backbones.
F
dx.doi.org/10.1021/bm400978a | Biomacromolecules XXXX, XXX, XXX−XXX

Biomacromolecules
Article
optimal MR visualization. In vivo implantations should now be
planned to confirm our approach and thoroughly evaluate the
benefit of MRI-visibility of MRI-PCLs.
dissociation. However, and to confirm this trend, future in vitro
release study should be undertaken in more complex mediums
like human serum or PBS completed with physiologically
Stability and Cytocompatibility of MRI-PCLs. In Vitro
Stability of MRI-PCLs. Our aim being to provide long-term
visualization through the use of hydrophobic and slow
degrading copolyesters, we evaluated the in vitro stability of
relevant concentrations of ZnCl (0.03 wt %) and albumin (4
2
wt %) to evaluate the influence of proteins and cations on the
3
8,39
MRI-PCLs complex stability.
One should also consider
future in vivo implantation to evaluate the possible influence of
enzymes, macrophages, and transient changes in pH (especially
pH changes during inflammation) that may appear in vivo.
3
+
compounds 4 to detect any release of free Gd , which has been
associated with nephrogenic systemic fibrosis disorder
3
6,37
3
+
(NFS).
Indeed, for long-term visualization and to avoid
These results should be compared to the Gd decom-
3
+
adverse effects in the body, it was important to determine
whether any gadolinium salts were released from the MRI-
PCLs. With regards to the removal of carboxylate groups, three
left against five in Magnevist, one can postulate that the stability
of the complex could be decreased with a higher dissociation
plexation after clinical injection of commercial Gd -based
3+
contrast agents. Indeed, when dealing with Gd -based contrast
agents, the question usually arises of associated NFS in patients
with renal failure. However, it is noteworthy that this
phenomenon is usually observed after a conventional MR
3+
constant (K ) and, thus, a material that is less stable than an
d
scan where patients receive 1.0−2.5 g Gd as a bolus
unmodified small molecule chelate. This was not the case for a
(
posology of 0.1−0.2 mmol/kg of contrast agent for an
3+
40−42
DTPA/Gd -PCL recently synthesized by our group, which
intravenous injection of Magnevist).
In vitro studies
6
,7
had four carboxylate groups. Given that compounds 4 are not
water-soluble, whereas thermodynamic and kinetic constants
are measured in aqueous solution, the stability of compounds 4
performed in native serum at pH 7.4 and 37 °C showed an
4
3
initial decomplexation rate of 0.16% per day, which
3+
corresponds to 1.6−4 mg of circulating free Gd before
3
+
was studied by titrating by ICP-MS the Gd released in
aqueous medium. Results are shown in Figure 5. Release
renal clearance. In the present work, a clear T enhancement
1
2
3+
was observed for films loaded with 13 μg/cm of Gd . To
compare with our previous results, if used to coat a classical 10
7
×
10 cm implanted mesh for soft tissue prolapses, this would
3+
correspond to an overall dose of 2 mg of Gd . One can note
3
+
the similarity between this overall dose of immobilized Gd
and the extent of circulating free Gd3+ estimated for a clinical
MR scan. Additionally, it is worth noting that, for a maximum
release of 0.4%, as obtained in this work after 90 days, the
3+
maximal amount of free Gd released by such a mesh would be
μg, which represents less than 0.5% of the circulating free
8
3+
Gd during a conventional scan.
In Vitro Cytocompatibility of MRI-PCLs. An evaluation of
murine fibroblast proliferation was performed on polymer films
to assess the cytocompatibility of MRI-PCLs and determine
whether they are suitable for cell culture and potential cell-
Figure 5. Stability study performed on mixed films of PCL and 42
contacting biomedical applications. Films containing 0.1 wt %
having high dose (HD, 0.4 wt %) and low dose (LD, 0.1 wt %) of Gd3
+
3+
of Gd were prepared from compound 4 for this evaluation.
2
(
PBS; 37 °C).
Figure 6 shows the proliferation at 1, 2, and 3 days of L929
murine fibroblasts on MRI-PCL films compared to native PCL
and TCPS. It may be concluded from this proliferation
experiment that the surface of MRI-PCL did not impede
fibroblast proliferation. In addition, this result tends to
kinetics were run in PBS at 37 °C and over a 90 day period.
