Synthesis of a novel amphiphilic GdPCTA-[12] derivative as a potential micellar MRI contrast agent

Article abstract of DOI:10.1016/j.tetlet.2008.07.160

A novel amphiphilic contrast agent, a GdPCTA-[12] derivative containing a dodecyl chain as lipophilic moiety, has been prepared. A convergent synthetic route from commercially available diethylene triamine and 3-hydroxypyridine is described. The target amphiphilic gadolinium complex was obtained in nine steps in 22% overall yield. Physicochemical properties and relaxivity measurements of this new contrast agent are described.

Full text of DOI:10.1016/j.tetlet.2008.07.160

Tetrahedron Letters 49 (2008) 5972–5975
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Synthesis of a novel amphiphilic GdPCTA-[12] derivative as a potential
micellar MRI contrast agent
Clotilde Ferroud a, , Hélène Borderies a,b, Elizabeth Lasri b, Alain Guy a, Marc Port b
*
a Laboratoire de Transformations Chimiques et Pharmaceutiques, UMR CNRS 7084, Conservatoire National des arts et métiers, Case 303, 2 rue Conté, 75003 Paris, France
b Guerbet, Centre de Recherche, BP 57400, 95943 Roissy CdG Cedex, France
a r t i c l e i n f o
a b s t r a c t
Article history:
A novel amphiphilic contrast agent, a GdPCTA-[12] derivative containing a dodecyl chain as lipophilic
moiety, has been prepared. A convergent synthetic route from commercially available diethylene tri-
amine and 3-hydroxypyridine is described. The target amphiphilic gadolinium complex was obtained
in nine steps in 22% overall yield. Physicochemical properties and relaxivity measurements of this new
contrast agent are described.
Received 30 June 2008
Revised 25 July 2008
Accepted 29 July 2008
Available online 31 July 2008
Ó 2008 Elsevier Ltd. All rights reserved.
Keywords:
GdPCTA-[12] derivative
Macrocyclic ligand
Gadolinium
Contrast agent
MRI
Amphiphilic
Micelles
The majority of paramagnetic contrast agents (CA) used for
diagnostic magnetic resonance imaging (MRI) comprise a para-
magnetic metal core, typically gadolinium(III), which is complexed
to a chelating ligand. In order to be considered as a potential CA, a
Gd complex must have a high thermodynamic stability and a
kinetic inertness, as free gadolinium ion is highly toxic for humans
even at low doses.1
For this purpose, an efficient approach is to slow the rotational
motion of the Gd-chelate and to increase the number of inner
sphere water molecules by designing heptadentate chelators. A
common approach to slow rotational motion has been to attach
the GdIII complex to a slowly tumbling macromolecule such as a
dendrimer,2 linear polymer,3 or protein4 or by supramolecular
aggregation of amphiphilic complexes.5,6
The high affinity of Gd (III) for a number of polyaminocarboxylic
acids, either cyclic or linear, has been exploited to form very stable
complexes (up to logKML >20), which are developed for clinical
applications.
Based on this rational drug design, we synthesized a new hep-
tacoordinate gadolinium complex Gd–PCTA-[12] (3,6,9,15-tetraaz-
abicyclo[9.3.1]pentadeca-1(15),11,13-triene-3,6,9-triacetate) able
to coordinate two inner sphere water molecules and containing a
dodecyl chain as lipophilic moiety in order to self-assemble into
micelles.
We have previously reported the synthesis of 12-membered
azapyridinomacrocycle PCTA 12 17 and a rigidified [PCTA-12]
derivative 2.8 In this Letter, we describe a convergent synthetic
route to a novel amphiphilic [PCTA-12] derivative 3 via a reaction
sequence based on macrocyclization involving tri-N-alkylated tri-
amine block (Scheme 1).
Contrast enhancement is due to the ability of the paramagnetic
Gd3+ cation to shorten the longitudinal (T1) and transverse (T2)
relaxation times of water protons in the surrounding tissues. The
effectiveness of gadolinium chelates as MRI contrast agents is usu-
ally assessed in vitro by measuring the corresponding relaxivities
r1 and r2, defined as the longitudinal and transverse relaxation
rates, respectively, for a millimolar solution of Gd complex.
The commercial CAs routinely used for clinical diagnosis have
longitudinal relaxivities (r1) ranging from 3.5 to 5 mMÀ1 sÀ1 1
.
