Synthesis of an Amphiphilic Bis-Aqua Gd(OBETA) Complex for the Preparation of High-Relaxivity Supramolecular Magnetic Resonance Imaging Probes

Article abstract of DOI:10.1002/cplu.201500366

Prompted by the favourable relaxometric, thermodynamic and kinetic properties of the bis-hydrated Gd(OBETA) (OBETA=2,2′-oxybis(ethylamine)-N,N,N′,N′-tetraacetic acid) complex, a novel derivative tailored with an n-hexadecyl chain was synthesised. The amphiphilic gadolinium complex was designed and prepared with the aim of obtaining high relaxivity supramolecular aggregates by self-assembly in micelles and liposomes. Thus, lipidic nanoparticles were prepared and characterised by dynamic light scattering and ¹H NMR relaxometry. Relaxivity values of up to 48.3 mm^-1 s^-1 (20 MHz and 298 K) were registered in liposomal aggregates. The binding to human serum albumin (HSA), evaluated both in terms of affinity and relaxometric properties of the supramolecular adduct, yielded exceptionally high relaxivity values (71.4 mm^-1 s^-1 at 30 MHz and 298 K).

Full text of DOI:10.1002/cplu.201500366

DOI: 10.1002/cplu.201500366
Full Papers
Synthesis of an Amphiphilic Bis-Aqua Gd(OBETA) Complex
for the Preparation of High-Relaxivity Supramolecular
Magnetic Resonance Imaging Probes**
Roberto Negri,[a] Fabio Carniato,[b] Mauro Botta,[b] Giovanni B. Giovenzana,*[a] and
Lorenzo Tei*[b]
Prompted by the favourable relaxometric, thermodynamic and
kinetic properties of the bis-hydrated Gd(OBETA) (OBETA=2,2’-
oxybis(ethylamine)-N,N,N’,N’-tetraacetic acid) complex, a novel
derivative tailored with an n-hexadecyl chain was synthesised.
The amphiphilic gadolinium complex was designed and pre-
pared with the aim of obtaining high relaxivity supramolecular
aggregates by self-assembly in micelles and liposomes. Thus,
lipidic nanoparticles were prepared and characterised by dy-
namic light scattering and 1H NMR relaxometry. Relaxivity
values of up to 48.3 mmÀ1 sÀ1 (20 MHz and 298 K) were regis-
tered in liposomal aggregates. The binding to human serum
albumin (HSA), evaluated both in terms of affinity and relaxo-
metric properties of the supramolecular adduct, yielded excep-
tionally high relaxivity values (71.4 mmÀ1 sÀ1 at 30 MHz and
298 K).
Introduction
Contrast-enhanced magnetic resonance imaging (MRI) is a pow-
erful diagnostic technique of widespread use in clinics and bio-
medical research. Contrast agents (CAs) allow a significant re-
duction in the relaxation times (T1 and T2) of water protons in
the tissues where they are distributed, thus increasing the
signal intensity and improving the diagnostic content of the
MR images.[1] The T1 agents are typically paramagnetic metal
complexes (GdIII and MnII) based on multidentate polyamino-
polycarboxylic acid ligands, able to tightly encapsulate the
metal ion in their coordination cage. Thereby, to avoid poten-
tial toxicity risks associated with long-term dissociation of the
metal ion from the corresponding complex, research efforts
have been devoted to enhance both the thermodynamic sta-
bility and kinetic inertness of the chelates.[2]
This is normally achieved by the use of octadentate ligands,
which form GdIII complexes with one coordinated water mole-
cule (q=1) as the archetypal and clinically approved Gd(DOTA)
(1; DOTA=1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic
acid)[3] and Gd(DTPA) (2; DTPA=diethylene triamine pentaace-
tic acid).[4] However, the search for ligands capable of forming
q=2 GdIII complexes without compromising thermodynamic
[a] Dr. R. Negri, Prof. G. B. Giovenzana
and kinetic stabilities has led to several examples of heptaden-
tate chelators able to form stable GdIII complexes. Among
them, ligands 3–7, namely, 6-amino-6-methylperhydro-1,4-dia-
zepine tetraacetic acid (AAZTA, 3a),[5] the tripodal hydroxypyri-
dinone (HOPO)-based ligands 4,[6] the pyridine-based macrocy-
clic 3,6,9,15-tetraazabicyclo[9.3.1]pentadeca-1(15),11,13-triene-
3,6,9-triacetic acid (PCTA, 5a),[7] the linear analogue 2,6-pyridi-
nebis(methylamino)-N,N,N’,N’-tetraacetic acid (PyMTA, 6a),[8]
and the picolinate derivative trans-1,2-diaminocyclohexane-N-
[(6-methylpicol-2-yl)methyl]-N,N’,N’-triacetic acid (CyPic3A, 7)[9]
appeared to be promising candidates.
Dipartimento di Scienze del Farmaco
Università degli Studi del Piemonte Orientale “A. Avogadro”
L.go Donegani 2/3, 28100 Novara (Italy)
E-mail: giovannibattista.giovenzana@uniupo.it
[b] Dr. F. Carniato, Prof. M. Botta, Prof. L. Tei
Dipartimento di Scienze e Innovazione Tecnologica
Università degli Studi del Piemonte Orientale “A. Avogadro”
V.le T. Michel 11, 15121 Alessandria (Italy)
E-mail: lorenzo.tei@uniupo.it
[**] OBETA=2,2’-oxybis(ethylamine)-N,N,N’,N’-tetraacetic acid
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cplu.201500366.