3
+
The amount of released Gd after 90 days was low whatever
the initial content of gadolinium in the films. It was limited to a
3
+
maximum concentration of released Gd of 0.45 and 13 μg/L
for LD and HD films, respectively (0.065 μg/L for PBS
negative control). The corresponding cumulative percentages
3+
of released Gd were therefore of 0.05% for the film containing
3
+
1
0 μg of Gd (LD film) and 0.4% for the film containing 40 μg
3+
of Gd (HD film). In addition, release profile in PBS showed a
limited and steady release. Taking into account the linear
portion of the curves, it was possible to extrapolate the period
3
+
required to release 1% of the total Gd amount under the used
conditions. One year for HD films and six years for LD films
were estimated. This result shows that, under the used
conditions, the measured release is indeed not significant and
that the PCL-based MMCAs can be considered as stable.
Comprehensive explanation of the parameters explaining the
observed stability despite a decreased number of a carboxylate
available for Gd3 complexation is beyond the scope of this
work but the following could be considered: (i) ester groups in
the polymer may possibly offset the loss of these carboxylate+
groups or (ii) perhaps the entropy of the system limits the Gd3
+
Figure 6. L929 proliferation on MRI-PCL films (4 , red) compared to
2
TCPS (positive control, gray) and PCL films (black); data are
expressed as mean ± SD and correspond to measurements in
triplicate.
G
dx.doi.org/10.1021/bm400978a | Biomacromolecules XXXX, XXX, XXX−XXX

Biomacromolecules
Article
demonstrate that the presence of free carboxylate on the
polymer backbone was not dentrimental to cell proliferation
although some of the free calcium present in the culture
medium may be complexed. Indeed, this is not surprising
considering the very low effective concentration of free ligand
on the MRI-PCL. Typically, if one considers the amount of
MRI-PCL used to coat a classical 10 × 10 cm implanted mesh
for soft tissue prolapses, it represents a negligible 6 μmol of
potential free ligand to be compared with the physiological
concentration of calcium between 2.2 and 2.6 mM. Taking into
account that PCL is widely recognized as a biocompatible
material, MRI-PCLs would appear to be suitable for the growth
of fibroblasts and cell-contacting applications.
REFERENCES
■
(
2
1) Gallez, B.; Baudelet, C.; Jordan, B. F. NMR Biomed. 2004, 17 (5),
40−262.
2) Fischer, T.; Ladurner, R.; Gangkofer, A.; Mussack, T.; Reiser, M.;
(
Lienemann, A. Eur. Radiol. 2007, 17 (12), 3123−3129.
(3) Lapray, J. F.; Costa, P.; Delmas, V.; Haab, F. Prog. Urol. 2009, 19
(13), 953−969.
(4) Birch, C. Best Pract. Res. Clin. Obstet. Gynaecol. 2005, 19 (6),
9
(
4
(
79−991.
5) Nottelet, B.; Coudane, J.; Vert, M. Biomaterials 2006, 27 (28),
948−4954.
6) Blanquer, S.; Guillaume, O.; Letouzey, V.; Lemaire, L.; Franconi,
F.; Paniagua, C.; Coudane, J.; Garric, X. Acta Biomater. 2012, 8 (3),
339−1347.
7) Guillaume, O.; Blanquer, S.; Letouzey, V.; Cornille, A.;
Huberlant, S.; Lemaire, L.; Franconi, F.; de Tayrac, R.; Coudane, J.;
Garric, X. Macromol. Biosci. 2012, 12 (10), 1364−1374.
8) Aime, S.; Castelli, D. D.; Crich, S. G.; Gianolio, E.; Terreno, E.
1
(
CONCLUSION
■
In this work, we presented the efficient convergent synthesis of
novel hydrophobic macromolecular agents based on poly(ε-
caprolactone). PCLs with controlled propargyl group ratios
yield MRI-PCLs with defined Gd grafting densities. H NMR,
FT-IR, and ICP-MS analyses confirmed the efficient coupling
of the complex on the PCL backbones via a main mechanism of
intramolecular cross-linking that led to MRI-PCLs containing
(
Acc. Chem. Res. 2009, 42 (7), 822−831.
3
+
1
(9) Beilvert, A.; Cormode, D. P.; Chaubet, F.; Briley-Saebo, K. C.;
Mani, V.; Mulder, W. J. M.; Vucic, E.; Toussaint, J. F.; Letourneur, D.;
Fayad, Z. A. Magn. Reson. Med. 2009, 62 (5), 1195−1201.
(
10) Ladd, D. L.; Hollister, R.; Peng, X.; Wei, D.; Wu, G.; Delecki,
D.; Snow, R. A.; Toner, J. L.; Kellar, K.; Eck, J.; Desai, V. C.; Raymond,
G.; Kinter, L. B.; Desser, T. S.; Rubin, D. L. Bioconjugate Chem. 1999,
10 (3), 361−370.