The lipophilic moiety is introduced into the 3-position of the
pyridine ring via a lipophilic amide function. The 3-position seems
to be sufficiently far away to limit any interference with the coor-
dination sphere of Gadolinium.9
Although these substances are widely used in clinical applications,
there is still a need to develop new compounds with improved per-
formances in terms of relaxivity.
The synthesis started with commercially available diethylene-
triamine 4, which was selectively dinosylated to form disulfon-
amide 5 in 74% yield by spontaneous crystallization from the
* Corresponding author. Tel.: +33 0 140272402; fax: +33 0 142710534.
E-mail address: clotilde.ferroud@cnam.fr (C. Ferroud).
0040-4039/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2008.07.160

C. Ferroud et al. / Tetrahedron Letters 49 (2008) 5972–5975
5973
PCTA-12
Cyclo PCTA-12
lipophilic PCTA-12
O
O
N
11
H
N
Gd3+
N
Gd3+
N
Gd3+
-OOC
-OOC
-OOC
N
N
N
N
N
N
n
n
COO-
COO-
COO-
N
N
N
COO-
-OOC
-OOC
3
1
2
n = 1, 2
Scheme 1.
reaction medium (Scheme 2) The nosyl group (2-nitrobenzenesulf-
onamide) plays several important roles in this reaction. In fact, it
allows the monoalkylation of either primary or secondary amine
functions of compound 4 to give the compound 7a after the selec-
tive protection of the primary amine functions. During the alkyl-
ation process, owing to the increased acidity, the deprotonation
of the remaining N–H bond of the sulfonamido groups is easily
accomplished even with relatively weak bases. On the other hand,
in the Richman–Atkins cyclization reaction, the bulky nature of
this group is, most likely, involved in a preorganization of the
intermediates, which favours the transition state leading to the
intramolecular cyclization, decreasing the importance of the alter-
native intermolecular oligomerization processes. Finally, the nosyl
group can be removed under mild conditions with soft
nucleophiles.10
The reaction of 1,7-dinosyl-1,4,7-triazaheptane 5 with tert-
butyl bromoacetate in refluxing acetonitrile and in the presence
of potassium carbonate led quantitatively to the functionalized
diprotected compound 6, which was pure enough to be used in
the next step without prior purification. The nosyl groups were
removed by treating the crude compound 6 with thiophenol in
the presence of sodium carbonate in dimethylformamide at 50 °C
for three hours. After purification, the expected functionalized
triamine 7a was isolated in 80% yield. Note that the lactamized
product 7b was preferentially formed under the hardest condi-
tions, i.e. longer reaction times at higher temperatures or more
reactive conditions (mercaptoethanol/DBU/1 h/rt/CH3CN).
O
COOEt
OH
a
N
9
N
8
OH
OH
b
O
COOEt
N
Br
Br
10a
Scheme 3. Reagents and conditions: (a) (i) aqueous formaldehyde 37%, NaOH, H2O,
90 °C, 6 h; (ii) ethyl bromoacetate, CH3CN–H2O, 80 °C, 2 h, 40%; (b) PPh3, CBr4,
CH3CN, rt, 4 h, 76%.
crystallization (MeOH) of the corresponding hydrochloride.9 Regio-
selective O-alkylation at the phenol hydroxyl group was then per-
formed in a one-pot reaction with ethylbromoacetate in a mixture
of acetonitrile/water (Scheme 3). After purification, the resulting
compound 9 was obtained in 40% overall yield. The functionalized
pyridine 9 was then treated with triphenylphosphine and carbon
tetrabromide in anhydrous acetonitrile at room temperature to
give the dibromide derivative 10a in 76% yield.
The macrocyclization reaction between bis(bromomethyl)pyr-
idine 10a and the functionalized triamine 7 was carried out in
the presence of a base in dimethylformamide at 80 °C for 3 h
(Scheme 4). Various parameters were studied to optimize the yield
of the reaction (Table 1).