ChemPlusChem 2016, 81, 235 – 241
235
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
We recently reported a comprehensive thermodynamic, ki-
netic and relaxometric study of the paramagnetic complex
formed by GdIII with the acyclic ligand 2,2’-oxybis(ethylamine)-
N,N,N’,N’-tetraacetic acid (OBETA, 8),[10] along with a detailed
structural characterization. Bis-hydrated Gd(OBETA) shows
good thermodynamic stability and kinetic inertness towards
transmetallation reactions with endogenous ions. Moreover, in-
teresting relaxivity values were observed, which resulted from
the presence of two coordinated water molecules in a fast-ex-
change regime. The high exchange rate (kex =1/tM) of the
inner sphere (IS) water molecules of Gd(OBETA) allows high re-
laxivity values to be obtained when the complex reduces its
rotational tumbling rate (1/tR) by binding to large systems
such as nanoparticles (NPs) or proteins. In fact, it is well estab-
lished that rotational dynamics represents one of the main fac-
tors controlling the efficiency of low-molecular-weight GdIII
chelates.[11] Thus, the use of complexes that self-aggregate into
lipid NPs such as micelles (5–50 nm) and liposomes (50–
500 nm) or strongly interact with the most abundant blood
plasma protein, human serum albumin (HSA), is one of the
most adopted strategies to achieve significant relaxivity en-
hancements for GdIII chelates.[11,12]
The ongoing interest in Gd-based micelles, liposomes[13] and
other kinds of lipidic or polymeric NPs—such as solid lipid
nanoparticles (SLN),[14] polylactic-co-glycolic acid (PLGA)[15] NPs,
Gd-labelled HDL lipoproteins[16] or nanodiscs[17]—relies on their
extensive use as vehicles for molecular and cellular imaging of
disease-specific markers and for image-guided drug delivery in
several pathologies such as tumours, myocardial infarction,
atherosclerosis or other pathophysiological processes, such as
angiogenesis and inflammation.[13]
Scheme 1. Synthesis of HD-OBETA (12). a) Br2, SOCl2, then EtOH; b) K2CO3, KI,
CH3CN; c) ClCH2COOEt, N,N-diisopropylethylamine (DIPEA), CH3CN; d) NaOH,
then HCl.
with ethyl 2-bromostearate in acetonitrile in the presence of
potassium carbonate/potassium iodide led to monosubstituted
derivative 10. Exhaustive alkylation of the amine groups of 10
with ethyl 2-chloroacetate in acetonitrile and in the presence
of DIPEA led to intermediate 11. Finally, hydrolysis of the ethyl
esters under basic conditions gave the desired ligand HD-
OBETA, which was isolated as the corresponding hydrochloride
by treatment with hydrochloric acid (12).
1H NMR relaxometric characterization
Currently, nearly all the amphiphilic Gd complexes used to
intercalate into these lipophilic NPs, apart from rare excep-
tions,[14,18–20] are q=1 Gd chelates, the relaxometric properties
of which are not optimal for attaining high r1 values when con-
jugated to slowly tumbling substrates.[12]
NMR relaxometry is a powerful tool for obtaining detailed
structural and dynamic information on paramagnetic chelates
and on the supramolecular structures that the metal complex
can form through self-assembly or noncovalent binding to pro-
teins or other hosting substrates. The measurement and analy-
sis of the nuclear magnetic longitudinal and transverse relaxa-
tion rates (R1 and R2) of the water proton nuclei as a function
of the applied field strength, temperature and pH permit de-
tailed information to be obtained, which define the efficiency
of the complex as an MRI contrast agent. In particular, the
water proton relaxation enhancement ability of a GdIII complex
is expressed by the parameter called relaxivity, ri (i=1, 2),
which represents the increase of the water proton relaxation
rates normalised to a 1.0 mm solution of the paramagnetic
system at a given temperature and magnetic-field strength
(typically 298/310 K and 20 MHz).
The favourable relaxometric properties of Gd(OBETA) encour-
aged us to design a derivative endowed with the ability to
self-assemble into lipidic nanoparticles or to bind reversibly to
HSA. Thus, in this study, an OBETA-based ligand bearing
a linear hexadecylic chain (HD-OBETA, Scheme 1) and the cor-
responding GdIII complex were synthesised.
A complete
1H NMR relaxometric study was carried out on the micelles ob-
tained by self-aggregation of Gd(HD-OBETA), on liposomes
containing Gd(HD-OBETA) embedded in the lipid bilayer and
on the supramolecular adduct achieved by means of the non-
covalent interactions of Gd(HD-OBETA) with HSA.
The presence of a long C16 alkyl chain in Gd(HD-OBETA) fa-
vours the formation of aggregates (micelles) even at submilli-
molar concentrations and the interaction with hydrophobic
pockets of HSA. The longitudinal relaxivity of Gd(HD-OBETA) at
Results and Discussion
Synthesis
298 K and 0.47 T is 34.2 mmÀ1 À1. This value is higher than that
s
The synthesis of HD-OBETA (Scheme 1, 12) takes advantage of
the tailored alkylating agent ethyl 2-bromostearate (Scheme 1,
9), prepared by a-halogenation of stearic acid with bromine
and esterification of the resulting brominated derivative. The
alkylation of commercially available 2,2’-oxybis(ethylamine)
found for other micellar aggregates of bis-hydrated (q=2) GdIII
chelates such as Gd(AAZTAC17) (3b), measured under identical
experimental conditions (r1 =30.0 mmÀ1 À1).[20] It is worth
s
noting that only a few examples of amphiphilic q=2 GdIII che-
lates capable of self-aggregation in micelles or intercalation
ChemPlusChem 2016, 81, 235 – 241
www.chempluschem.org
236
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
into the membrane of liposomes are known. To the best of our
knowledge, in addition to the above cited Gd(AAZTAC17), just
two GdIII complexes based on PCTA and PyMTA derivatives
bearing a C12 chain (5b and 6b) with r1 value at 298 K of 29.2
(20 MHz)[21] and 30.7 mmÀ1 sÀ1 (6.95 MHz),[22] respectively, have
been reported.
dilute conditions. A dynamic light scattering (DLS) study at
concentrations above and below 0.12 mm was carried out. As
reported in Figure 1b, the DLS data show in both cases the for-
mation of micellar aggregates, but with different hydrodynam-
ic diameter, that is, 43.6 nm (polydispersity index, PDI=0.268)
for micelles at [Gd(HD-OBETA)]>0.12 mm and 18.0 nm (PDI=
0.276) for those below 0.12 mm. In the range 0.05–0.12 mm,
the observed proton relaxation rate shows a linear depend-
ence from the concentration; therefore, we can only infer that
the cmc value is lower than 0.05 mm. Nevertheless, cmc values
lower than 0.05 mm have been reported previously for com-
plexes functionalised with lipophilic chains of comparable
length.[23] On the base of these results, we hypothesised that
only in very diluted conditions (<0.12 mm) do Gd(HD-OBETA)
chelates assemble into spherical micelles, whereas at concen-
trations higher than 0.12 mm a higher degree of aggregation
occurs.