3
+
1
.0, 2.6, and 3.6 wt % of Gd . T relaxation times were highly
1
impacted with a 70% reduction for grafting densities as low as
.8 mol %. T relaxation time measurements also showed that
0
(11) Lee, H. Y.; Jee, H. W.; Seo, S. M.; Kwak, B. K.; Khang, G.; Cho,
S. H. Bioconjugate Chem. 2006, 17 (3), 700−706.
1
low grafting ratios were sufficient to obtain MRI-visible
materials. No further influence is detectable for substitution
ratios above 2.1 mol %. This was confirmed by the MR
(
2
12) Lucas, R. L.; Benjamin, M.; Reineke, T. M. Bioconjugate Chem.
008, 19 (1), 24−27.
(13) Shiraishi, K.; Kawano, K.; Minowa, T.; Maitani, Y.; Yokoyama,
3
+
experiments with Gd -containing films. All of them presented
M. J. Controlled Release 2009, 136 (1), 14−20.
a T signal enhancement and best results were obtained for
1
3
+
(14) Toth, E.; Helm, L.; Kellar, K. E.; Merbach, A. E. Chem.−Eur. J.
films containing more than 2 μg of Gd per mg of PCL. MR
experiments also demonstrated the role of Gd grafting
1
(
999, 5 (4), 1202−1211.
3
+
15) Vaccaro, M.; Accardo, A.; Tesauro, D.; Mangiapia, G.; Lof, D.;
Schillen, K.; Soderman, O.; Morelli, G.; Paduano, L. Langmuir 2006,
2 (15), 6635−6643.
16) Zarabi, B.; Nan, A. J.; Zhuo, J. C.; Gullapalli, R.; Ghandehari, H.
density. A fine control and a compromise between the overall
3+
3+
Gd content and the Gd grafting density on the polymer
backbone is necessary for optimal MR visualization. These
novel MMCAs have been found to be stable over a 90 day
2
(
Macromol. Biosci. 2008, 8 (8), 741−748.
3
+
period with less than 0.5% of Gd released in PBS. Finally,
they also were found to be cytocompatible, with good
proliferation of fibroblasts, which confirms the high potential
of these novel hydrophobic PCL-based MMCAs for biomedical
applications.
(17) Zong, Y.; Guo, J.; Ke, T.; Mohs, A. M.; Parker, D. L.; Lu, Z. R. J.
Controlled Release 2006, 112 (3), 350−356.
(
18) Fu, Y. J.; Raatschen, H. J.; Nitecki, D. E.; Wendland, M. F.;
Novikov, V.; Fournier, L. S.; Cyran, C.; Rogut, V.; Shames, D. M.;
Brasch, R. C. Biomacromolecules 2007, 8 (5), 1519−1529.
19) Grogna, M.; Cloots, R.; Luxen, A.; Jerome, C.; Passirani, C.;
(
Lautram, N.; Desreux, J. F.; Detrembleur, C. Polym. Chem. 2010, 1 (9),
1485−1490.
ASSOCIATED CONTENT
■
*
S
Supporting Information
H NMR spectrum of 1, MTB test of 2 and 4, H NMR, FT-IR,
and SEC analyses of 3, and MR visualization of 4 with higher
contrast. This material is available free of charge via the Internet
at http://pubs.acs.org.
(20) Pressly, E. D.; Rossin, R.; Hagooly, A.; Fukukawa, K. I.;
Messmore, B. W.; Welch, M. J.; Wooley, K. L.; Lamm, M. S.; Hule, R.
A.; Pochan, D. J.; Hawker, C. J. Biomacromolecules 2007, 8 (10),
1
1
3
(
126−3134.
21) Shokeen, M.; Pressly, E. D.; Hagooly, A.; Zheleznyak, A.;
Ramos, N.; Fiamengo, A. L.; Welch, M. J.; Hawker, C. J.; Anderson, C.
J. ACS Nano 2011, 5 (2), 738−747.
AUTHOR INFORMATION
■
*
(22) Zhang, G. D.; Zhang, R.; Wen, X. X.; Li, L.; Li, C.
Biomacromolecules 2008, 9 (1), 36−42.
Corresponding Author
E-mail: benjamin.nottelet@univ-montp1.fr.
(23) Franconi, F.; Roux, J.; Garric, X.; Lemaire, L. Magn. Reson. Med.
2
013, DOI: 10.1002/mrm.24666.
24) Villaraza, A. J. L.; Bumb, A.; Brechbiel, M. W. Chem. Rev. 2010,
110 (5), 2921−2959.
25) Darcos, V.; El Habnouni, S.; Nottelet, B.; El Ghzaoui, A.;
Coudane, J. Polym. Chem. 2010, 1 (3), 280−282.
26) Carboni, B.; Benalil, A.; Vaultier, M. J. Org. Chem. 1993, 58
14), 3736−3741.
27) Perez-Baena, I.; Loinaz, I.; Padro, D.; Garcia, I.; Grande, H. J.;
Notes
(
The authors declare no competing financial interest.
(
ACKNOWLEDGMENTS
■
(
The authors wish to thank the French Ministry of Education
and Research for S.E.H.’s fellowship, the Erasmus program for
B.P.’s fellowship, Chantal Douchet and Olivier Bruguier for
ICP-MS analyses and Sylvie Hunger and Cedric Paniagua for
NMR analyses.
(
(
Odriozola, I. J. Mater. Chem. 2010, 20 (33), 6916−6922.
(28) Brown, R.; Clarke, D. W.; Daffner, R. H. Am. J. Roentgenol. 2000,
175, 1087−1090.
H
dx.doi.org/10.1021/bm400978a | Biomacromolecules XXXX, XXX, XXX−XXX

Biomacromolecules
Article
(
29) Vonarbourg, A.; Sapin, A.; Lemaire, L.; Franconi, F.; Menei, P.;
Jallet, P.; Le Jeune, J. J. Magn. Reson. Mater. Phys., Biol. Med. 2004, 17
3−6), 133−139.
30) Anelli, P. L.; Fedeli, F.; Gazzotti, O.; Lattuada, L.; Lux, G.;
Rebasti, F. Bioconjugate Chem. 1999, 10 (1), 137−140.
31) Bryson, J. M.; Chu, W. J.; Lee, J. H.; Reineke, T. M. Bioconjugate
Chem. 2008, 19 (8), 1505−1509.
32) Fernandez-Trillo, F.; Pacheco-Torres, J.; Correa, J.; Ballesteros,
(
(
(
(
P.; Lopez-Larrubia, P.; Cerdan, S.; Riguera, R.; Fernandez-Megia, E.
Biomacromolecules 2011, 12 (8), 2902−2907.
(
33) Parac-Vogt, T. N.; Kimpe, K.; Laurent, S.; Pierart, C.; Elst, L. V.;
Muller, R. N.; Binnemans, K. Eur. J. Inorg. Chem. 2004, 17, 3538−
543.
34) Product Monograph: MAGNEVIST Gadopentetate Dimeglumine
Injection; Bayer Inc.: Toronto, Ontario, 2012; submission control No.
56553, pp 1−38.
35) Villringer, A.; Rosen, B. R.; Belliveau, J. W.; Ackerman, J. L.;
3
(
1
(
Lauffer, R. B.; Buxton, R. B.; Chao, Y. S.; Wedeen, V. J.; Brady, T. J.
Magn. Reson. Med. 1988, 6 (2), 164−174.
(
36) Goulle, J. P.; Saussereau, E.; Mahieu, L.; Bouige, D.; Groenwont,
S.; Guerbet, M.; Lacroix, C. J. Anal. Toxicol. 2009, 33 (2), 92−98.
(
37) Thakral, C.; Abraham, J. L. Radiol. Clin. 2009, 47 (5), 841−853.
(
38) Baranyai, Z.; Palinkas, Z.; Uggeri, F.; Brucher, E. Eur. J. Inorg.
Chem. 2010, 13, 1948−1956.
39) Laurent, S.; Elst, L. V.; Henrotte, V.; Muller, R. N. Chem.
(
Biodiversity 2010, 7 (12), 2846−2855.
(
(
(
40) VIDAL: le Dictionnaire, 88th ed.; Vidal: Paris, 2012.
41) Grobner, T. Nephrol. Dial. Transpl. 2006, 21 (4), 1104−1108.
42) Idee, J. M.; Port, M.; Raynal, I.; Schaefer, M.; Le Greneur, S.;
Corot, C. Fund. Clin. Pharmacol. 2006, 20 (6), 563−576.
43) Frenzel, T.; Lengsfeld, P.; Schirmer, H.; Hutter, J.; Weinmann,
H. J. Invest. Radiol. 2008, 43 (12), 817−828.
(
I
dx.doi.org/10.1021/bm400978a | Biomacromolecules XXXX, XXX, XXX−XXX