The reaction sequence adopted for the other building block is
given in Scheme 3. 3-Hydroxypyridine 8 as treated with aqueous
formaldehyde and sodium hydroxide to form the expected 2,6-
dihydroxymethyl-3-hydroxypyridine, which was purified by
H
H
N
N
a
H2N
NH2
NsHN
NHNs
5
4
b
tBuOOC
N
6
tBuOOC
N
N
COOtBu
Ns
Ns
c
O
N
tBuOOC
tBuOOC
N
tBuOOC
N
H
N
COOtBu
N
N
COOtBu
H
H
7a
7b
Scheme 2. Reagents and conditions: Ns = 2-NO2–C6H4 (a) (i) NaOH, H2O, rt; (ii) 2-nitrobenzenesulfonyl chloride, THF–Et2O (1:6), rt, overnight, 74%; (b) tert-butyl
bromoacetate, K2CO3, CH3CN, reflux, 2 h, 100%; (c) thiophenol, Na2CO3, DMF, 50 °C, 3 h, 80%.

5974
C. Ferroud et al. / Tetrahedron Letters 49 (2008) 5972–5975
O
O
O
O
N
11
OEt
H
N
N
tBuOOC
N
N
tBuOOC
b
a
N
N
N
N
7a
10a
+
COOtBu
COOtBu
tBuOOC
12
tBuOOC
11
c
O
O
O
O
N
N
H
11
N
H
11
N
Gd3+
-OOC
N
N
HOOC
d
N
N
N
COO-
N
COOH
-OOC
13
3
HOOC
Scheme 4. Reagents and conditions: (a) Na2CO3, DMF, 80 °C, 3 h, 87%; (b) AlMe3, dodecylamine, toluene, 50 °C, overnight, 87%; (c) HCl 2 M in Et2O, CH2Cl2, rt, 24 h, 100%; (d)
Gd2O3, H2O, 80 °C, 3 h, 50%.
Table 1
product by chromatography on silica gel must be performed.
Secondly, dodecylamine was chemoselectively coupled with mac-
rocycle 11 in the presence of DIBAL-H.11 Impurities were formed
in addition to the expected product, and, as previously, a purifica-
tion was therefore required. Lastly, we applied a procedure previ-
ously described by Weinreb12 in which trimethylaluminium is
used as a condensing agent. The preactivation of one equivalent
of dodecylamine with one equivalent of AlMe3 was followed by
addition of macrocycle 11. The reaction was performed in toluene
at 50 °C overnight and was followed by a basic treatment. The
effective complexation of both the amine function and the ester
function by alkylaluminium13 led chemoselectively to the pure
amine derivative 12 in 87% yield. Moreover, these treatments were
simplified, as no byproduct is formed.
Entry
Substrate concentration
in DMFa (M)
Baseb
Yield of 8 (%)
1
2
3
4
5
0.05
0.10
0.10
0.10
0.10
K2CO3
K2CO3
Cs2CO3
Na2CO3
Li2CO3
25–33
48–54
31
65 (87c)
48
a
All reactions were performed in DMF at 80 °C for 3 h using 0.77 mmol of
substrates introduced at rt.
b
Four equivalents of base
Isolated in large scale (up to 13 mmol).
c
First, we studied the influence of the substrate concentration.
The macrocyclization reaction was performed with 0.1 and
0.05 M solutions of substrates in the presence of potassium car-
bonate (Table 1, entries 1 and 2). The best yield was obtained for
a 0.1 M solution.
Second, given that the dibromide derivative 10a is sensitive to
temperature, it was introduced by two different methods, at room
temperature just before heating or dropwise at 80 °C. No difference
in yield was observed between these two methods.
The role of the nature of the base-counterion was then exam-
ined during nucleophilic displacement of the halide in the ring clo-
sure step. The results (Table 1 entries 2–5) show a template effect
from Na+. After optimization, the macrocyclization reaction was
performed with a 0.1 M solution of substrates in DMF at 80 °C
and in the presence of sodium carbonate and gave compound 11
in large scale (up to 13 mmol) in 87% yield.
It is to be noticed that when the dibromo derivative 10a was
replaced by the corresponding dichloro compound 10 b in the
same conditions as entry 2, compound 11 was obtained in less than
15% yield versus 48–54%.
Finally, ligand 13 was obtained quantitatively by treatment of
tri-tert-butyl ester 12 in dichloromethane with a 2 M ether solu-
tion of hydrochloric acid at room temperature.
The corresponding Gd complex 314 was obtained, after precipi-
tation from the reaction, by heating the ligand with gadolinium
oxide (Gd2O3) in water at 80 °C while maintaining pH between
5.2 and 5.5 in 50% yield. The electrospray MS data confirmed the
presence of the Gd complex. The presence of free gadolinium(III)
ions was checked by the usual Arsenazo test, and was estimated
to be 0.4 mol %/mol.