Generally, for an amphiphilic complex, the measure of R1 as
a function of concentration of the Gd complex affords a clear
inflection point that defines the critical micellar concentration
(cmc). As expected, the aliphatic chain of Gd(HD-OBETA) is re-
sponsible for a concentration-dependent effect on the relaxivi-
ty. However, for Gd(HD-OBETA), the plot of r1 (20 MHz; 298 K)
versus the concentration of the complex over the range 0.05–
0.6 mm shows an ill-defined variation from linearity at about
0.12 mm (Figure 1a), which allows two regions characterised by
different relaxivity to be discriminated. Below the inflection
point at 0.12 mm, r1 assumes the value of 25.0 mmÀ1 sÀ1
,
whereas above it the relaxivity is 34.0 mmÀ1 sÀ1. The r1 value at
low concentrations (<0.12 mm) is too high for a monomeric
complex and suggests some degree of aggregation even in
The above qualitative considerations find support through
the measurement and analysis of the magnetic-field depend-
ence of the relaxivity, the so-called nuclear magnetic relaxation
dispersion (NMRD) profile. The NMRD profiles of Gd(HD-OBETA)
at concentrations above and below 0.12 mm and at 298 and
310 K were recorded in the frequency range 0.01–70 MHz,
which correspond to magnetic-field strengths varying between
2.34”10À4 and 1.64 T (Figure 2). The profiles are characterised
Figure 2. NMRD profiles of Gd(HD-OBETA) micellar aggregates at concentra-
tions below 0.12 mm (open circles), above 0.12 mm (black circles), and of
Gd(HD-OBETA) incorporated in the liposomal membrane (open squares). The
solid lines represent the best results of the fitting to the experimental
points.
by different amplitudes but present a common general shape
with a broad hump centred at about 25–30 MHz, typical of
slowly tumbling systems. Only the NMRD data of the more di-
luted system were analyzed as the size of the micelles in this
solution is more similar to analogous micellar systems reported
in the literature.[12] Moreover, only the high-field region (>
3 MHz) of the NMRD profile was used for the fitting procedure
because of the known limitations of the Solomon–Bloember-
gen–Morgan (SBM) theory to properly account for the behav-
1
Figure 1. Top: plot of the water H longitudinal relaxation rate at 20 MHz
and 298 K as a function of the concentration of Gd(HD-OBETA). Bottom: DLS
measurement for micellar aggregates of Gd(HD-OBETA) at concentrations
above (black squares) and below (open circles) 0.12 mm.
ChemPlusChem 2016, 81, 235 – 241
www.chempluschem.org
237
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
iour of slowly tumbling systems at low magnetic-field 1.7), in agreement with the results obtained for Eu(OBETA) (Fig-
strengths.[2] Owing to the presence of a relatively fast local ro- ure S1 in the Supporting Information).[10]
tation of the complex about the main axis of the aliphatic
There is a large difference between the correlation times as-
chain superimposed onto the global motion of the micellar ag- sociated with the overall tumbling motion and the local rota-
gregate, the analysis of the NMRD profiles requires the incor- tion to indicate that the systems are highly flexible and their
poration of the model-free Lipari–Szabo approach into the relaxivity is severely limited by the relatively short value of tRL.
SBM equations for the IS relaxation mechanism.[24] This model The local rotational correlation time (tRL =0.12 ns) and the
allows the contribution of the overall global rotation of the order parameter S2 (0.28) for Gd(HD-OBETA) in a micellar form
paramagnetic micelle (tRG) to be separated from the contribu- are much lower than those reported for Gd(AAZTAC17), proba-
tion of a faster local motion (tRL) due to the local rotation of bly as a result of a higher internal rotation of the GdIII complex
the coordination cage around the pendant arm. The correla- around the aliphatic chain. The data for Gd(PCTA-C12) were
tion of the two types of motions is described by the parameter fitted without the use of the Lipari–Szabo approach and thus
S2 whose value may vary between zero (completely independ- are not directly comparable.[21] Interestingly, the relaxivity
ent motions) and one (totally correlated motions). A least- values at 310 K for Gd(HD-OBETA) are almost superimposable
squares fit of the data was performed by treating as variable to those at 298 K for both micelles in the diluted solution and
parameters tRG, tRL, S2 and the parameters characterizing the in the further aggregated form (see Figures S2–S5 in the Sup-
electron spin relaxation, that is, the correlation time for the porting Information).
