AAZTA-based bifunctional chelating agents for the synthesis of multimeric/dendrimeric MRI contrast agents

Article abstract of DOI:10.1039/c0ob00096e

Monomeric and dimeric bifunctional chelates based on the AAZTA (6-amino-6-methylperhydro-1,4-diazepinetetraacetic acid) platform bearing arylamino and isothiocyanate groups for conjugation to biomolecules were synthesised. Both systems were used for the p

Full text of DOI:10.1039/c0ob00096e

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
AAZTA-based bifunctional chelating agents for the synthesis of
multimeric/dendrimeric MRI contrast agents†
Giuseppe Gugliotta, Mauro Botta and Lorenzo Tei*
Received 9th May 2010, Accepted 9th July 2010
DOI: 10.1039/c0ob00096e
Monomeric and dimeric bifunctional chelates based on the AAZTA
(6-amino-6-methylperhydro-1,4-diazepinetetraacetic acid) platform bearing arylamino and
isothiocyanate groups for conjugation to biomolecules were synthesised. Both systems were used for the
preparation of high relaxivity dendrimeric (PAMAM G1) and multimeric octa-GdIII complexes. These
systems show enhanced relaxometric properties attributed to the presence of two coordinated, fast
exchanging, water molecules (q = 2) on each metal ion and to a rather rigid and compact molecular
structure.
strengths.9,10 Thus, the synthesis of dimeric BFCAs could be very
Introduction
useful both for increasing the payload of GdIII chelates per unit of
functional group on the biological or nano-sized carrier and for
the straightforward preparation of multimeric systems for high
field MRI or MR-molecular imaging applications.
Complexes of transition and lanthanide(III) metal ions have found
a broad range of applications in chemistry, biology and medicine
such as diagnostic imaging,1 molecular imaging,2 tumour therapy3
and luminescent materials.4 Among all the chelating ligands
reported so far, those based on a polyamino polycarboxylic
structure are particularly useful and widely employed for their
large versatility and strong coordination properties. Bifunctional
chelating agents (BFCAs) are molecules containing two different
moieties: a strong metal chelating unit and a suitable function able
to link the probe to a given biomolecule.5 The linkage must be
stable under physiological conditions (i.e. hydrolytic, oxidizing,
and reducing conditions in an aqueous solution near neutrality
containing several anions, metal cations, small molecules, and
enzymes) and possibly performed under mild conditions to avoid
vector and/or probe degradation.5 Moreover, the conjugation
step should be quantitative, selective, fast and easy to perform.
For example, BFCAs have been extensively used to label specific
biological carriers or targeting vectors (e.g. peptides, proteins or
antibody fragments) with a (radio)metal ion.6,7
Paramagnetic contrast agents (CAs) based on GdIII chelates
have proved essential in differentiating soft tissues in Magnetic
Resonance Imaging (MRI) protocols.1,8 However, as MRI is
characterized by an intrinsically low sensitivity, the most inno-
vative molecular or cellular imaging applications require highly
efficient CAs that produce a sufficient relaxation effect by a
limited mass of the agent with high local concentration. A possible
strategy to obtain this effect is the delivery of a high number of
imaging reporters at the site of interest by conjugating several
GdIII chelates to macromolecules or nanoparticles linked to a
targeting vector.2 Moreover, for high field MR scanners (1.5–
4.7 T), multimeric GdIII T1 agents of medium molecular weight
(~ 2–6 kDa) have been proposed as good candidates as they show
excellent relaxation properties over a broad range of imaging field
Among the various polyamino polycarboxylic BFCAs em-
ployed for MRI studies, one of the most promising is represented
by the heptadentate ligand AAZTA.11 In fact, as [GdAAZTA]-
shows excellent properties in terms of thermodynamic stability
and relaxivity (e.g. two water molecules in the inner coordination
sphere of GdIII with a fast rate of exchange), a functionalised
derivative of [GdAAZTA]- can be considered as a prototype for the
development of new classes of high relaxivity multimeric agents.
We recently reported a general synthetic procedure towards
a series of AAZTA bifunctional prochelators for coupling both
to NH2-containing biomolecules and to other reactive functional
groups of relevance for biomedical applications.12 Among these,
the AAZTA-like chelator with a free hydroxymethyl group and the
carboxylates protected as t-butyl esters (AAZTA-OH, Scheme 1)
was used as starting material to react with succinic anhydride to
form an ester or with thionyl chloride to form the alkyl chloride
derivative. However, the versatility of these bifunctional chelators
is limited by the formation of a cyclic 2-oxomorpholine derivative
in both basic or acid conditions.
In order to prevent the intramolecular cyclization of AAZTA-
OH and exploit this ligand as a basic unit for the synthesis of
mono- and dimeric BFCAs, aromatic isocyanates were selected
as they react with alcohols at neutral pH and room temperature
forming carbamates. Using this procedure, mono- and dimeric
AAZTA derivatives containing arylamino and arylisothiocyanate
reacting groups were synthesised. The utility of these BFCAs was
demonstrated by their attachment onto a polyamine or PAMAM
G1 dendrimer to form octameric ligands. The synthesis of the
octa-GdIII complexes and their relaxivity is also reported.
Results and discussion
Dipartimento di Scienze dell’Ambiente e della Vita, Universita` del Piemonte
Orientale “Amedeo Avogadro”, Viale T. Michel 11, 15121, Alessandria, Italy.