The micellar concentration of Gd–PCTA-[12] derivative 3 was
determined at 0.19 mM by plotting the T1 relaxation rate
(20 MHz, 37 °C) as a function of Gd–PCTA-[12] derivative 3
concentration.
The r1-relaxivity value obtained in aqueous solutions at 37 °C
and 20 MHz for the new micellar Gd–PCTA-[12] derivative 3 is
higher than that for the previously reported micellar Gd(DOTA-
C14) complex6a (33 vs 18.8 mMÀ1 sÀ1) under the same conditions
and for other DOTA, DTPA and PCTA amphiphilic derivatives cited
in Table 2. Most likely, the number of H2O molecules on the Gd3+ is
responsible for the increased relaxivity compared to the Gd(DOTA-
C14) complex. This new contrast agent has higher r1-relaxivity
than the Gd(PCTA-O-C12) complex10a (33 vs 28.5 mMÀ1 sÀ1, i.e.,
16% enhancement). A possible explanation would be related to
The last step of synthesis involves the formation of the amide
linkage between dodecylamine and the macrocycle derivative 11.
Firstly, by using conventional conditions in the presence of
N-hydroxysuccinimide/EDCI, a tedious purification of the crude

C. Ferroud et al. / Tetrahedron Letters 49 (2008) 5972–5975
5975
Table 2
5. (a) Glogard, C.; Stensrud, G.; Hovland, R.; Fossheim, S. L.; Klaveness, J. Int. J.
Pharm. 2002, 233, 131–140; (b) Masotti, A.; Mangiola, A.; Sabatino, G.; Maira,
G.; Denaro, L.; Conti, F.; Ortaggi, G.; Capuani, G. Int. J. Immunopathol. Pharmacol.
2006, 19, 379–390.
Comparison of r1-relaxivities measured at 20 MHz for various amphiphilic Gd
complexes
Gd complexes
r1-relaxivities
6. (a) Glogard, C.; Hovland, R.; Fossheim, S. L.; Aasen, A. J.; Klaveness, J. J. Chem.
Soc., Perkin Trans. 2 2000, 1047; (b) Hovland, R.; Glogard, C.; Aasen, A. J.;
Klaveness, J. J. Chem. Soc., Perkin Trans. 2 2001, 929–933; (c) Muller, R. N.;
Vander Elst, L.; Roch, A.; Peters, J. A.; Csajbok, E.; Gillis, P.; Gossuin, Y. Advances
in Inorganic Chemistry, Elsevier Ed., Oxford, 2005, Vol 57, pp 239-289; (d)
Masotti, A.; Remollino, L.; Carafa, M.; Marianecci, C.; Santucci, E.; Ortaggi, G.
Synlett 2006, 2815–2817; (e) Nicolle, G. M.; Toth, E.; Eisenwiener, K-P.; Mäcke,
H. R.; Merbach, A. E. J. Biol. Inorg. Chem. 2002, 7, 757–769.
7. (a) Siaugue, J. M.; Segat-Dioury, F.; Favre-Réguillon, A.; Madic, C.; Foos, J.; Guy,
A. Tetrahedron Lett. 2000, 41, 7443–7446; (b) Siaugue, J. M.; Segat-Dioury, F.;
Sylvestre, I.; Favre-Réguillon, A.; Foos, J.; Madic, C.; Guy, A. Tetrahedron 2001,
57, 4713–4718.
(mMÀ1 sÀ1), 37 °C
Gd(PCTA–CH2CONH–C12) 3
Gd(PCTA–O–C12)10a
33
28.5
18.8
Gd(DOTA–C14)6a
Gd(DTPA–CONH–C12) andGd(DTPA–CONH–C14)
andGd(DTPA–CONH–C16) andGd(DTPA–
CONH–C18)15
<15 from NMRD profiles
Gd(DTPA–bisCONH–C14) andGd(DTPA–
<14 from NMRD profiles
<10 from NMRD profile
bisCONH–C16)16
Gd(DTPA–bisCONH–C18)16
8. (a) Dioury, F.; Guéné, E.; Di Scala-Roulleau, A.; Ferroud, C.; Guy, A.; Port, M.
Tetrahedron Lett. 2005, 46, 611–613; (b) Dioury, F.; Sambou, S.; Guéné, E.;
Sabatou, M.; Ferroud, C.; Guy, A.; Port, M. Tetrahedron 2007, 63, 204–214.