modulation of the zero-field-splitting interaction (tV) and the
Long-circulating liposomes incorporating the amphiphilic
mean square zero-field-splitting energy (D2). For the residence Gd(HD-OBETA) complexes in the lipidic bilayer were prepared
lifetime (tM) we used as a starting value the one obtained for as described previously[18] using a formulation of membrane
Gd(OBETA), assuming that the introduction of a lipophilic components that allows high water permeability.[26] Hence, it
chain does not alter the water exchange rate to a large was possible to fit the NMRD profile of the paramagnetic lipo-
extent.[10] This value is best determined by performing varia- somes without taking into account differential contributions
ble-temperature 17O NMR spectroscopy measurements. Un- from the GdIII chelates pointing in- and outwards from the lipi-
fortunately, in the present case it was not possible to obtain dic membrane. DLS measurements showed the formation of
the necessary concentration required for this experiment. The vesicles of 150.6 nm average size (PDI=0.104). The relaxivity
value of liposomes incorporating Gd(HD-OBETA) at
298 K and 20 MHz is 48.3 mmÀ1 sÀ1, a value much
Table 1. Best-fit parameters obtained by analysis of the 1H NMRD profiles for Gd(HD-
OBETA) in the form of a micelle (diluted solution) and embedded in the membrane bi-
greater than those of corresponding liposomes em-
bedding q=1 GdIII complexes.[12] To our knowledge,
layer of a liposome compared to the related q=2 systems (Gd(AAZTAC17) and
there are only a few examples of paramagnetic lipo-
Gd(PCTA-C12)).[a]
somes that reached or overcame an r1 value of
Parameters
Gd[HD-OBETA] Gd[HD-OBETA] Gd[AAZTAC17][b] Gd[PCTA-C12][c]
40 mmÀ1 sÀ1 at 298 K and 20 MHz: those incorporat-
ing GdPyMTAOC12 (6b, r1 =43.9 mmÀ1 sÀ1 [26]
those embedding GdDOTA(GAC12)2
)
and
[micelle]
[liposome]
D2 [1019 sÀ2
t2V98 [ps]
]
1.7Æ0.1
36.0Æ0.5
80[d]
0.8Æ0.1
14.5Æ0.2
80[d]
1.43
43
67
2.54
0.29
3.10
0.40
2.3
43
100
0.47
–
3.10
–
(r1 =
40.0 mmÀ1 sÀ1), a GdIII complex suitably designed to
restrict the local rotational motion of the chelates in
the liposome bilayer.[18,27] The NMRD profiles of the
Gd(HD-OBETA) liposomes at 298 and 310 K were re-
corded in the frequency range 0.01–70 MHz and
fitted only in the high-field region (>3 MHz) using
the Lipari–Szabo free model for the description of
the rotational dynamics (Figure 2). The results, report-
ed in Table 1, show that the hexadecyl chain of
Gd(HD-OBETA) rotates about its axis more slowly
t2M98 [ns]
298
RG
298
t
t
[ns]
[ps]
20Æ5
50Æ5
0.12Æ0.02
0.35Æ0.02
RL
r [Š]
S2
3.00[e]
3.00[e]
0.28Æ0.02
0.40Æ0.01
[a] For the parameters q, a and 298D the values of 2, 4.0 Š, 2.24”10À5 cm2 sÀ1 were
used, respectively. [b] Ref. [20]. [c] Ref. [21]. [d] Ref. [10]. [e] Values were fixed during
the fitting.
best-fit parameters are listed in Table 1 and are compared to than in the case of micellar aggregates (S2 =0.40 and tRL =
other aggregated systems based on q=2 GdIII amphiphilic che- 0.35 ns) and this slower degree of internal rotation is the main
lates.
Moreover, to confirm that the formation of micellar aggre-
reason for the observed 93% increase in relaxivity.
gates does not influence the number of inner-sphere water
molecules (q), the hydration number of the analogue EuIII com-
Binding to HSA
plex ([Eu(HD-OBETA)]ꢀ0.1 mm) was determined by lumines- Molecules containing long carbon chains typically have a high
cence lifetime measurements following a method reported in affinity for HSA, the most abundant blood serum protein and
the literature.[25] The lifetime of the Eu(5D0) transition was mea- an interesting biological target for CAs, especially those aimed
sured in solutions in both D2O and H2O and the values were at acting as blood-pool agents. In fact, chelates bearing hydro-
used to calculate the q parameter. The luminescence lifetime phobic moieties can increase their blood retention time
data leads to a value of the hydration number close to 2 (q= through formation of supramolecular adducts with HSA, which
ChemPlusChem 2016, 81, 235 – 241
www.chempluschem.org
238
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
also extends the tumbling rate of the paramagnetic complex
by increasing their relaxivity, especially at low magnetic fields
(0.5–1.5 T, 20–60 MHz).[28]
ing an octadecyl chain[23] and Gd(AAZTAC17).[20] As shown by
the NMRD profile of the Gd(HD-OBETA)/HSA adduct (Fig-
ure 3b), recorded at 298 K and in the frequency range 0.01–
70 MHz, the relaxivity peaks at 30 MHz with a value of
71.4 mmÀ1 sÀ1 are close to that reported for the Gd(AAZTAC17)/
HSA conjugate (84 mmÀ1 sÀ1 for the adduct with defatted HSA).
The field dependency of the relaxivity is quite typical for
slowly tumbling macromolecular systems, with values of tRL
and S2 higher than those of micelles and liposomes (S2 =0.53
and tRL =0.76 ns), which indicates a restriction of the internal
rotation of the Gd(HD-OBETA) chelates around the alkyl chain
interacting with the protein. Finally, as already noted for the
lipid NPs discussed earlier, rb1 shows only a limited variation
with temperature (Figure S7 in the Supporting Information).
The noncovalent hydrophobic interaction of Gd(HD-OBETA)
with HSA was investigated by titrating a dilute solution of the
complex (0.065 mm) with the protein up to a 0.9 mm concen-
tration and by following the resulting increase of the observed
longitudinal proton relaxation rate, R1, at 20 MHz and 298 K
(Figure 3a). A pronounced enhancement of R1 is observed at
Conclusion
The GdIII complex containing a novel amphiphilic ligand de-
rived from the heptadentate chelator OBETA functionalised
with a n-hexadecyl chain, Gd(HD-OBETA), was studied in detail
by 1H NMR relaxometric techniques. The aliphatic chain of
Gd(HD-OBETA) favours the formation of micellar aggregates
and the inclusion in the lipidic bilayer of liposomes and in the
hydrophobic pockets of HSA. The relaxivity of the lipidic NPs is
remarkable (48.3 mmÀ1 sÀ1 at 20 MHz and 298 K in case of lipo-
somes) owing to their macromolecular structure and to the
presence of two coordinated water molecules in the Gd(HD-
OBETA) complex. The most striking result, however, was ob-
tained by measuring the relaxivity of the supramolecular
adduct formed with HSA (71.4 mmÀ1 sÀ1 at 30 MHz and 298 K).