E-mail: lorenzo.tei@mfn.unipmn.it
† Electronic supplementary information (ESI) available: Instrumental
details and HPLC methods. See DOI: 10.1039/c0ob00096e
A detailed study on the thermodynamic and kinetic stabilities
of AAZTA with several lanthanide ions was recently reported
by Baranyai et al.13 In the case of several GdDOTA and
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 4569–4574 | 4569
©

View Article Online
Scheme 1 Synthesis of monomeric AAZTA and GdAAZTA BFCAs. Reagents and conditions: i: CH2Cl2, rt, 2 h; ii: H2, Pd/C 10%, MeOH, rt, 18 h; iii:
CSCl2, CH2Cl2/KHCO3 sat. (1 : 1 v/v), rt, 2 h; iv: TFA/CH2Cl2 (1 : 1 v/v), rt, 18 h; v: GdCl3, H2O, pH = 7.
Scheme 2 Synthesis of dimeric AAZTA-based BFCAs. Reagents and conditions: i: SOCl2, reflux, 18 h; ii: NaN3, NaOH 1 M/CH2Cl2, rt, 1 h; iii: Toluene,
reflux, 3 h; iv: CH2Cl2, rt, 2 h; v: H2, Pd/C 10%, MeOH, rt, 18 h; vi: CSCl2, CH2Cl2/KHCO3 sat. (1 : 1 v/v), rt, 2 h; vii: TFA/CH2Cl2 (1 : 1 v/v), rt, 18 h;
viii: GdCl3, H2O, pH = 7, to obtain Gd2(10).
GdDTPA derivatives and of their mono- and bisamide analogs
the stability constants of the complexes vary only marginally
with the modification of the basic ligand structure. This is
because the coordination cage of the paramagnetic metal ion
remains unaffected. In accordance with these observations, the
modifications of the ligand AAZTA reported herein occur on
the exocyclic methyl group and therefore are not expected to
influence the coordinating ability towards lanthanide(III) ions
as also demonstrated by the relaxometric data reported for the
AAZTA-OH precursor.12
The synthesis of the monomeric BFCAs 2, 3 and GdL1 started
from the reaction of AAZTA-OH with 4-nitrophenyl isocyanate
as shown in Scheme 1 to obtain the 4-nitrophenyl carbamate 1 in
high yield. Interestingly, carbamates are known to be stable both
at the acidic pH values necessary for tert-butyl ester deprotection
and under physiological conditions. The nitro group was then
reduced to amine (2) by using hydrogen at ambient pressure and
10 wt% Pd/C catalyst, and finally the amine was reacted with
thiophosgene to form almost quantitatively the isothiocyanate
derivative (3). This latter can readily react with a variety of amino-
containing biomolecules. It is worth noting that derivative 2 is
also a BFCA that can be coupled to carboxylic acids by forming
stable amide bonds. Interestingly, deprotection of the t-butyl esters
of 2 yielded the ligand 4 which was complexed with GdCl3 and
subsequently reacted with thiophosgene to give GdL1, a BFCA
that can react with NH2 groups also in aqueous media (Scheme 1)
Then, a functionalised AAZTA dimer was designed as a multi-
meric BFCA for use in the construction of higher oligomers. The
synthesis was carried out employing a nitro-isophthalic acyl azide
transformed into the isocyanate through Curtius rearrangement
by a modification of the reported procedures.14 In particular, nitro-
isophthalic acid was converted into acyl chloride with thionyl
chloride and then into acyl azide with sodium azide in a mixture
of NaOH (1.0 mol L-1) and CH2Cl2 (Scheme 2). Heating to reflux
(3 h) in toluene quantitatively yielded the nitro isophthaloyl diiso-
cyanate that was then reacted with two equivalents of AAZTA-
OH to obtain the dimeric ligand (6) after column chromatographic
purification. The nitro group was converted into amine and then to
isocyanate following the analogous procedure reported to obtain
3 from 1. As for the monomeric system, deprotection of the t-butyl
esters of 7 yielded the ligand 9 which was complexed with GdCl3
and subsequently reacted with thiophosgene to give Gd2L2, a
dimeric Gd-based bifunctional chelate that can react with amines
also in aqueous media. It might be worth outlining that derivatives
7, 8, Gd2(10) and Gd2L2 (Scheme 2) are among the first examples
of dimeric bifunctional chelating agents.15
4570 | Org. Biomol. Chem., 2010, 8, 4569–4574
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Scheme 3 Synthesis of octameric ligands L3 and L4. Reagents and conditions: i: TFA/CH2Cl2 (1 : 1 v/v), rt, 18 h.
As an exemplification of the use of 3 and 8, the reaction with
aliphatic primary polyamines was carried out in CH2Cl2 in order
to obtain two octameric ligands (Scheme 3). While 3 was reacted
with octaamino PAMAM G1 dendrimer, for the dimeric ligand
8 the reaction was carried out with N,N¢-bis-(2-aminoethyl)-
ethylenediamine, prepared as described in the literature.16 Oc-
tameric protected ligands were purified by crystallization from
diethyl ether/petroleum ether. Then, the t-butyl esters were
deprotected in a 1 : 1 mixture of CH2Cl2 and trifluoroacetic acid
to obtain the octameric ligands L3 and L4. Both ligands were
also purified and desalted by passing through a sephadex G25 gel
filtration column.
The GdIII complexes of these octameric ligands were prepared by
adding small volumes of a stock solution of GdCl3 to a solution of
the ligand and maintaining 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 as a
function of the concentration of gadolinium(III).
The relaxivity, r1, per Gd of Gd8L3 and Gd8L4 is 21.9 and
24.4 mM-1 s-1 (20 MHz, 298 K), respectively. To the best of our
knowledge, Gd8L3 is the first example of a dendrimer conjugated
to Gd-complexes possessing two water molecules (q = 2) in
the first coordination sphere.17 The higher relaxivity of Gd8L4,
composed of four dimeric AAZTA units, can be attributed to
its more compact and rigid structure that affects the effective
reorientational correlation time, tR. To account for this difference,
mass relaxivities or densities of relaxivity have been defined as
the enhancement of the relaxation rate by a unit mass (g L-1) of
CA.10 High density of relaxivity is required for applications such
as cell imaging in which a sufficient relaxation effect has to be
produced by a limited mass of the agent. The values obtained
for Gd8L2 and Gd8L3 are 24.7 and 36.5 (g L-1)-1 s-1, respectively
(20 MHz, 298 K). The highest density of relaxivity so far reported
is that of “metallostar” a supramolecular system based on the
self assembly of a dimeric Gd-DTTA-bipyridine complex into a
Fe(III) complex (DTTA = diethylenetriaminotetraacetic acid).10
The resulting hexameric Gd-complex has a density of relaxivity at
20 MHz and 298 K of 43.3 (g L-1)-1 s-1. This value is more than
double that of Gadomer17, an intravascular dendrimeric MRI
contrast agent under development (Bayer Schering Pharma AG)
for MR angiography, containing 24 Gd-DOTA-monoamide units
(22.7 (g L-1)-1 s-1).18 Thus, our octameric Gd-complexes Gd8L3
and Gd8L4 show densities of relaxivity intermediate between
these multimeric/dendrimeric systems. In addition, these systems
show enhanced relaxometric properties when compared to systems
of similar size/molecular weight because of the concomitant
occurrence of fast rate of water exchange and high hydration
number (q = 2).
Conclusions
In conclusion, the synthesis of monomeric and dimeric BFCAs
offers the possibility to design and construct a series of CAs
containing a large payload of highly efficient GdIII chelates by
a modular synthetic approach. These modules have been proved
to be useful for the synthesis of multimeric/dendrimeric systems
endowed with high relaxivity both per Gd and per molecule. They
can also be very versatile since by varying linkers or scaffold
it is possible to design multimeric systems suitable for different
requirements. At the same time, these oligomeric BFCAs may be
of interest for increasing the number of GdIII units attached to
various types of particles and thus for obtaining high relaxivity
nano-sized systems. Also direct conjugation to targeting vectors
would permit the synthesis of multimeric imaging probes in few
steps to allow the delivery of highly efficient CAs at the target site.
Finally, it is worth noting that, as AAZTA is a highly efficient
ligand for the coordination of lanthanide or other trivalent metal
ions, such systems could also be useful for the complexation
of other diagnostic (111In, 67/68Ga) or therapeutic (90Y, 177Lu)
radiometals for nuclear medicine applications.
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 4569–4574 | 4571
©

View Article Online
was stirred for 2 h at rt. The organic phase was separated and
extracted with water (3 ¥ 10 mL); then the organic phase was
dried over NaSO4, filtered and evaporated to yield a yellow solid.
TLC (silica gel 60 F254, petroleum ether/EtOAc 8 : 2, detection:
UV 254 nm): Rf 0.41. 1H-NMR (CDCl3) 400 MHz d = 7.41
(bs, 2H, CH), 7.20 (bs, 2H, CH), 7.15 (bs, 1H, NH), 4.20 (s,
2H, CH2OCO), 3.72 (s, 4H, CH2CO), 3.21 (s, 4H, CH2CO), 3.12
(d, 2H, J = 14.3 Hz, CH2C), 2.87 (d, 2H, J = 14.3 Hz, CH2C),
2.70–2.64 (m, 4H, CH2CH2), 1.44 (s, 18H, CH3), 1.43 (s, 18H,
CH3).13C-NMR (CDCl3) 100 MHz d = 172.9, 170.4 (CO), 153.3
(NHCOO), 140.0 (C), 128.0 (C), 110.3 (CH), 106.7 (CH), 81.4,
81.2 (CCH3), 67.6 (CH2OCO), 63.4 (C), 62.2, 62.3 (CH2CO), 59.0,
51.6 (CH2cyclo), 28.2, 28.0 (CH3). ESI-MS (m/z): 778.47 (M + H+);
calc for C38H60N5O10S: 778.41. IR spectrum (KBr disk): 3322,
2971, 2111, 1739, 1548, 1218, 1155 cm-1.
Experimental
General experimental conditions
All reactants were used as supplied from commercial sources
unless stated otherwise. H and 13C NMR spectra were recorded
1
on a JEOL ECP 400 spectrometer. Mass spectra with elec-
trospray ionization (ES) were recorded on a SQD 3100 Mass
Detector (Waters). The HPLC-MS analysis was carried out
on a Waters system equipped with Waters 1525 binary HPLC
Pump and Waters 2489 UV/vis and Waters SQD 3100 detec-
tors. The water proton longitudinal relaxation rates of aque-
ous solutions of Gd8L3 and Gd8L4 were measured by using
a Stelar Spinmaster spectrometer (Mede, Italy) operating at
0.5 T and 298 K. Infrared (IR) spectra were recorded in the
range 4000–400 cm-1 using a Bruker Equinox 55 spectrometer.
Elemental analysis were performed on a EuroVector EA 3000
instrument. AAZTA-OH [1,4-bis(t-butoxycarbonylmethyl)-6-
[bis(t-butoxycarbonylmethyl)]amino-6-hydroxymethylperhydro-
1,4-diazepine] was synthesised as described previously.12
General procedure for t-butyl ester deprotection
To a solution of polyaminocarboxylate-t-butyl esters in CH2Cl2
(ca. 0.4 N) was gradually added trifluoroacetic acid (equal volume)
and the mixture was stirred at rt overnight. The solution was then
evaporated in vacuo and the product was recovered with excess
diethyl ether, isolated by centrifugation, washed thoroughly with
diethyl ether and dried in vacuo obtaining pure desired product
as the trifluoroacetate salt. Final products were purified from
salts and low molecular weight impurities by gel filtration on a
Sephadex G25 column (30 cm ¥ 1.6 cm) (Pharmacia Biotech,
Uppsala, Sweden) using H2O as eluent. Analytical HPLC-MS
runs on final ligands (10 mL of a 2 mg mL-1 solution in H2O) were
carried out using Method 1 or 2 (see Electronic Supplementary
Information†) and H2O–TFA 0.1% (A) and CH3CN–TFA 0.1%
(or CH3OH–TFA 0.1%) (B) as eluents.
1. A solution of AAZTA-OH (600 mg; 1.00 mmol) in
dry CH2Cl2 (5 mL) was added under N2 to a solution of
p-nitrophenylisocyanate (165 mg; 1.00 mmol) in dry CH2Cl2
(5 mL). The reaction mixture was stirred for 2 h at rt and then
evaporated in vacuo. The crude product was purified on silica
gel chromatography column (petroleum ether/ethyl acetate 8/2)
yielding 1 as pale yellow solid (734 mg; 96%). TLC (silica gel 60
F254, petroleum ether/EtOAc 8 : 2, detection: UV 254 nm): Rf 0.45.
1H-NMR (CDCl3) 400 MHz d = 8.53 (bs, 1H, NH), 8.12 (d, 2H,
J = 9.2 Hz, CH), 7.57 (m, 2H, CH), 4.21 (s, 2H, CH2OCO), 3.65
(s, 4H, CH2CO), 3.22 (s, 4H, CH2CO), 3.05 (d, 2H, J = 14.3 Hz,
CH2C), 2.76–2.65 (m, 6H, CH2cyclo), 1.40 (s, 18H, CH3), 1.39 (s,
18H, CH3).13C-NMR (CDCl3) 100 MHz d = 172.8, 170.7 (CO),
153.1 (NHCOO), 144.7, 142.7 (C), 125.2, 117.7 (CH), 81.0, 80.9
(CCH3), 68.0 (CH2OCO), 63.3 (C), 62.0, 60.4 (CH2CO), 59.1,
51.8 (CH2cyclo), 28.2, 28.1 (CH3). ESI-MS (m/z): 766.55 (M + H+);
calc for C37H60N5O12: 766.42. IR spectrum (KBr disk): 3323, 2979,
2932, 1745, 1513, 1383, 1223, 1167 cm-1.
4. HPLC-MS, method 1, retention time 4.80 min, purity 95%.
1H-NMR (D2O) 400 MHz d = 7.43 (d, 2H, J = 8.4 Hz, CH), 7.31
(d, 2H, J = 8.4 Hz, CH), 4.11 (s, 2H, CH2OCO), 3.81 (s, 4H,
CH2CO), 3.77 (s, 4H, CH2CO), 3.49–3.43 (m, 8H, CH2cyclo). 13C-
NMR (CDCl3) 100 MHz d = 176.9, 171.4 (CO), 154.5 (NHCOO),
138.2, 125.5 (C), 122.9, 121.0 (CH), 66.1 (CH2OCO), 61.7 (C),
59.0, 53.3 (CH2CO), 58.1, 51.7 (CH2cyclo). ESI-MS (m/z): 512.20
(M + H+); calc for C21H30N5O10: 512.20. IR spectrum (KBr disk):
3407, 2919, 2610, 1729, 1659, 1198, 1123 cm-1.
2. A suspension of 10% Pd/C (70 mg) in water (1 mL) was
added to a solution of 1 (730 mg; 0.95 mmol) in MeOH (10 mL).
The mixture, under H2 atmosphere, was stirred overnight and
then filtered over celite. The filtrate was evaporated in vacuo
affording pure 2 (683 mg; 93%). TLC (silica gel 60 F254, petroleum
ether/EtOAc 8 : 2, detection: UV 254 nm): Rf 0.13. 1H-NMR
(CDCl3) 400 MHz d = 7.22 (s, 2H, J = 5.9 Hz, CH), 6.64
(d, 2H, J = 5.9 Hz, CH), 4.16 (s, 2H, CH2OCO), 3.71 (s, 4H,
CH2CO), 3.40 (bs), 2.88 (bs) (12H, CH2CO + CH2cyclo), 1.43 (s,
36H, CH3).13C-NMR (CDCl3) 100 MHz d = 172.7, 170.6 (bs)
(CO), 153.6 (NHCOO), 140.0, 130.1 (C), 120.5, 115.7 (CH), 81.2
(bs) (CCH3), 67.4 (bs) (CH2OCO), 61.7 (bs), 58.8 (bs), 51.8 (bs)
(C, CH2CO, CH2cyclo), 28.2 (CH3). ESI-MS (m/z): 736.25 (M +
H+); calc for C37H62N5O10: 736.45. IR spectrum (KBr disk): 3364,
2971, 1729, 1527, 1218, 1144 cm-1.
General procedure for gadolinium complex preparation
The GdIII complexes were prepared by mixing stoichiometric
amounts of the ligands and GdCl3 solution (final concentration
ca. 1 mM). Unchelated Gd3+ ions were eliminated by precipitation
of the hydroxide at basic pH by adding aliquots of a concentrated
NaOH solution.
[GdL1]. A solution of thiosphogene (54 mL; 0.70 mmol) in
dry CH2Cl2 (5 mL) was added dropwise in 2 min to a solution
of 4 (100 mg; ca. 0.14 mmol) in H2O (5 mL). The mixture was
vigorously stirred for 2 h at rt. Then, after adjustment of the pH
to 5.5, the water solution was washed 3 times with CH2Cl2 (5 mL)
and then evaporated in vacuo. ESI-MS (m/z): 708.06 (M + H+);
calc for C22H24N5O10SGd: 708.05 (100,0%), (isotopic distribution
consistent with Gd complex).
3. A solution of thiosphogene (83 mL; 1.09 mmol) in dry
CH2Cl2 (2 mL) was added drop wise in 2 min to a ice-bath
cooled mixture of 2 (670 mg; 0.91 mmol) in saturated potassium
bicarbonate (5 mL) and CH2Cl2 (5 mL), under vigorous magnetic
stirring. The ice-bath was removed after 5 min and the suspension
1-Nitro-3,5-benzenedicarbonyldiazide.
A
solution of 1-
nitroisophthaloyl dichloride (1.168 g; 4.73 mmol) in dry CH2Cl2
4572 | Org. Biomol. Chem., 2010, 8, 4569–4574
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
(4 mL) was added drop wise in 2 min to an ice-bath cooled
mixture of NaN3 (3.075 g; 47.3 mmol) in 1.0 mol L-1 NaOH
solution (10 mL) and CH2Cl2 (8 mL), under vigorous magnetic
stirring. The ice-bath was removed after 5 min and the mixture
was stirred for 45 min at rt. The organic phase was separated
and the aqueous phase was extracted with CH2Cl2 (3 ¥ 10 mL);
then, all organic phases were collected, dried over NaSO4, filtered
7.15 (bs, 2H, NH), 4.21 (s, 4H, CH2OCO), 3.69 (s, 8H, CH2CO),
3.25 (s, 8H, CH2CO), 3.09 (d, 4H, J = 14.3 Hz, CH2C), 2.87 (d,
4H, J = 14.3 Hz, CH2C), 2.70–2.64 (m, 8H, CH2CH2), 1.43 (s,
36H, CH3), 1.42 (s, 36H, CH3).13C-NMR (CDCl3) 100 MHz d =
172.8, 170.4 (CO), 153.1 (NHCOO), 140.0 (C), 128.0 (C), 110.3
(CH), 106.7 (CH), 81.3, 81.0 (CCH3), 67.5 (CH2OCO), 63.2 (C),
62.3, 62.1 (CH2CO), 59.0, 51.6 (CH2cyclo), 28.2, 28.0 (CH3). ESI-
MS (m/z): 1421.38 (M + H+); calc for C69H114N9O20S: 1420.79.
IR spectrum (KBr disk): 3320, 2968, 2109, 1739, 1548, 1208,
1155 cm-1.
1
and evaporated to yield a white solid (1.045 g, 83%). H-NMR
(CDCl3) 400 MHz d = 9.0 (d, 2H, J = 1.8 Hz), 8.9 (t, 1H, J =
1.8 Hz). 13C-NMR (CDCl3) 100 MHz d = 169.7 (CO), 148.8 (C),
135.1 (CH), 133.3 (2C), 128.8 (2CH).
9. HPLC-MS, method 1, retention time 6.51 min, purity 92%.
1H-NMR (D2O) 400 MHz d = 7.10 (bs, 3H, CH), 4.06 (s, 4H,
CH2OCO), 3.80 (s, 8H, CH2CO), 3.67 (s, 8H, CH2CO), 3.49–3.40
(m, 16H, CH2cyclo).13C-NMR (CDCl3) 100 MHz d = 177.1, 171.2
(CO), 154.5 (NHCOO), 137.1, 125.0 (C), 122.5, 121.0 (CH), 66.2
(CH2OCO), 61.5 (C), 59.4, 53.2 (CH2CO), 58.4, 51.7 (CH2cyclo).
ESI-MS (m/z): 930.45 (M + H+); calc for C36H52N9O20: 930.33.
IR spectrum (KBr disk): 3329, 2957, 2586, 1720, 1410, 1187,
1137 cm-1.
6. 1-Nitro-3,5-benzenedicarbonyldiazide (100 mg; 0.38 mmol)
was heated to reflux for 3 h in dry toluene, under N2. After cooling
to rt, a solution of AAZTA-OH (460 mg; 0.76 mmol) in dry CH2Cl2
(3 mL) was added under N2 atmosphere. The reaction mixture
was stirred overnight at rt, and then evaporated in vacuo. The
crude product was purified on a silica gel chromatography column
(petroleum ether/ethyl acetate 75/25) yielding pure 6 as a pale
yellow solid (472 mg, 88%). TLC (silica gel 60 F254, petroleum
ether/EtOAc 8 : 2, detection: UV 254 nm): Rf 0.35. 1H-NMR
(CDCl3) 400 MHz d = 8.03 (s, 2H, CH), 7.93 (bs, 2H, NH), 7.79 (s,
1H, CH), 4.23 (s, 4H, CH2OCO), 3.68 (s, 8H, CH2CO), 3.25 (s, 4H,
CH2CO), 3.24 (s, 4H, CH2CO), 3.08 (d, 4H, J = 14.3 Hz, CH2C),
2.78 (d, 4H, J = 14.3 Hz, CH2C), 2.80–2.63 (m, 8H, CH2 CH2), 1.42
(s, 36H, CH3), 1.41 (s, 36H, CH3).13C-NMR (CDCl3) 100 MHz d =
172.8, 170.7 (CO), 153.2 (NHCOO), 149.3 (C), 140.0 (2C), 113.0
(CH), 107.6 (2CH), 81.0, 80.8 (CCH3), 68.1 (CH2OCO), 63.3 (C),
62.3, 62.1 (CH2CO), 59.0, 51.7 (CH2cyclo), 28.3, 28.2 (CH3). ESI-
MS (m/z): 1408.78 (M + H+); calc for C68H114N9O22: 1408.81. IR
spectrum (KBr disk): 3331, 2973, 2942, 1729, 1549, 1370, 1212,
1158 cm-1.
[GdL2]. A solution of thiosphogene (31 mL; 0.40 mmol) in
dry CH2Cl2 (5 mL) was added dropwise in 2 min to a solution
of 9 (100 mg; ca. 0.08 mmol) in H2O (5 mL). The mixture was
then stirred for 2 h at rt. After adjustment of the pH to 5.5,
the water solution was washed 3 times with CH2Cl2 (5 mL) and
then evaporated in vacuo. ESI-MS (m/z): 1279.09 (M + H+); calc
for C38H46N9O19SGd2: 1279.11 (100,0%), (isotopic distribution
consistent with bis-Gd complex).
General procedure for octameric ligands
To a solution of the appropriate poly primary amine and triethy-
lamine (2 eq for each amino group) in dry CH2Cl2 (10 mN), under
N2 atmosphere, cooled at T £ 10 ◦C with an ice-bath, a solution
of 3 or 8 (1.2 eq for each amino group) in dry CH2Cl2 (2 mM) was
added dropwise. The reaction mixture was stirred overnight at rt
and then washed with water (3 ¥ 5 mL), HCl 0.1 mol L-1 (2 ¥ 5 mL)
and saturated NaHCO3 (2 ¥ 5 mL). The organic phase was dried
over NaSO4, filtered and evaporated to yield the crude product,
that was purified through crystallizations from diethyl ether with
petroleum ether.
7. 10% Pd/C (50 mg) suspended in water (1 mL) was added to
a solution of 6 (472 mg; 0.33 mmol) in MeOH (10 mL). The
mixture was stirred overnight under H2 atmosphere and then
filtered. The filtrate was evaporated in vacuo affording pure 7
(430 mg; 93%). TLC (silica gel 60 F254, petroleum ether/EtOAc
8 : 2, detection: UV 254 nm): Rf 0.05. 1H-NMR (CDCl3) 400 MHz
d = 7.12 (bs, 2H, NH), 6.79 (bs, 1H, CH), 6.61 (bs, 2H, CH), 4.14
(s, 4H, CH2OCO), 3.66 (s, 8H, CH2CO), 3.22 (s, 8H, CH2CO),
3.04 (d, 4H, J = 14.3 Hz, CH2C), 2.74 (d, 4H, J = 14.3 Hz, CH2C),
2.87–2.61 (m, 8H, CH2CH2), 1.37 (s, 36H, CH3), 1.36 (s, 36H,
CH3).13C-NMR (CDCl3) 100 MHz d = 172.8, 170.6 (CO), 153.2
(NHCOO), 148.4 (C), 139.5 (2C), 100.1 (2CH), 98.7 (CH), 81.0,
80.8 (CCH3), 67.5 (CH2OCO), 63.1 (C), 62.1–62.0 (CH2CO), 58.7,
51.4 (CH2cyclo), 28.2, 28.1 (CH3). ESI-MS (m/z): 1379.43 (M + H+);
calc for C68H116N9O20: 1378.83. IR spectrum (KBr disk): 3360,
2965, 1745, 1513, 1208, 1124 cm-1.
10. (yield 81%, calculated on polyamine). TLC (silica gel 60
F254, petroleum ether/EtOAc 1 : 1, detection: UV 254 nm): Rf 0.22.
1H-NMR (CDCl3) 400 MHz d = 7.36 (bs, 32H, CH), 4.20 (s, 16H,
CH2OCO), 3.84 (bs, 8H, CH2) 3.71 (s, 32H, CH2CO), 3.40 (bs,
16H, CH2), 3.26 (s, 32H, CH2CO), 3.09 (d, 16H, J = 13.9 Hz,
CH2C), 2.92 (bs, 8H, CH2), 2.83–2.63 (m, 32H, CH2cyclo), 2.64 (bs,
20H, CH2), 2.54 (bs, 24H, CH2), 2.38 (bs, 24H, CH2), 1.42 (s, 288H,
CH3).13C-NMR (CDCl3) 100 MHz d = 173.6, 172.8, 170.9 (CO),
153.5 (NHCOO), 136.2, 133.0 (bs, C), 125.7, 119.1 (bs, CH), 81.0–
80.8 (CCH3), 67.7 (CH2OCO), 63.3 (C), 62.3 (CH2CO), 59.0–51.6
(CH2cyclo), 55.2, 51.7, 50.4, 44.3, 39.1, 37.6, 34.0 (CH2), 28.3–28.2
(CH3). IR spectrum (KBr disk) 3391, 3241, 3002, 1746, 1641,
1522, 1356, 1221, 1148, 1073 cm-1. Elemental analysis calcd (%)
for C366H600N66O92S8: C, 57.44; H, 7.90; N, 12.08; S, 3.35. found:
C, 57.10; H, 7.75; N, 11.89; S, 3.21.
8. A solution of thiosphogene (31 mL; 0.4 mmol) in dry CH2Cl2
(2 mL) was added drop wise in 2 min to a ice-bath cooled mixture
of 7 (430 mg; 0.31 mmol) in saturated KHCO3 (5 mL) and
CH2Cl2 (5 mL), under vigorous magnetic stirring. The ice-bath
was removed after 5 min and the suspension was stirred for 2 h
at rt. The organic phase was separated and extracted with water
(3 ¥ 10 mL); then the organic phase was dried over NaSO4, filtered
and evaporated to yield a yellow solid. TLC (silica gel 60 F254
,
L3. HPLC, method 2, retention time 13.90 min, purity 85%.
1H-NMR (D2O) 400 MHz d = 7.36 (bs, 16H, CH), 7.17 (bs, 16H,
CH), 4.10 (s, 16H, CH2OCO), 3.68 (bs, 16H, CH2) 3.51 (s, 64H,
petroleum ether/EtOAc 8 : 2, detection: UV 254 nm): Rf 0.29. 1H-
NMR (CDCl3) 400 MHz d = 7.41 (bs, 2H, CH), 7.20 (bs, 1H, CH),
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 4569–4574 | 4573
©

View Article Online
CH2CO), 3.46 (bs, 16H, CH2), 3.27 (bs, 64H, CH2cyclo), 3.05 (bs,
32H, CH2), 2.27 (bs, 12H, CH2), 2.54 (bs, 24H, CH2). 13C-NMR
(D2O) 100 MHz d = 180.0, 175.2, 173.9 (CO), 154.8 (NHCOO),
136.7, 132.4 (bs, C), 126.8, 120.7 (bs, CH), 67.2 (CH2OCO), 63.2
(C), 60.8 (CH2CO), 58.1, 53.4 (CH2cyclo), 52.7, 49.8, 44.0, 39.1,
36.1, 31.4 (CH2). IR spectrum (KBr disk) 3259, 2897, 1739, 1633,
1527, 1389, 1293, 1129, 1048 cm-1. Elemental analysis calcd (%)
for C238H325Cl5N66Na24O92S8: C, 43.52; H, 4.99; N, 14.07; S, 3.91.
found: C, 43.35; H, 4.85; N, 13.97; S, 3.82.
References
1 V. Kubicek and E. Toth, Adv. Inorg. Chem., 2009, 61, 63; C. F. G. C.
Geraldes and S. Laurent, Contrast Media Mol. Imaging, 2009, 4, 1.
2 S. Aime, D. Delli Castelli, S. Geninatti Crich, E. Gianolio and E.
Terreno, Acc. Chem. Res., 2009, 42, 822; S. Aime, S. Geninatti Crich,
E. Gianolio, G. B. Giovenzana, L. Tei and E. Terreno, Coord. Chem.
Rev., 2006, 250, 1562.
3 D. Chen, V. Milacic, M. Frezza and Q. P. Dou, Curr. Pharm. Des., 2009,
15, 777; M. J. Heeg and S. S. Jurisson, Acc. Chem. Res., 1999, 32, 1053.
4 S. V. Eliseevaa and J.-C. G. Bu¨nzli, Chem. Soc. Rev., 2010, 39, 189; L.
Armelao, S. Quici, F. Barigelletti, G. Accorsi, G. Bottaro, M. Cavazzini
and E. Tondello, Coord. Chem. Rev., 2010, 254, 487.
5 G. T. Hermanson, Bioconjugate Techniques, Academic Press, 1996.
6 M. Woods and A. D. Sherry, Chemistry Today, 2005, 23, 31; M. L. De
Leo´n Rodriguez and Z. Kovacs, Bioconjugate Chem., 2008, 19, 391; A.
Barge, L. Tei, D. Upadhyaya, F. Fedeli, L. Beltrami, R. Stefan`ıa, S.
Aime and G. Cravotto, Org. Biomol. Chem., 2008, 6, 1176.
7 S. Liu, Adv. Drug Delivery Rev., 2008, 60, 1347; J. Fichna and A.
Janecka, Bioconjugate Chem., 2003, 14, 3.
8 P. Caravan, J. J. Ellison, T. J. McMurry and R. B. Lauffer, Chem. Rev.,
1999, 99, 2293; S. Aime, M. Botta and E. Terreno, Adv. Inorg. Chem.,
2005, 57, 173.
9 P. Caravan, C. T. Farrara, L. Frullano and U. Ritika, Contrast Media
Mol. Imaging, 2009, 4, 89; Y. Song, E. K. Kohlmeir and T. J. Meade,
J. Am. Chem. Soc., 2008, 130, 6662.
10 J. B. Livramento, E. Toth, A. Sour, A. Borel, A. E. Merbach and R.
Ruloff, Angew. Chem., Int. Ed., 2005, 44, 1480; J. B. Livramento, A.
Sour, A. Borel, A. E. Merbach and E. Toth, Chem.–Eur. J., 2006, 12,
989.
11 S. Aime, L. Calabi, C. Cavallotti, E. Gianolio, G. B. Giovenzana, P.
Losi, A. Maiocchi, G. Palmisano and M. Sisti, Inorg. Chem., 2004, 43,
7588; G. B. Giovenzana, G. Palmisano, M. Sisti, C. Cavallotti, S. Aime
and L. Calabi, PCT Int. Appl., 2003, WO0308390.
12 G. Gugliotta, M. Botta, G. B. Giovenzana and L. Tei, Bioorg. Med.
Chem. Lett., 2009, 19, 3442.
11. (yield 24%, calculated on polyamine). TLC (silica gel 60
F254, petroleum ether/EtOAc 5 : 5, detection: UV 254 nm): Rf 0.20.
1H-NMR (CDCl3) 400 MHz d = 7.78 (bs, 12H, CH), 4.19 (s, 16H,
CH2OCO), 3.98 (bs, 2H, CH2NHCS), 3.71 (s, 32H, CH2CO), 3.26
(s, 32H, CH2CO), 3.09 (d, 16H, J = 13.9 Hz, CH2C), 2.77–2.63
(m, 48H, CH2cyclo), 2.16 (bs, 20H, CH2N), 1.42 (s, 288H, CH3).13C-
NMR (CDCl3) 100 MHz d = 172.8, 170.9 (CO), 153.3 (NHCOO),
140.1 (bs, C), 120.1 (bs, CH), 80.9, 80.7 (CCH3), 67.9 (CH2OCO),
63.3 (C), 62.3 (CH2CO), 59.0–51.6 (CH2cyclo), 51.8 (CH2NHCS),
62.3, 53.5, 41.8 (CH2), 28.3, 28.2 (CH3). IR spectrum (KBr disk):
3357, 2984, 1710, 1627, 1422, 1220, 1073 cm-1. Elemental analysis
calcd (%) for C286H480N42O80S4: C, 58.07; H, 8.18; N, 9.94; S, 2.17.
found: C, 57.90; H, 8.03; N, 9.87; S, 2.09.
L4. HPLC, method 2, retention time 14.77 min, purity 81%.
1H-NMR (D2O) 400 MHz d = 7.63–7.42 (bs, 12H, CH), 4.37 (s,
16H, CH2OCO), 3.60 (s, 32H, CH2CO), 3.43 (s, 32H, CH2CO),
3.09–2.84 (m, 64H, CH2cyclo), 2,18, 1.95 (bs, 20H, CH2N). 13C-
NMR (D2O) 100 MHz d = 180.6, 179.4 (CO), 155.1 (NHCOO),
139.7 (bs, C), 103.5 (bs, CH), 66.2 (CH2OCO), 64.3 (C), 63.0
(CH2CO), 60.6, 56.6 (CH2cyclo), 51.8 (CH2NHCS), 64.3, 57.3, 36.6
(CH2). IR spectrum (KBr disk) 3418, 1618, 1410, 1218, 1070 cm-1.
Elemental analysis calcd (%) for C158H201ClN42Na24O80S4: C, 40.51;
H, 4.33; N, 12.56; S, 2.74. found: C, 40.39; H, 4.25; N, 12.44; S,
2.63.
13 Z. Baranyai, F. Uggeri, G. B. Giovenzana, A. Benyei, E. Brucher and
S. Aime, Chem.–Eur. J., 2009, 15, 1696.
14 M. Yoneno, S. Abe and K. Namba, Kogyo Kagaku Zasshi, 1967, 70,
1442; V. D. Jadhav and F. P. Schmidtchen, J. Org. Chem., 2008, 73,
1077.
15 L. Tei, G. Gugliotta, S. Avedano, G. B. Giovenzana and M. Botta, Org.
Biomol. Chem., 2009, 7, 4406; Z. Zhang, J. C. Amedio, P. D. Caravan,
S. Dumas, A. Kolodziej and T. J. McMurry, US Patent 2003/0180222.
16 K. N. Raymond, T. M. Corneillie, J. Xu, US Patent 2008/0213917.
17 S. Langereis, A. Dirksen, T. M. Hackeng, M. H. P. van Genderen and E.
W. Meijer, New J. Chem, 2007, 31, 1152; V. J. Venditto, A. I. S. Regino
and M. V. Brechbiel, Mol. Pharmaceutics, 2005, 2, 302.
Acknowledgements
This research was supported by funding from Regione Piemonte
(PIIMDMT and Nano IGT projects). ESF COST Action D38 is
also acknowledged.
18 G. M. Nicolle, E. Toth, H. Schmitt-Willich, B. Radu¨chel and A. E.
Merbach, Chem. Eur. J., 2002, 8, 1040.
4574 | Org. Biomol. Chem., 2010, 8, 4569–4574
This journal is
The Royal Society of Chemistry 2010
©