9. Aime, S.; Botta, M.; Frullano, L.; Crich, S. G.; Giovenzana, G. B.; Pagliarin, R.;
Palmisano, G.; Sisti, M. Chem. Eur. J. 1999, 5, 1253–1260.
10. (a) Hovland, R.; Glogard, C.; Aasen, A. J.; Klaveness, J. Org. Biomol. Chem. 2003, 1,
644–647; (b) Miller, S. C.; Scanlan, T. S. J. Am. Chem. Soc. 1997, 119, 2301–2302.
11. Huang, P.-Q.; Zheng, X.; Deng, X.-M. Tetrahedron Lett. 2001, 42, 9039–9041.
12. Basha, A.; Lipton, M.; Weinreb, S. M. Tetrahedron Lett. 1977, 18, 4171–4172.
13. (a) Sidler, D. R.; Lovelace, T. C.; McNamara, J. M.; Reider, P. J. J. Org. Chem. 1994,
59, 1231–1233; (b) Alvarez-Ibarra, C.; G Csákÿ, A.; Gómez de la Oliva, O.;
Rodrıguez, E. Tetrahedron Lett. 2001, 42, 2129–2131.
the secondary amide function of the lipophilic side arm that can
form hydrogen bonds between monomers, resulting in more rigid-
ified micelles. This would be responsible for enhancement of the
local rotational correlation time and for an increased relaxivity of
the new Gd complex.
In conclusion, the amphiphilic GdPCTA-[12] derivative 3 was
synthesized in nine steps from commercially available diethylene
triamine 4 and 3-hydroxypyridine 8 in 22% overall yield. In
comparison, Hovland et al.10a have described the synthesis of a
derivative, bearing a dodecylamine chain directly connected to
the aromatic hydroxyl on the pyridine, in six steps from two
commercially unavailable compounds in 3% overall yield.
The physicochemical properties of compound 3 are being
further investigated, and will be published shortly.
14. Analytical data for all compounds: Compound 5: 1H NMR (400 MHz, DMSO-d6):
d 2.52 (t, J = 6.5 Hz, 4H), 2.92 (t, J = 6.5 Hz, 4H), 7.86–7.91 (m, 4H), 7.97–8.00
(m, 2 H), 8.00–8.07 (m, 2H); 13C (100 MHz, DMSO-d6): d 42.6, 47.8, 124.4,
129.5, 132.6, 132.7, 134.0, 147.7. Compound 6: 1H NMR (400 MHz, CDCl3): d
1.34 (s, 18H), 1.44 (s, 9H), 2.86 and 3.45 (2 Â t, J = 6.8 Hz, 8H), 3.24 (s, 2H), 4.15
(s, 4H), 7.51–7.61 (m, 2H), 7.62–7.72 (m, 4H), 8.03–8.11 (m, 2H); 13C (100 MHz,
CDCl3): d 27.9, 28.2, 46.8, 49.5, 53.2, 56.2, 81.5, 82.3, 124.0, 131.0, 131.9, 133.3,
133.5, 148.0, 167.8, 170.3; HRMS (ES+) calcd for C34H50N5O14S2 + H+, 816.2796;
found, 816.2787. Compound 7: 1H NMR (400 MHz, CDCl3): d 1.39 (s, 9H), 1.41
(s, 18H), 2.61–2.65 (m, 4H), 2.75–2.79 (m, 4H), 3.27 and 3.29 (s, 6H); 13C
(100 MHz, CDCl3):
d 28.1, 47.2, 51.2, 171.0, 171.1, 81.0, 81.3, 53.6, 55.8.
Compound 9: 1H NMR (400 MHz, CDCl3): d 1.23 (t, J = 7.2 Hz, 3H), 4.19 (q,
J = 7.2 Hz, 2H), 4.60, 4.61 and 4.73 (3s, 6H), 7.00 and 7.15 (2d, J = 8.5 Hz, 2H);
13C (100 MHz, CDCl3): d 14.1, 60.2, 61.7, 64.2, 65.5, 168.4, 148.2, 150.0, 151.3,
119.2, 119.9. Compound 10a: 1H NMR (400 MHz, CDCl3): d 1.30 (t, J = 7.2 Hz,
3H), 4.27 (q, J = 7.2 Hz, 2H), 4.52 and 4.67 (2s, 4H), 4.74 (s, 2H), 7.06 and 7.09 (2
d, J = 8.6 Hz, 2H); 13C (100 MHz, CDCl3): d 14.2, 28.8, 33.3, 61.8, 65.8, 120.2,
124.5, 146.7, 149.3, 151.7, 167.9. Compound 11: 1H NMR (400 MHz, CDCl3): d
1.27 (t, J = 7.0 Hz, 3 H), 1.42 (s, 9H), 1.46 (s, 9H), 1.47 (s, 9H), 1.99–2.22 (m, 4H),
2.48–2.67 (m, 4H), 3.11 (2d, J = 17.6 Hz, AB, 2H), 3.32 and 3.36 (2d, J = 3.0 Hz,
AB, 2H), 3.42 and 3.47 (2 d, J = 7.0 Hz, AB, 2H), 3.68 and 3.93 (2d, J = 14.6 Hz,
AB, 2H), 3.74 and 4.13 (2d, J = 15.8 Hz, AB, 2H), 4.21 (q, J = 7.0 Hz, 2H), 4.67 and
4.72 (2d, J = 16.6 Hz, AB, 2H), 7.16 and 7.20 (2d, J = 8.5 Hz, 2H); 13C (100 MHz,
CDCl3): d 13.7, 27.5, 52.7, 53.2, 53.3, 53.5, 55.9, 56.0, 58.9, 59.3, 61.1, 61.0, 65.1,
82.2, 82.37, 82.42, 120.1, 121.8, 147.1, 149.1, 150.6, 167.7, 172.5, 172.6, 173.1;
LRMS (IC) calcd for C33H54N4O9 + Na+, 673; found, 673.47; LRMS (ES+) calcd for
Acknowledgement
Financial support from Laboratoires Guerbet is gratefully
acknowledged.
References and notes
1. (a) Aime, S.; Botta, M.; Fasano, M.; Terreno, E. Chem. Soc. Rev. 1998, 27, 19–29;
(b) Idee, J.-M.; Port, M.; Raynal, I.; Schaefer, M.; Le Greneur, S.; Corot, C. Fund.
Clin. Pharmacol. 2007, 206, 563–576; (c) Port, M.; Idée, J.-M.; Medina, C.; Robic,
C.; Sabatou, M.; Corot, C. Biometals 2008. doi:10.2007/s10534-008-9135-x.
2. (a) Toth, E.; Pubanz, D.; Vauthey, S.; Helm, L.; Merbach, A. E. Chem. Eur. J. 1996,
2, 1607–1615; (b) Krause, W.; Hackmann-Schlichter, N.; Maier, F. K.; Muller, R.
N. Top. Curr. Chem. 2000, 210, 261–308.
3. (a) Vexler, V. S.; Clement, O.; Schmitt-Willich, H.; Brasch, R. C. J. Magn. Res.
Imaging 1994, 4, 381–388; (b) Ladd, D. L.; Hollister, R.; Peng, X.; Wie, 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, 361–370; (c)
Corsi, D. M.; Vander Elst, L.; Muller, R. N.; van Bekkum, H.; Joop, A. P. Chem Eur.
J. 2001, 7, 64–71.
C
C
33H54N4O9 + H+, 651 found, 651.40. Compound 12: LRMS (IC) calcd for
43H75N5O8 + H+, 790; found, 790.44; LRMS (ES+) calcd for C43H75N5O8 + H+,
790; found, 790.51). Compound 13: LRMS (ES+) calcd for C31H51N5O8 + H+, 622;
found, 622.37. Compound 3: LRMS (ES+) calcd for C31H48GdN5O8 + H+, 776;
found, 776.00.
15. Parac-Vogt, T. N.; Kimpe, K.; Laurent, S.; Pierart, C.; Vander Elst, L.; Muller, R.
N.; Binnemans, K. Eur. J. Inorg. Chem. 2004, 17, 3538–3543.
16. Kimpe, K.; Parac-Vogt, T. N.; Laurent, S.; Pierart, C.; Vander Elst, L.; Muller, R.
N.; Binnemans, K. Eur. J. Inorg. Chem. 2003, 16, 3021–3027.
4. Toth, E.; Connac, F.; Helm, L.; Adzamli, K.; Merbach, A. E. J. Biol. Inorg. Chem.
1998, 3, 606–613.