These results, coupled with the good stability data recently re-
ported for Gd(OBETA),[10] promise the future applicability of
suitably modified amphiphilic OBETA-like Gd complexes for en-
capsulation in liposomes or other biocompatible, hydrophobic
NPs as nanosized MRI probes or as blood-pool MR agents.
Experimental Section
All chemicals were purchased from Sigma–Aldrich, Alfa Aesar, or
TCI Europe and were used without further purification. 1H and
13C NMR spectra were recorded on a JEOL Eclipse ECP 300 spec-
trometer operating at 7.05 T. Chemical shifts were reported relative
to tetramethylsilane (TMS) and were referenced by using the resid-
ual proton resonances of the solvent. Electrospray ionization mass
spectra (ESIMS) were recorded on a SQD 3100 mass detector
(Waters), operating in positive- or negative-ion mode, with 1%v/v
formic acid in methanol as the carrier solvent. DLS experiments
were carried out by using a Zetasizer NanoZS, Malvern, UK, operat-
ing in a particle-size range from 0.6 nm to 6 mm and equipped
with a He/Ne laser with l=633 nm. Photoluminescence emission
spectra were recorded on a Horiba Jobin–Yvon Model IBH FL-322
Fluorolog 3 spectrometer equipped with a 450 W xenon arc lamp,
double-grating excitation and emission monochromators
(2.1 nmmmÀ1 dispersion; 1200 grooves per mm) and a Hamamatsu
Model R928 photomultiplier tube. Time-resolved measurements
were performed by using the time-correlated single-photon count-
ing (TCSPC) option. A 370 nm spectra LED laser was used to excite
the sample. Signals were collected by using an IBH DataStation
Hub photon-counting module. Data analysis was performed by
Figure 3. Top: water proton relaxation rate of a 0.065 mm aqueous solution
of Gd(HD-OBETA) as a function of increasing amounts of HSA (20 MHz,
298 K). Bottom: 1H NMRD profile of the Gd(HD-OBETA)/HSA adduct at 298 K.
low HSA concentration to indicate a very strong affinity of the
complex for the hydrophobic pockets of the protein. The non-
linear fitting of the experimental data according to the para-
magnetic relaxation enhancement (PRE) equations provides
the values for the association constant, KA =1.3”106 mÀ1, and
for the relaxivity of the protein conjugate, rb1 =68.1 mmÀ1 sÀ1
.
The KA value provides just a simple indication of the binding
strength since different binding sites of different affinity are
likely to be present on the protein.[28] Remarkably, the KA value
is higher than those found for the interaction with high-affinity
fatty acid sites on HSA of a Gd(DOTA-monoamide) dimer bear-
ChemPlusChem 2016, 81, 235 – 241
www.chempluschem.org
239
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
using the commercially available DAS6 software (HORIBA JobinY-
von IBH).
HD-OBETA-Et4 (11)
Ethyl 2-[2-(2-aminoethoxy)ethylamino]hexadecanoate (500 mg,
1.2 mmol) was dissolved in acetonitrile (10 mL). Ethyl 2-chloroace-
tate (510 mL, 4.8 mmol, 4 equiv) and DIPEA (1.05 mL, 6 mmol,
5 equiv) were added sequentially and the resulting reaction mix-
ture was stirred overnight at room temperature. Additional ethyl 2-
chloroacetate (130 mL, 1.2 mmol, 1 equiv) and DIPEA (200 mL,
1.2 mmol, 1 equiv) were added and the reaction mixture was
heated to reflux for 2 h. The solvent was removed by evaporation
under reduced pressure. The residue was dissolved in diethyl ether
and was then washed three times with a Na2CO3 saturated solution
and twice with a NaCl saturated solution. The organic phase was
dried over anhydrous Na2SO4, filtered, and the solvent was re-
moved by evaporation under reduced pressure. The residue was
separated by gravimetric chromatography on a silica-gel column
(petroleum ether/ethyl acetate 9:1 as eluent) giving HD-OBETA-Et4
as a colourless oil (720 mg, 1.07 mmol, 89%). 1H NMR (CDCl3,
300 MHz, 298 K): d=4.15 (q, J=7.1 Hz, 4H), 4.14 (q, J=7.0 Hz, 2H),
4.13 (q, J=7.0 Hz, 2H), 3.59–3.42 (m, 10H), 3.38 (t, J=7.4 Hz, 1H),
2.97–2.83 (m, 2H), 2.92 (t, J=5.6 Hz, 2H), 1.75–1.51 (m, 2H), 1.48–
1.09 (m, 40H), 0.88 ppm (t, J=6.7 Hz, 3H); 13C NMR (CDCl3,
75.4 MHz, 298 K): d=173.6 (C), 172.2 (2C), 171.4 (C), 70.7 (2CH2),
70.3 (CH2), 65.4 (CH), 60.4 (2CH2), 60.3 (CH2), 60.2 (CH2), 56.0 (CH2),
53.8 (CH2), 53.2 (CH2), 52.2 (CH2), 32.0 (CH2), 30.6 (CH2), 29.7 (9CH2),
29.6 (CH2), 29.5 (CH2), 26.2 (CH2), 22.7 (CH2), 14.4 (CH3), 14.3 (2CH3),
14.1 ppm (CH3).
2-Bromostearoyl chloride
Stearic acid (40 g, 140.6 mmol) was dissolved in SOCl2 (50 mL,
0.7 mol, 5 equiv) and the homogeneous solution was heated to
reflux for one hour. Bromine (44.5 g, 2 equiv, 278 mmol) was added
dropwise while heating at reflux and the resulting reaction mixture
was stirred at reflux for 3 h and overnight at room temperature.
Volatile compounds were removed under reduced pressure to
yield 2-bromostearoyl chloride as a light yellowish viscous oil
(53.4 g, 139.8 mmol, 99%). 1H NMR (CDCl3, 300 MHz, 298 K): d=
4.49 (t, J=6.6 Hz, 1H), 2.23–1.95 (m, 2H), 1.58–1.20 (m, 28H),
0.87 ppm (t, J=6.7 Hz, 3H); 13C NMR (CDCl3, 75.4 MHz, 298 K): d=
170.2 (C), 54.2 (CH), 34.9 (CH2), 32.0 (CH2), 29.8 (6CH2), 29.65 (CH2),
29.51 (CH2), 29.46 (CH2), 29.30 (CH2), 28.8 (CH2), 26.9 (CH2), 22.8
(CH2), 14.2 ppm (CH3).
Ethyl 2-bromostearoate (9)
2-Bromostearoyl chloride (10 g, 26.2 mmol) was dissolved in etha-
nol (100 mL) and the resulting reaction mixture was heated to
reflux for one hour, monitoring periodically by using thin-layer cho-
romatography (TLC) (petroleum ether/ethyl acetate 95:5). An
excess amount of ethanol was removed by evaporation under re-
duced pressure to give ethyl 2-bromostearate as a white solid
(10.3 g, 26.4 mmol, 96%). 1H NMR (CDCl3, 300 MHz, 298 K): d=
4.24–4.12 (m, 3H), 2.10–1.85 (m, 2H), 1.35–1.12 (m, 31H), 0.85 ppm
(t, J=6.1 Hz, 3H); 13C NMR (CDCl3, 75.4 MHz, 298 K): d=174.5 (C),
61.9 (CH2), 46.2 (CH), 35.0 (CH2), 32.0 (CH2), 29.73 (7CH2), 29.65
(CH2), 29.54 (CH2), 29.43 (CH2), 29.39 (CH2), 28.9 (CH2), 27.3 (CH2),
22.8 (CH2), 14.1 (CH3), 14.0 ppm (CH3).
HD-OBETA·2HCl (12)
HD-OBETA-Et4 (700 mg, 1.04 mmol) was dissolved in ethanol (5 mL)
and solid NaOH (210 mg, 5.2 mmol, 5 equiv) was added at room
temperature. The resulting reaction mixture was stirred at room
temperature overnight. The hydrolysis was checked by 1H NMR
spectroscopy. Na4(HD-OBETA): 1H NMR (CD3OD, 300 MHz, 298 K):
d=3.50 (t, J=4.6 Hz, 4H), 3.14 (d, J=16.6 Hz, 1H), 3.01 (d, J=
16.5 Hz, 1H), 3.07–2.93 (m, 5H), 2.70–2.62 (m, 2H), 2.59–2.51 (m,
2H), 1.69–1.43 (m, 2H), 1.40–1.20 (m, 28H), 0.89 ppm (t, J=6.7 Hz,
3H). The resulting solution was acidified to pH 2 with concentrated
hydrochloric acid. The white precipitate was centrifuged, washed
with cold water, and the residual solvent was removed under re-
Ethyl 2-[2-(2-aminoethoxy)ethylamino]hexadecanoate (10)
Ethyl 2-bromostearate (4.5 g, 11.5 mmol, 3 equiv) was dissolved in
acetonitrile (75 mL) and 2,2’-oxybisethylamine (400 mg, 3.8 mmol,
1 equiv) was added in one portion. Powdered potassium carbonate
(8.0 g, 57.5 mmol, 5 equiv) was added and the resulting reaction
mixture was vigorously stirred and heated to reflux for 6 h, moni-
toring periodically by TLC. Potassium iodide (190 mg, 1.15 mmol,
0.1 equiv) was added and the reaction mixture was stirred for 2 h
at reflux and overnight at room temperature. The solvent was re-
moved by evaporation under reduced pressure. The residue was
dissolved with dichloromethane and washed three times with
a Na2CO3 saturated solution and twice with a NaCl saturated solu-
tion. The organic phase was dried over anhydrous Na2SO4, filtered,
and the solvent was removed under reduced pressure. The residual
oil was separated by gravimetric chromatography on a silica-gel
column (eluent: petroleum ether/ethyl acetate, 9:1!6:4 gradient),
yielding ethyl 2-[2-(2-aminoethoxy)ethylamino]hexadecanoate as
duced pressure leading to HD-OBETA·2HCl as
a white solid
1
(650 mg, 1.03 mmol). H NMR (CD3OD, 300 MHz, 323 K): d=3.91 (s,
4H), 3.77–3.63 (m, 7H), 3.37 (t, J=4.9 Hz, 2H), 3.20 (t, J=5.1 Hz,
2H), 1.93–1.68 (m, 2H), 1.55–1.18 (m, 28H), 0.89 ppm (t, J=6.7 Hz,
3H); 13C NMR (CD3OD, 75.4 MHz, 323 K): d=170.0 (4C), 67.6 (CH2),
66.8 (CH), 66.5 (CH2), 56.1 (2CH2), 55.1 (CH2), 53.7 (CH2), 53.5 (CH2),
48.0 (CH2), 31.7 (CH2), 29.4 (11CH2), 29.1 (CH2), 28.6 (CH2), 26.3
(CH2), 22.3 (CH2), 11.6 ppm (CH3).
Gd(HD-OBETA)
Gd(HD-OBETA) was synthesised by adding small volumes of a stock
solution of GdCl3·6H2O to a solution of the ligand, while maintain-
ing the pH at 6.5 with diluted NaOH. The complexation process
was monitored by measuring the change in the longitudinal water
proton relaxation rate (R1) at 20 MHz and 258C (pH 6.5) as a func-
tion of the concentration of GdIII.
1
a colourless oil (740 mg, 1.8 mmol, 47%). H NMR (CDCl3, 300 MHz,
298 K): d=4.21–4.09 (m, 2H), 3.58–3.38 (m, 4H), 3.20 (t, J=6.6 Hz,
1H), 2.83 (t, J=5.2 Hz, 2H), 2.82–2.74 (m, 1H), 2.60 (dt, J=11.9, J=
5.0 Hz, 1H), 1.81 (brs, 3H), 1.64–1.54 (m, 2H), 1.33–1.18 (m, 31H),
0.84 ppm (t, J=6.7 Hz, 3H); 13C NMR (CDCl3, 75.4 MHz, 298 K): d=
175.5 (C), 73.1 (CH2), 70.6 (CH2), 61.7 (CH), 60.5 (CH2), 47.6 (CH2),
41.9 (CH2), 33.6 (CH2), 32.0 (CH2), 29.74 (8CH2), 29.62 (CH2), 29.51
(CH2), 29.42 (CH2), 28.8 (CH2), 22.7 (CH2), 14.4 (CH3), 14.2 ppm (CH3);
ESI+MS: m/z: 415.25 [M+H]+.
Liposome preparation
Long-circulating liposomes were prepared as described previous-
ly.[18] The total amount of phospholipids and the amphiphilic com-
ChemPlusChem 2016, 81, 235 – 241
www.chempluschem.org
240
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
plex was 20 mgmLÀ1. Briefly, appropriate amounts of dipalmitoyl-
phosphatidylcholine (DPPC), cholesterol, DSPE-PEG2000 (DSPE=
1,2-distearoyl-sn-glycero-3-phosphoethanolamine, PEG=polyethy-
lene glycol), and Gd(HD-OBETA) in a molar ratio of 55:30:5:10, re-
spectively, were dissolved in chloroform/methanol (95:5 by
volume) in a round-bottomed flask. A lipid film was prepared after
slow solvent removal under reduced pressure on a rotary evapora-
tor. The film was then dried under a stream of nitrogen for 2 h. Lip-
osomes were formed by adding an isotonic buffer at pH 7.4, com-
posed of 10 mm 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid
(HEPES) and 135 mm NaCl, to the lipidic thin film. The hydration
was performed at 558C and was accompanied by vigorous shak-
ing. The obtained suspension was extruded several times (Lipex ex-
truder, Northern Lipids Inc.) through polycarbonate filters with pro-
gressively reduced pore diameters from 400 to 100 nm. Liposomes
were dialysed briefly (4 h) to remove any unincorporated material.
Chemistry of Contrast Agents in Medical Magnetic Resonance Imaging,
2nd ed. (Eds.: A. Merbach, L. Helm, É. Tóth), Wiley, New York, 2013.
[3] G. J. Stasiuk, N. J. Long, Chem. Commun. 2013, 49, 2732–2746.
[4] S. Laurent, C. Henomount, L. Vander Elst, R. N. Muller, Eur. J. Inorg.
Chem. 2012, 1889–1915.
[5] a) S. Aime, L. Calabi, C. Cavallotti, E. Gianolio, G. B. Giovenzana, P. Losi,
A. Maiocchi, G. Palmisano, M. Sisti, Inorg. Chem. 2004, 43, 7588–7590;
b) G. Gugliotta, M. Botta, L. Tei, Org. Biomol. Chem. 2010, 8, 4569–4574.
[6] a) S. Hajela, M. Botta, S. Giraudo, J. Xu, K. N. Raymond, S. Aime, J. Am.
Chem. Soc. 2000, 122, 11228–11229; b) A. Datta, K. N. Raymond, Acc.
Chem. Res. 2009, 42, 938–947.
[7] S. Aime, M. Botta, S. Geninatti Crich, G. B. Giovenzana, R. Pagliarin, M.
Sisti, E. Terreno, Magn. Reson. Chem. 1998, 36, S200–S208.
[8] L. Pellegatti, J. Zhang, B. Drahos, S. Villette, F. Suzenet, G. Guillaumet, S.
Petoud, É. Tóth, Chem. Commun. 2008, 6591–6593.
[9] E. M. Gale, N. Kenton, P. Caravan, Chem. Commun. 2013, 49, 8060–8062.
[10] a) Z. Baranyai, M. Botta, M. Fekete, G. B. Giovenzana, R. Negri, L. Tei, C.
Platas-Iglesias, Chem. Eur. J. 2012, 18, 7680–7685; b) R. Negri, Z. Bara-
nyai, L. Tei, G. B. Giovenzana, C. Platas-Iglesias, J. Bodnàr, A. Vàgner, M.
Botta, Inorg. Chem. 2014, 53, 12499–12511.
[11] M. Botta, L. Tei, Eur. J. Inorg. Chem. 2012, 1945–1960.
[12] a) A. Accardo, D. Tesauro, A. Luigi, C. Pedone, G. Morelli, Coord. Chem.
Rev. 2009, 253, 2193–2213; b) S. Aime, M. Botta, M. Fasano, S. G. Crich,
E. Terreno, J. Biol. Inorg. Chem. 1996, 1, 312–319.
[13] a) S. Langereis, T. Geelen, H. Grüll, G. J. Strijkersand, K. Nicolay, NMR
Biomed. 2013, 26, 728–744; b) W. T. Al-Jamal, K. Kostarelos, Acc. Chem.
Res. 2011, 44, 1094–1104.
[14] G. A. Rolla, M. Botta, L. Tei, C. Cabella, S. Ghiani, C. Brioschi, A. Maiocchi,
Chem. Eur. J. 2013, 19, 11189–11193.
[15] R. N. Mariano, D. Alberti, J. C. Cutrin, S. Geninatti Crich, S. Aime, Mol.
Pharm. 2014, 11, 4100–4106.
[16] D. P. Cormode, J. C. Frias, Y. Ma, W. Chen, T. Skajaa, K. Briley-Saebo, A.
Barazza, K. J. Williams, W. J. M. Mulder, Z. A. Fayad, E. A. Fisher, Clin. Lipi-
dol. 2009, 4, 493–500.
[17] C. E. Carney, I. L. Lenov, C. J. Baker, K. W. MacRenaris, A. L. Eckermann,
S. G. Sligar, T. J. Meade, Bioconjugate Chem. 2015, 26, 899–905.
[18] F. Kielar, L. Tei, E. Terreno, M. Botta, J. Am. Chem. Soc. 2010, 132, 7836–
7837.
[19] L. Tei, G. Gugliotta, Z. Baranyai, M. Botta, Dalton Trans. 2009, 9712–
9714.
[20] E. Gianolio, G. B. Giovenzana, D. Longo, I. Longo, I. Menegotto, S. Aime,
Chem. Eur. J. 2007, 13, 5785–5797.
Relaxometric measurements
The water proton longitudinal relaxation rates as a function of the
magnetic-field strength were measured in nondeuterated aqueous
solutions on a fast field-cycling Stelar SmarTracer relaxometer
(Stelar srl, Mede (PV), Italy) over a continuum of magnetic-field
strengths from 0.00024 to 0.25 T (corresponding to 0.01–10 MHz
proton Larmor frequencies). The relaxometer operates under com-
puter control with an absolute uncertainty in 1/T1 of Æ1%. Addi-
tional longitudinal relaxation data in the range 15–70 MHz were
obtained on a Stelar Relaxometer connected to a Bruker WP80
NMR electromagnet adapted to variable-field measurements (15–
80 MHz proton Larmor frequency). The exact concentration of GdIII
was determined by measurement of bulk magnetic susceptibility
shifts of a tBuOH signal or by inductively coupled plasma mass
spectrometry (ICP-MS, Element-2, Thermo-Finnigan, Rodano (MI),
1
Italy). The H T1 relaxation times were acquired by the standard in-
version recovery method with a typical 908 pulse width of 3.5 ms,
and 16 experiments of 4 scans. The temperature was controlled
with a Stelar VTC-91 airflow heater equipped with a calibrated
copper-constantan thermocouple (uncertainty of Æ0.18C). The
noncovalent interaction of Gd(HD-OBETA) with HSA was evaluated
by using the proton relaxation enhancement (PRE) method: R1 was
monitored as a function of the amount of HSA added to 0.065 mm
Gd(HD-OBETA) solution, at 298 K and 20 MHz, while maintaining
the pH at 6–7.
[21] a) R. Hovland, C. Gløgard, A. J. Aasen, J. Klaveness, Org. Biomol. Chem.
2003, 1, 644–647; b) C. Ferroud, H. Borderies, E. Lasri, A. Guy, M. Port,
Tetrahedron Lett. 2008, 49, 5972–5975.
ˇ
[22] C. S. Bonnet, L. Pellegatti, F. Buron, C. M. Shade, S. Villette, V. Kubícek, G.
Guillaumet, F. Suzenet, S. Petoud, É. Tóth, Chem. Commun. 2010, 46,
124–126.
[23] L. Tei, G. Gugliotta, S. Avedano, G. B. Giovenzana, M. Botta, Org. Biomol.
Chem. 2009, 7, 4406–4414.
[24] a) G. Lipari, S. Szabo, J. Am. Chem. Soc. 1982, 104, 4546–4559; b) G.
Lipari, S. Szabo, J. Am. Chem. Soc. 1982, 104, 4559–4570.
[25] S. Aime, E. Gianolio, E. Terreno, G. B. Giovenzana, R. Pagliarin, M. Sisti, G.
Palmisano, M. Botta, M. P. Lowe, D. Parker, J. Biol. Inorg. Chem. 2000, 5,
488–497.
[26] S. Lacerda, C. S. Bonnet, A. Pallier, S. Villette, F. Foucher, F. Westall, F.
Buron, F. Suzenet, C. Pichon, S. Petoud, É. Tóth, Small 2013, 9, 2662–
2666.
Acknowledgements
The financial support from “Compagnia di San Paolo” (CSP-2012,
NANOPROGLY Project) is gratefully acknowledged. This study was
performed under the auspices of EU COST Action TD1004.
[27] E. Cittadino, M. Botta, L. Tei, F. Kielar, R. Stefania, E. Chiavazza, S. Aime,
E. Terreno, ChemPlusChem 2013, 78, 712–722.
[28] a) S. Dumas, V. Jacques, W. C. Sun, J. S. Troughton, J. T. Welch, J. M.
Chasse, H. Schmitt-Willich, P. Caravan, Invest. Radiol. 2010, 45, 600–612;
b) V. Jacques, S. Dumas, W. C. Sun, J. S. Troughton, M. T. Greefield, P. Car-
avan, Invest. Radiol. 2010, 45, 613–624.
Keywords: amphiphiles
lanthanides · micelles
·
chelates
·
imaging agents
·
[1] a) M. A. Brown, R. P. Semelka, MRI-Basic Principles and Application, John
Wiley & Sons, Hoboken, 2003; b) P. Caravan, Chem. Soc. Rev. 2006, 35,
512–523; c) S. Aime, M. Botta, E. Terreno, Adv. Inorg. Chem. 2005, 57,
173–237.
Manuscript received: August 11, 2015
Accepted Article published: September 1, 2015
Final Article published: September 17, 2015
[2] a) R. B. Lauffer, Chem. Rev. 1987, 87, 901–927; b) P. Caravan, J. J. Ellison,
T. J. McMurry, R. B. Lauffer, Chem. Rev. 1999, 99, 2293–2252; c) The
ChemPlusChem 2016, 81, 235 – 241
www.chempluschem.org
241
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim