Synthesis and physicochemical characterization of carbon backbone modified [Gd(TTDA)(H2O)]2- derivatives

Article abstract of DOI:10.1021/ic101799c

The present study was designed to exploit optimum lipophilicity and high water-exchange rate (k_ex) on low molecular weight Gd(III) complexs to generate high bound relaxivity (r₁^b), upon binding to the lipophilic site of human serum albumin (HSA). Two new carbon backbone modified TTDA (3,6,10-tri(carboxymethyl)-3,6,10-triazadodecanedioic acid) derivatives, CB-TTDA and Bz-CB-TTDA, were synthesized. The complexes [Gd(CB-TTDA)(H ₂O)]^2- and [Gd(Bz-CB-TTDA)(H₂O)]^2- both display high stability constant (log K_GdL = 20.28 and 20.09, respectively). Furthermore, CB-TTDA (log K_(Gd/Zn) = 4.22) and Bz-CB-TTDA (log K_(Gd/Zn) = 4.12) exhibit superior selectivity of Gd(III) against Zn(II) than those of TTDA (log K_(Gd/Zn) = 2.93), EPTPA-bz-NO₂ (log K_(Gd/Zn) = 3.19), and DTPA (log K _(Gd/Zn) = 3.76). However, the stability constant values of [Gd(CB-TTDA)(H₂O)]^2- and [Gd(Bz-CB-TTDA)(H ₂O)]^2- are lower than that of MS-325. The parameters that affect proton relaxivity have been determined in a combined variable temperature ¹⁷O NMR and NMRD study. The water exchange rates are comparable for the two complexes, 232 × 10⁶ s^-1 for [Gd(CB-TTDA)(H₂O)]^2- and 271 × 10⁶ s ^-1 for [Gd(Bz-CB-TTDA)(H₂O)]^2-. They are higher than those of [Gd(TTDA)(H₂O)]^2- (146 × 10 ⁶ s^-1), [Gd(DTPA)(H₂O)]^2- (4.1 × 10⁶ s^-1), and MS-325 (6.1 × 10⁶ s^-1). Elevated stability and water exchange rate indicate that the presence of cyclobutyl on the carbon backbone imparts rigidity and steric constraint to [Gd(CB-TTDA)(H₂O)]^2-and [Gd(Bz-CB-TTDA) (H₂O)]^2-. In addition, the major objective for selecting the cyclobutyl is to tune the lipophilicity of [Gd(Bz-CB-TTDA)(H ₂O)]^2-. The binding affinity of [Gd(Bz-CB-TTDA)(H ₂O)]^2- to HSA was evaluated by ultrafiltration study across a membrane with a 30 kDa MW cutoff, and the first three stepwise binding constants were determined by fitting the data to a stoichiometric model. The binding association constants (K_A) for [Gd(CB-TTDA)(H ₂O)]^2- and [Gd(Bz-CB-TTDA)(H₂O)]^2- are 1.1 × 10² and 1.5 × 10³, respectively. Although the K_A value for [Gd(Bz-CB-TTDA)(H₂O)] ^2- is lower than that of MS-325 (K_A = 3.0 × 10 ⁴), the r₁^b value, r₁^b = 66.7 mM^-1 s^-1 for [Gd(Bz-CB-TTDA)(H₂O)] ^2-, is significantly higher than that of MS-325 (r₁ ^b = 47.0 mM^-1 s^-1). As measured by the Zn(II) transmetalation process, the kinetic stabilities of [Gd(CB-TTDA)(H ₂O)]^2-, [Gd(Bz-CB-TTDA)(H₂O)]^2-, and [Gd(DTPA)(H₂O)]^2- are similar and are significantly higher than that of [Gd(DTPA-BMA)(H₂O)]^2-. High thermodynamic and kinetic stability and optimized lipophilicity of [Gd(CB-TTDA)(H₂O)]^2- make it a favorable blood pool contrast agent for MRI.

Full text of DOI:10.1021/ic101799c

Inorg. Chem. 2011, 50, 1275–1287 1275
DOI: 10.1021/ic101799c
Synthesis and Physicochemical Characterization of Carbon Backbone Modified
[Gd(TTDA)(H2O)]2- Derivatives
Ya-Hui Chang,† Chiao-Yun Chen,‡,§ Gyan Singh,|| Hsing-Yin Chen,† Gin-Chung Liu,‡,§ Yih-Gang Goan,#,3 Silvio
Aime,O and Yun-Ming Wang*,||,^
†Department of Medicinal and Applied Chemistry, ‡Department of Medical Imaging, Kaohsiung Medical
University Hospital, §Department of Radiology, Kaohsiung Medical University, 100 Shih-Chuan first Road,
Kaohsiung 807, Taiwan, ||Department of Biological Science and Technology, and ^Institute of Molecular
Medicine and Bioengineering, National Chiao Tung University, 75 Bo-Ai Street, Hsinchu 300, Taiwan,
#Department of Surgery, Kaohsiung Veterans General Hospital, Kaohsiung, Taiwan, 3Department of Nursing,
Yuh-Ing Junior College of Health Care & Management, Kaohsiung, Taiwan, and ODepartment of Chemistry
IFM and Molecular Imaging Center, University of Torino, Torino, Italy
Received September 2, 2010
The present study was designed to exploit optimum lipophilicity and high water-exchange rate (kex) on low molecular weight
Gd(III) complexs to generate high bound relaxivity (r1b), upon binding to the lipophilic site of human serum albumin (HSA).
Two new carbon backbone modified TTDA (3,6,10-tri(carboxymethyl)-3,6,10-triazadodecanedioic acid) derivatives, CB-
TTDA and Bz-CB-TTDA, were synthesized. The complexes [Gd(CB-TTDA)(H2O)]2- and [Gd(Bz-CB-TTDA)(H2O)]2- both
display high stability constant (log KGdL = 20.28 and 20.09, respectively). Furthermore, CB-TTDA (log K(Gd/Zn) =4.22) and Bz-
CB-TTDA (log K(Gd/Zn) = 4.12) exhibit superior selectivity of Gd(III) against Zn(II) than those of TTDA (log K(Gd/Zn) = 2.93),
EPTPA-bz-NO2 (log K(Gd/Zn) = 3.19), and DTPA (log K(Gd/Zn) = 3.76). However, the stability constant values of [Gd(CB-
TTDA)(H2O)]2- and [Gd(Bz-CB-TTDA)(H2O)]2- are lower than that of MS-325. The parameters that affect proton relaxivity
have been determined in a combined variable temperature 17O NMR and NMRD study. The water exchange rates are
comparable for the two complexes, 232 ꢀ 106 s-1 for [Gd(CB-TTDA)(H2O)]2- and 271 ꢀ 106 s-1 for [Gd(Bz-CB-
TTDA)(H2O)]2-. They are higher than those of [Gd(TTDA)(H2O)]2- (146ꢀ106 s-1), [Gd(DTPA)(H2O)]2- (4.1ꢀ106 s-1),
and MS-325 (6.1ꢀ106 s-1). Elevated stability and water exchange rate indicate that the presence of cyclobutyl on the
carbon backbone imparts rigidity and steric constraint to [Gd(CB-TTDA)(H2O)]2-and [Gd(Bz-CB-TTDA)(H2O)]2-. In
addition, the major objective for selecting the cyclobutyl is to tune the lipophilicity of [Gd(Bz-CB-TTDA)(H2O)]2-. The binding
affinity of [Gd(Bz-CB-TTDA)(H2O)]2- to HSA was evaluated by ultrafiltration study across a membrane with a 30 kDa MW
cutoff, and the first three stepwise binding constants were determined by fitting the data to a stoichiometric model. The binding
association constants (KA) for [Gd(CB-TTDA)(H2O)]2- and [Gd(Bz-CB-TTDA)(H2O)]2- are 1.1 ꢀ 102 and 1.5 ꢀ 103,
respectively. Although the KA value for [Gd(Bz-CB-TTDA)(H2O)]2- is lower than that of MS-325 (KA = 3.0 ꢀ 104), the r1b
value, r1b = 66.7 mM-1 s-1 for [Gd(Bz-CB-TTDA)(H2O)]2-, is significantly higher than that of MS-325 (r1b = 47.0 mM-1 s-1).
As measured by the Zn(II) transmetalation process, the kinetic stabilities of [Gd(CB-TTDA)(H2O)]2-, [Gd(Bz-CB-
TTDA)(H2O)]2-, and [Gd(DTPA)(H2O)]2- are similar and are significantly higher than that of [Gd(DTPA-BMA)(H2O)]2-
.
High thermodynamic and kinetic stability and optimized lipophilicity of [Gd(CB-TTDA)(H2O)]2- make it a favorable blood pool
contrast agent for MRI.
Introduction
normal and abnormal tissue.1 The major disadvantage of
MRI is its inherent low sensitivity. To enhance the quality of
image, contrast agents (CAs) have often been used prior to
MR imaging, which serve as catalysts in shortening the
Magnetic resonance imaging (MRI) has proven invaluable
in the clinical imaging of several anatomical and physiologi-
cal abnormalities. The technique exploits in vivo water
present in tissue at ∼90 M to increase contrast between
(1) Caravan, P. Chem. Soc. Rev. 2006, 35, 512–523.
ꢁ
ꢀꢁ ꢀ
(2) (a) Polasek, M.; Sedinova, M.; Kotek, J.; Vander Elst, L.; Muller,
*To whom correspondence should be addressed. Mailing address:
Department of Biological Science and Technology, Institute of Molecular
Medicine and Bioengineering, National Chiao Tung University, 75 Bo-Ai
Street, Hsinchu 300, Taiwan. Telephone: 886-3-5721212 ext 56972. Fax: 886-
3-5729288. E-mail:ymwang@mail.nctu.edu.tw.
ꢁ
R. N.; Hermann, P.; Lukes, I. Inorg. Chem. 2009, 48, 455–465. (b) Frangioni,
J. V. Nat. Biotechnol. 2006, 24, 909–913. (c) Louie, A. Y.; Huber, M. M.; Ahrens,
E. T.; Rothbacher, U.; Moats, R.; Jacobs, R. E.; Fraser, S. E.; Meade, T. J. Nat.
Biotechnol. 2000, 18, 321–325.
r
2011 American Chemical Society
Published on Web 01/19/2011
pubs.acs.org/IC

1276 Inorganic Chemistry, Vol. 50, No. 4, 2011
Chang et al.
longitudinal and transverse relaxation time (i.e., T1 and T2)
of water protons.2 Primary clinical CAs, such as [Gd-
(DTPA)(H2O)]2- (Magnevist), [Gd(DTPA-BMA)(H2O)]
(Omniscan), and [Gd(DOTA)(H2O)]- (Dotarem), display
relaxivities of around 4-5 mM-1 s-1 in the range of magnetic
fields (0.5-1.5 T) used in clinical MRI at 37 ꢀC.3 However,
the Solomon-Bloembergen-Morgan (SBM) theory pre-
dicts an over 20-fold relaxivity enhancement for the Gd(III)
complex with single hydration state (q=1) at regular imag-
ing fields used for clinical imaging, upon the simultaneous
optimization of key parameters, such as the electronic relaxa-
tion rate (1/T1,2e), the water exchange rate (kex), and the
molecular rotational correlation time (τR).4,5 However, in
practice, the electronic relaxation rate (1/T1,2e) of the para-
magnetic center is still hard to alter.3
In the past few years many endeavors have been made to
develop hypersensitive CAs. In this front, CAs endowed with
much higher relaxivity,6 targeting specific organ and tissue,7
and behaving as smart CAs8 have been extensively explored.
Recently, special attention has been devoted to the optimiza-
tion of water exchange rate on Gd(III) complexes.9-11 The
inner-sphere water exchange rate has been shown to reach its
optimal value when steric compression is created around the
water binding site in the Gd(III) complex.12 Steric crowding
facilitates in squeezing out the bound water molecules in the
rate determining step and consequently accelerates the water
exchange. This phenomenon has been observed in both
acyclic13 and macrocyclic14 Gd(III) complexes. However,
for low molecular weight Gd(III) complexes, high water
exchange rate has only limited advantages since relaxivity is
mainly restricted by fast molecular motion.15 T1 relaxation is
optimum when the frequency of molecular motion is close to
the Larmor frequency. However, the small gadolinium com-
plexes are characterized by rotational correlation times rang-
ing between 50 and 100 ps, which corresponds to rotational
frequencies of 10-20 GHz in contrast to the optimal imaging
field strength for clinical imaging ranging from 0.5 to 1.5 T
which corresponds to Lamor frequencies of 20-65 MHz.16
Fortunately, it is well established from previous studies that
the metal complex fluctuates at frequency somewhat closer to
the proton Larmor frequency on slowing down rotational
diffusion and, consequently, relaxivity gain can be achieved.17
A systematic review identifies a number of strategies em-
ployed to design high molecular weight CAs. Small molec-
ular weight Gd(III) complexes can be covalently or non-
covalently bound to macromolecules, for instance, linear
polymers,18 dendrimers,19 viral capsids,20 proteins,21 polysa-
ccharides,22 DNA quadruplex scaffolds,23 self- assembled
peptide amphiphile nanofibers,24 and liposomes.25 However,
some of these approaches suffer from internal flexibility or
nonrigid attachment of ligand to macromolecule.26 Nonco-
valent interaction of small Gd(III) complexes to in vivo
proteins has drawn much attention due to diverse physiolo-
gical congenial reasons. This approach is often called recep-
tor-induced magnetization enhancement (RIME).27 The
commercial contrast agent MS-325 (Vasovist) is an example
of a rationally designed RIME CA for angiographic applica-
tion. It possesses a lipophilic diphenylcyclohexyl residue wh-
ich facilitates noncovalent interaction with human serum
albumin (HSA).28,29 The binding augments the blood plasma
circulation time and also slows down its tumbling rate,
leading to greater contrast enhancements of blood vessel
MR image.30 It is apparent from previous studies that
(16) Caravan, P.; Cloutier, N. J.; Greenfield, M. T.; McDermid, S. A.;
Dunham, S. U.; Bulte, J. W.; Amedio, J. C., Jr.; Looby, R. J.; Supkowski,
R. M.; Horrocks, W. D., Jr.; McMurry, T. J.; Lauffer, R. B. J. Am. Chem.
Soc. 2002, 124, 3152–3162.
(17) (a) Lauffer, R. B.; Brady, T. J. Magn. Reson. Imaging 1985, 3, 11–16.
(b) Lauffer, R. B.; Brady, T. J.; Brown, R. D., III; Baglin, C.; Koenig, S. H. Magn.
Reson. Med. 1986, 3, 541–548.
(18) (a) Lucas, R. L.; Benjamin, M.; Reineke, T. M. Bioconjugate Chem.
2008, 19, 24–27. (b) Zhang, G.; Zhang, R.; Wen, X.; Li, L.; Li, C. Biomacro-
molecules 2008, 9, 36–42. (c) Fu, Y.; Raatschen, H. J.; Nitecki, D. E.; Wendland,
M. F.; Novikov, V.; Fournier, L. S.; Cyran, C.; Rogut, V.; Shames, D. M.; Brasch,
R. C. Biomacromolecules 2007, 8, 1519–1529.
(19) Xu, H.; Regino, C. A.; Koyama, Y.; Hama, Y.; Gunn, A. J.;
Bernardo, M.; Kobayashi, H.; Choyke, P. L.; Brechbiel, M. W. Bioconjugate
Chem. 2007, 18, 1474–1482.
ꢀ
ꢀ
(3) Costa, J.; Toth, E.; Helm, L.; Merbach, A. E. Inorg. Chem. 2005, 44,
4747–4755.
(4) Solomon, I. Phys. Rev. 1955, 99, 559–565.
(5) Bloembergen, N.; Morgan, L. O. J. Chem. Phys. 1961, 34, 842–850.
ꢀ
€
(6) (a) Fatin-Rouge, N.; Toth, E.; Meuli, R.; Bunzli, J. C. G. J. Alloys
Compd. 2004, 374, 298–302. (b) Vander Elst, L.; Port, M.; Raynal, I.; Simonot,
C.; Muller, R. N. Eur. J. Inorg. Chem. 2003, 2495–2501. (c) Aime, S.; Cabella,
C.; Colombatto, S.; Crich, S. G.; Gianolio, E.; Maggioni, F. J. Magn. Reson.
Imaging 2002, 16, 394–406.
(20) Anderson, E. A.; Isaacman, S.; Peabody, D. S.; Wang, E. Y.; Canary,
J. W.; Kirshenbaum, K. Nano Lett. 2006, 6, 1160–1164.
(21) (a) Yang, J. J.; Yang, J.; Wei, L.; Zurkiya, O.; Yang, W.; Li, S.; Zou,
J.; Zhou, Y.; Maniccia, A. L.; Mao, H.; Zhao, F.; Malchow, R.; Zhao, S.;
Johnson, J.; Hu, X.; Krogstad, E.; Liu, Z. R. J. Am. Chem. Soc. 2008, 130,
9260–9267. (b) Karfeld, L. S.; Bull, S. R.; Davis, N. E.; Meade, T. J.; Barron, A. E.
Bioconjugate Chem. 2007, 18, 1697–1700.
(7) (a) Weinmann, H. J.; Ebert, W.; Misselwitz, B.; Schmitt-Willich, H.
ꢁ
Eur. J. Radiol. 2003, 46, 33–44. (b) Kubícek, V.; Rudovskꢀy, J.; Kotek, J.;
Hermann, P.; Vander Elst, L.; Muller, R. N.; Kolar, Z. I.; Wolterbeek, H. T.; Peters,
ꢁ
J. A.; Lukes, I. J. Am. Chem. Soc. 2005, 127, 16477–16485.
(8) (a) Chang, Y. T.; Cheng, C. M.; Su, Y. Z.; Lee, W. T.; Hsu, J. S.; Liu,
G. C.; Cheng, T. L.; Wang, Y. M. Bioconjugate Chem. 2007, 18, 1716–1727.
(b) Duimstra, J. A.; Femia, F. J.; Meade, T. J. J. Am. Chem. Soc. 2005, 12847–
12855.
(22) Sirlin, C. B.; Vera, D. R.; Corbeil, J. A.; Caballero, M. B.; Buxton,
R. B.; Mattrey, R. F. Acad. Radiol. 2004, 11, 1361–1369.
(23) Cai, J.; Shapiro, E. M.; Hamilton, A. D. Bioconjugate Chem. 2009,
20, 205–208.
(24) Bull, S. R.; Guler, M. O.; Bras, R. E.; Meade, T. J.; Stupp, S. I. Nano
Lett. 2005, 5, 1–4.
(9) Wang, Y. M.; Lee, C. H.; Liu, G. C.; Sheu, R. S. Dalton Trans. 1998,
24, 4113–4118.
(25) (a) Laurent, S.; Vander Elst, L.; Thirifays, C.; Muller, R. N.
Langmuir 2008, 24, 4347–4351. (b) Laurent, Vander Elst, L.; Thirifays, C.;
Muller, R. N. Eur. Biophys. J. 2008, 37, 1007–1014.
(10) Mato-Iglesias, M.; Platas-Iglesias, C.; Djanashvili, K.; Peters, J. A.;
ꢀ
ꢀ
Toth, E.; Balogh, E.; Muller, R. N.; Vander Elst, L.; de Blas, A.; Rodrı
´
guez-
Blas, T. Chem. Commun. 2005, 4729–4731.
ꢁ
ꢁ
(26) (a) Hermann, P.; Kotek, J.; Kubı
´
cek, V.; Lukes, I. Dalton Trans.
ꢀ
(11) Balogh, E.; Mato-Iglesias, M.; Platas-Iglesias, C.; Toth, E.;
ꢀ
ꢀ
2008, 3027–3047. (b) Jaszberenyi, Z.; Moriggi, L.; Schmidt, P.; Weidensteiner,
Djanashvili, K.; Peters, J. A.; de Blas, A.; Rodrı
2006, 45, 8719–8728.
´
guez-Blas, T. Inorg. Chem.
ꢀ
ꢀ
C.; Kneuer, R.; Merbach, A. E.; Helm, L.; Toth, E. J. Biol. Inorg. Chem. 2007,
12, 406–420. (c) Rudovskꢀy, J.; Botta, M.; Hermann, P.; Hardcastle, K. I.; Lukes,
I.; Aime, S. Bioconjugate Chem. 2006, 17, 975–987.
(27) (a) Jenkins, B. G.; Armstrong, E.; Lauffer, R. B. Magn. Reson. Med.
1991, 17, 164–178. (b) Lauffer, R. B. Magn. Reson. Med. 1991, 22, 339–342.
(28) Lauffer, R. B.; Parmelee, D. J.; Dunham, S. U.; Ouellet, H. S.; Dolan,
R. P.; Witte, S.; McMurry, T. J.; Walovitch, R. C. Radiology 1998, 207, 529–
538.
(29) Parmelee, D. J.; Walovitch, R. C.; Ouellet, H. S.; Lauffer, R. B.
Invest. Radiol. 1997, 32, 741–747.
ꢁ
(12) Powell, D. H.; Ni Dhubhghaill, O. M.; Pubanz, D.; Helm, L.;
Lebedev, Y. S.; Schlaepfer, W.; Merbach, A. E. J. Am. Chem. Soc. 1996,
118, 9333–9346.
(13) Aime, S.; Barge, A.; Borel, A.; Botta, M.; Chemerisov, S.; Merbach,
A. E.; Muller, U.; Pubanz, D. Inorg. Chem. 1997, 36, 5104–5112.
(14) Ruloff, R.; Toth, E.; Scopelliti, R.; Tripier, R.; Handel, H.; Merbach,
€
ꢀ
ꢀ
A. E. Chem. Commun. 2002, 2630–2631.
ꢀ
(15) Burai, L.; Toth, E.; Sour, A.; Merbach, A. E. Inorg. Chem. 2005, 44,
3561–3568.

Article
Inorganic Chemistry, Vol. 50, No. 4, 2011 1277
Scheme 1. Schematic Representation of the Synthesis of CB-TTDAa
a (I) DBU, DMF; (II) MeOH; (III) BH3-THF, THF; (IV) BrCH2COOt-Bu, K2CO3, CH3CN; and (V) HCl.
balanced modifications of the ligand carbon backbone with
different structural moieties can affect not only its size,
charge, and lipophilicity but also coordinating ability
with Gd(III).31 Furthermore, the presence of a hydrophobic
residuewilllead to a greater uptake in certain organs and may
have potential to serve as organs specific agents.32
was not enough to induce sufficient lipophilicity for
angiographic application. Hence, apart from increasing the
basicity of the ligand and the rigidity of Gd(III) complex, the
primary intention for selecting the cyclobutyl is to tune the
lipophilicity of [Gd(Bz-CB-TTDA)(H2O)]2- to closely mimic
that of MS-325. The ligand protonation constants and the
thermodynamic and conditional stability constants of the
ligands chelated with Gd(III), Cu(II), Zn(II), and Ca(II) and
their selectivity for Gd(III) over endogenously available
metal ions were estimated with the aim of assessing safety
considerations for the two Gd(III) complexes as potential
medical MRI CAs. The two key parameters affecting the
proton relaxivity, water exchange and rotation dynamics,
have been evaluated by combining variable temperature 17O
nuclear magnetic resonance (17O NMR) and nuclear mag-
netic relaxation dispersion (NMRD) profile studies. Finally,
the binding affinity of [Gd(CB-TTDA)(H2O)]2- and [Gd(Bz-
CB-TTDA)(H2O)]2- toward HSA was evaluated by ultra-
filtration and proton relaxation enhancement methods.
For the past few years, our group has been actively
engaged in the development of [Gd(TTDA)(H2O)]2- deriva-
tives with the objective to study the effects of the substitution
at different site of the TTDA on water exchange, rotational
dynamics, and thermodynamic stability.33,34 Here, we report
the synthesis and physicochemical properties of two new TTDA
derivatives, 6-carboxymethyl-3-{{[1-(N,N-dicarboxymethyl)-
2-aminomethyl]-cyclobut-1-yl}methyl}-3,6-diazaoctanedioic
acid (CB-TTDA) (Scheme 1) and 5-benzyl-6-carboxymethyl-
3-{{[1-(N,N-dicarboxymethyl)-20-aminomethyl]-cyclobut-10-yl}-
methyl}-3,6-diazaoctanedioic acid (Bz-CB-TTDA) (Scheme 2).
The cyclobutyl35 and benzyl36 residues are introduced on the
carbon backbone of [Gd(TTDA)(H2O)]2- as a means to impart
lipophilicity to the Gd(III) complexes. From our previous
study, we have discovered that simply introducing the benzyl
group on the carbon backbone of [Gd(TTDA)(H2O)]2-36,37
Materials and Methods
Chemicals. Reagents and solvents were purchased from com-
mercial sources with the highest quality grade and were used
without further purification, unless stated otherwise. 17O-enriched
water (20.3 atom %) was purchased from Isotec Inc. Human
serum albumin (HSA, product no. A-1653, Fraction V Powder
96-99%) was purchased from Sigma and was used without any
further purification,38 and the molecular weight was assumed to
be 69 kDa. The 50 mM phosphate buffer was used to maintain
the pH of all solutions (pH 7.4) containing HSA. 1H (400 MHz),
13C (100 MHz), and 17O (54.2 MHz) NMR spectra were
recorded on a Varian Gemini-400 spectrometer with 5 mm
(30) Caravan, P.; Comuzzi, C.; Crooks, W.; McMurry, T. J.; Choppin,
G. R.; Woulfe, S. R. Inorg. Chem. 2001, 40, 2170–2176.
(31) (a) Artali, R.; Botta, M.; Cavallotti, C.; Giovenzana, G. B.;
Palmisano, G.; Sisti, M. Org. Biomol. Chem. 2007, 5, 2441–2447. (b) Aime,
S.; Geninatti Crich, S.; Gianolio, E.; Giovenzana, G. B.; Tei, L.; Terreno, E.
Coord. Chem. Rev. 2006, 250, 1562–1579. (c) Aime, S.; Botta, M.; Terreno, E.
Adv. Inorg. Chem. 2005, 57, 173–237. (d) Aime, S.; Botta, M.; Fasano, M.;
Terreno, E. Chem. Soc. Rev. 1998, 27, 19–29.
(32) Sun, Y.; Martell, A. E.; Reibenspies, J. H.; Reichert, D. E.; Welch,
M. J. Inorg. Chem. 2000, 39, 1480–1486.
(33) Wang, Y. M.; Li, C. R.; Huang, Y. C.; Ou, M. H.; Liu, G. C. Inorg.
Chem. 2005, 44, 382–392.
sample tubes. Methyl cyanoacetate, ethylenediamine, BH3
THF, cyclobutane-1,1-dicarboxylic acid diamide, L-phenylalanine,
thionylchloride, gadoliniumchloride, andeuropiumchloride were
3
(34) Ou, M. H.; Tu, C. H.; Tsai, S. C.; Lee, W. T.; Liu, G. C.; Wang, Y. M.
Inorg. Chem. 2006, 45, 244–254.
(35) Liage, S. U.S. Patent 5905176, 1999.
(36) Hsu, J. S.; Jaw, T. S.; Liu, G. C.; Wang, Y. M.; Chen, S. H.; Kuo, Y. T.;
Jao, J. C.; Li, C. W.; Tsai, K. J. Magn. Reson. Imaging 2004, 20, 632–639.
(37) Cheng, T. H.; Lee, T. M.; Ou, M. H.; Li, C. R.; Liu, G. C.; Wang,
Y. M. Helv. Chim. Acta 2002, 85, 1033–1050.
(38) Cohn, E. J.; Strong, L. E.; Hughes, W. L., Jr.; Mulford, D. J.;
Ashworth, J. N.; Melin, M.; Taylor, H. L. J. Am. Chem. Soc. 1946, 68, 459–
475.

1278 Inorganic Chemistry, Vol. 50, No. 4, 2011
Chang et al.
Scheme 2. Schematic Representation of the Synthesis of Bz-
Potentiometric Measurements. The protonation constants
of the ligands were determined by potentiometric titrations of
1.0 mM ligand solutions using an automatic titrator system. To
determine the stability constants for [Gd(CB-TTDA)(H2O)]2-
and [Gd(Bz-CB-TTDA)(H2O)]2-, competition titrations with
EDTA were performed. The ratio of 1 Gd(III)/1 ligand/1 EDTA
was employed. Under this condition, the metal ion is partitioned
between ligand and EDTA, and the rate of transmetalation is fast
enough between pH 2-4. However, complexation was usually
rapid (3-5 min per point to give a stable pH reading) in the Cu(II),
Ca(II), and Zn(II) studies. The same values of the stability constants
were obtained by using the direct or back-titration. The autotitrat-
ing system consists of a 702 SM titroprocessor, a 728 stirrer, and a
PT-100 combination pH electrode (Metrohm). The protonation
constants of the ligands were calculated using a FORTRAN
computer program, PKAS,40 written for polyprotonic weak acid
equilibria. The overall stability constants of the various metal
complexes formed in aqueous solution were determined from the
titration data with the FORTRAN computer program BEST.40
Variable-Temperature 17O NMR Measurement. The measure-
ment of the 17O transverse relaxation rates, longitudinal relaxation
rates, and chemical shifts was carried out with a Varian Gemini-
400 spectrometer, equipped with a 10 mm probe, by using an
external D2O lock. The Varian 600 temperature control unit was
used to stabilize the temperature in the range 278-338 K. Solutions
containing 5.5% of the 17O isotope were used. The inversion
recovery sequence was applied to measure longitudinal relaxation
rates, 1/T1, and the Carr-Purcell-Meiboom-Gill spin-echo
technique was used to obtain transverse relaxation rates, 1/T2.
The solution was introduced into spherical glass containers, fitting
into ordinary 10 mm NMR tubes, in order to eliminate magnetic
susceptibility corrections to chemical shifts.
CB-TTDAa
Preparation of 4.5% (w/v) Human Serum Albumin (HSA) and
GdL/HSA Solutions. HSA was dissolved in a 10 mM sodium phos-
phate and 150 mM sodium chloride (pH 7.4) solution. Two 4.5%
(w/v) HSA solutions were prepared: one containing the GdL chelate
at a concentration of 5.3 mM and another without the GdL chelate.
Ultrafiltration Measurements of Binding of GdL Chelate to
HSA. GdL/HSA solutions ranging from 0.05 to 5.3 mM GdL
chelate in 4.5% (w/v) HSA were made by combining appropriate
amounts of 4.5% (w/v) HSA and 5.3 mM GdL in 4.5% (w/v) HSA
solution. Aliquots (400 μL) of these solutions were placed in 30
kDa ultrafiltration units, incubated at 37.0 (0.1 ꢀC for 20 min, and
then centrifuged at 3500g for 10 min. The filtrates obtained from
these ultrafiltration units were used to measure the concentration
of free GdL chelate. Duplicate aliquots were operated for each
sample. The Gd(III) concentrations of the GdL/HSA solutions
and ultrafiltrates were obtained using ICP-MS.
a (I) BH3-THF, THF; (II) SOCl2, MeOH; (III) TEA, MeOH; (IV) BH3-
THF, THF; (V) BrCH2COOt-Bu, CH3CN, K2CO3; and (VI) HCl.
purchased from Aldrich. 1,8-Diazabicyclo[5,4,0]undec-7-ene
(DBU) and 1,3-dibromopropane were purchased from Alfa
aesar. tert-Butyl bromoacetate was synthesized according to the
reported procedure.39 The details of the synthesis of CB-TTDA
and Bz-CB-TTDA are reported in the Supporting Information,
and refer to Schemes 1 and 2 for the formula.
Preparation of [Ln(CB-TTDA)(H2O)]2- and [Ln(Bz-CB-TTDA)-
(H2O)]2- Complexes. The Ln(III) = Eu(III) and Gd(III)) com-
plexes were prepared by dissolving the ligand (0.05 mmol) in H2O
(3 mL), and the pH of the solution was adjusted to 6.5 with 1 M
NaOH. To these solutions, 2.5 mL of an aqueous solution of LnCl3
(0.05 mmol) was added dropwise, maintaining the pH at 6.5 with 1
M NaOH. The Ln(III) chelates were instantaneously formed at
room temperature. For the 17O NMR sample, Gd(ClO4)3 stock
solution was prepared according to the literature.33 The aqueous
solutions of Gd(III) chelates were prepared by mixing the ligand
and Gd(ClO4)3 stock solution in stoichiometric amounts. A slight
ligand excess was used, and the pH of all solutions was adjusted
using 1 M NaOH or HClO4. The solutions were then evaporated
under reduced pressure, and 1 equiv weight of 5.5% 17O-enriched
water was added. The absence of free metal in the solutions was
checked by the xylenol orange test.34
Butanol Buffer Partition Coefficient. The Gd(III) chelated Bz-
CB-TTDA ([Gd]T = 0.07 mM in 10 mL of PBS) was equilibrated
at room temperature for 1-2 h with PBS-saturated butanol
(10 mL). The vials were centrifuged at 2000g for 5 min to ensure
that the layers were separated. Aliquots (5 mL) from each phase
were removed and analyzed via inductively coupled plasma-
atomic emission spectroscopy (ICP-AES, PerkinElmer OPTIMA
2000 PV). The partition coefficient (P) was calculated as follows.41
avg conc of GdðIIIÞ complex in butanol
P ¼
avg conc of GdðIIIÞ complex in PBS
Transmetalation Experiments. This technique is based on
measurement of the evolution of the water proton paramagnetic
longitudinal relaxation rate (R1m) of a phosphate buffer solution
Relaxation Time Measurement. Relaxation times T1 of aqu-
eous solutions of gadolinium(III) complexes were measured to
determine relaxivity r1. All measurements were made using a
relaxometer operating at 20 MHz and 37.0 ( 0.1 ꢀC (NMS 120
minispec, Bruker) by means of the standard inversion-recovery
pulse sequence.
(40) Martell, A. E.; Motekaitis, R. J. Determination and Use of Stability
Constants, 2nd ed.; VCH: New York, 1992.
(41) McMurry, T. J.; Parmelee, D. J.; Sajiki, H.; Scott, D. M.; Ouellet,
H. S.; Walovitch, R. C.; Tyeklar, Z.; Dumas, S.; Bernard, P.; Nadler, S.;
Midelfort, K.; Greenfield, M.; Troughton, J.; Lauffer, R. B. J. Med. Chem.
2002, 45, 3465–3474.
(39) Vollmar, A.; Dunn, M. S. J. Org. Chem. 1960, 25, 387–390.

Article
Inorganic Chemistry, Vol. 50, No. 4, 2011 1279
Table 1. Protonation Constants of Bz-CB-TTDA, CB-TTDA, TTDA, MS-325, EPTPA-bz-NO2, and DTPA at I = 0.1 M Me4NCla and T = 25.0 ( 0.1 ꢀC
log Ki
TTDAa
MS-325b
EPTPA-bz-NO2
DTPAd
c
equilibrium
Bz-CB-TTDA
CB-TTDA
[HL]/[L][H]
[H2L]/[HL][H]
[H3L]/[H2L][H]
[H4L]/[H3L][H]
ΣpKa
10.72 (0.04)
9.40 (0.01)
5.54 (0.09)
2.43 (0.03)
28.09
11.08 (0.03)
9.17 (0.05)
5.23 (0.02)
3.11 (0.04)
28.59
10.60
8.92
5.12
2.80
27.44
11.15 (0.12)
8.62 (0.06)
4.51 (0.09)
2.96 (0.06)
27.24
10.86
8.91
4.70
3.25
27.72
10.49
8.60
4.28
2.64
26.01
a Data were obtained from ref 33. b Data were obtained from ref 30. c Data were obtained from ref 44. d Data were obtained from ref 45.
containing 2.5 mM Gd(III) complex and 2.5 mM ZnCl2.42,43
Relaxation rate (R1m) measurements were performed at 37.0 (
0.1 ꢀC and 20 MHz (Bruker Minispec 120).
which were calculated from potentiometric titration curves,
are reported in Table 1 along with those of TTDA, MS-325
(4-(R)-[(4,4-diphenyl-cyclohexyl)-phosphanooxyethyl]3,6,
9-triaza-3,6,9-tris(methoxy-carbonyl)-undecanedioato
gadolinium(III)) ligand, EPTPA-bz-NO2 ((4-p-nitroben-
zyl-3,6,10-tri(carboxymethyl)-3,6,10-triazadodecanedioic
acid),44 and DTPA (3,6,9-tri(carboxymethyl)-3,6,9-tria-
zaundecanedioic acid)45 (Chart 1). The overall basicity of
each ligand can be quantified by calculating ΣpKa.46A
comparison of the ΣpKa values of CB-TTDA (28.59) and
Bz-CB-TTDA (28.09) to that of its parent TTDA (27.44)
indicates that the introduction of the four member ring on
the carbon backbone of TTDA significantly increases the
ΣpKa value. This result implies that the presence of
cyclobutyl in CB-TTDA and Bz-CB-TTDA considerably
increases their basicity. The overall basicity decreases in
the following order: CB-TTDA > Bz-CB-TTDA ≈ EP-
TPA-bz-NO2 ≈ TTDA ≈ MS-325 > DTPA.
Data Analysis. The simultaneous least-squares fitting of 17O
NMR data and the binding parameters to HSA were determined
by fitting the experimental data using the program SCIENTIST
for Windows by MICROMATH, version 2.0.
Results and Discussion
Synthesis of [Gd(CB-TTDA)(H2O)]2-. The ligand CB-
TTDA was synthesized in four steps, as shown in Scheme 1
(the detailed synthesis method is summarized in the
Supporting Information). In the first step methyl cyanoa-
cetate was treated with 1,3-dibromopropane in the presence
of DBU in DMF at 70 ꢀC. The compound 1-cyano-cyclo-
butanecarboxylic methyl ester (1) was obtained. Subse-
quently, compound 1 was reacted with ethylenediamine
in methanol at room temperature for 15 h. It gave rise to
1-cyano-cyclobutanecarboxylic acid (2-aminoethyl) amide
All of the metal complexes show titration curve depres-
sion (Figures 5S and 6S in the Supporting Information)
relative to the titration curve for the ligand, reflecting
complex formation. The thermodynamic stability of a
metal complex is given by the thermodynamic stability
constants (KML(therm)), following eq 1
(2). The solution of 2 and BH3 THF was stirred for 15 h
3
at 60 ꢀC, followed by the addition of the 6 M HCl resulted
in the formation of compound 3. Finally, compound 3
was treated with tert-butylbromoacetate in the presence
of anhydrous K2CO3 in CH3CN at 80 ꢀC to provide CB-
TTDA(4). [Gd(CB-TTDA)(H2O)]2- identified by LC-MS
was obtained by complexation of Gd(III) with CB-TTDA
in water at room temperature.
½MLꢁ
KMLðthermÞ
¼
ð1Þ
½Mꢁ½Lꢁ
Synthesis of [Gd(Bz-CB-TTDA)(H2O)]2-. The synthetic
scheme of Bz-CB-TTDA ligand is depicted in Scheme 2 (the
detail synthesis method is summarized in the Supporting
Information). The synthesis procedure begins with reduc-
tion of amide in cyclobutane-1,1-dicarboxylic acid diamide
where [M], [L], and [ML] are the concentrations of the
metal ion, the deprotonated ligand, and the complex,
respectively. The thermodynamic stability constants of
CB-TTDA, Bz-CB-TTDA, TTDA, EPTPA-bz-NO2, and
DTPA complexes with different metal ions are summar-
ized in Table 2. The order of increase in the stability
constant for these ligands chelated with Gd(III) is
[Gd(TTDA)(H2O)]2- ≈ [Gd(EPTPA-bz-NO2)(H2O)]2- <
[Gd(CB-TTDA)(H2O)]2- ≈ [Gd(Bz-CB-TTDA)(H2O)]2- <
[Gd(DTPA)(H2O)]2- ≈ MS-325. The higher stability
constants of the CB-TTDA and Bz-CB-TTDA with Gd-
(III), Ca(II), Cu(II), and Zn(II) compared to those of
TTDA and EPTPA-bz-NO2 can be explained in terms of
steric effect and basicity. The stability of Gd(III) com-
plexes is influenced by their coordination geometry which
is dictated by the ligand structure and flexibility.47 The
with BH3 THF in THF at 60 ꢀC to obtain (1-(amino-
3
methyl)cyclobutyl)methamine (5). Subsequently, compound
5 was treated with L-phenylalanine methyl ester in methanol
at room temperature for 15 h to obtain 2-amino-N-((1-
aminomethyl)cyclobutyl)-3-phenylpropanamide (7).Finally,
treatment of compound 8 with tert-butylbromoacetate in
the presence of anhydrous K2CO3 in CH3CN at 90 ꢀC
provides Bz-CB-TTDA (9) which was characterized by 1H
NMR and LC-MS. [Gd(Bz-CB-TTDA)(H2O)]2- identified
by LC-MS was obtained by complexation of Gd(III) with
Bz-CB-TTDA in water at room temperature.
Protonation and Stability Constant. The potentiometric
titration curves for Bz-CB-TTDA and CB-TTDA have
been deposited as Supporting Information (Figures 5S
and 6S in the Supporting Information). The protonation
constants (log Ki) of Bz-CB-TTDA and CB-TTDA,
ꢀ
(44) Laus, S.; Ruloff, R.; Toth, E.; Merbach, A. E. Chem.;Eur. J. 2003,
9, 3555–3566.
(45) Smith, R. M.; Martell, A. E. Critical Stability Constants; Plenum:
New York, 1974.
(46) Kumar, K.; Tweedle, M. F.; Malley, M. F.; Gougoutas, J. Z. Inorg.
Chem. 1995, 34, 6472–6480.
(42) Laurent, S.; Vander Elst, L.; Muller, R. N. Contrast Media Mol.
Imaging 2006, 1, 128–137.
(43) Laurent, S.; Vander Elst, L.; Copoix, F.; Muller, R. N. Invest. Radiol.
2001, 36, 115–122.
(47) Tei, L.; Baranyai, Z.; Botta, M.; Piscopo, L.; Aime, S.; Giovenzana,
G. B. Org. Biomol. Chem. 2008, 6, 2361–2368.
(48) Rabasso, N.; Louaisil, N.; Fadel, A. Tetrahedron. 2006, 62, 7445–
7454.

1280 Inorganic Chemistry, Vol. 50, No. 4, 2011
Chang et al.
Chart 1
Table 2. Stability Constants, Conditional Stability Constants, Selectivity Constants, and Modified Selectivity Constants of Gd3þ, Zn2þ, Ca2þ, and Cu2þ Complexes with
Bz-CB-TTDA, CB-TTDA, TTDA, MS-325, EPTPA-bz-NO2, and DTPA at I = 0.1 M Me4NCl; T = 25.0 ( 0.1 ꢀCa
TTDAb
MS-325c
EPTPA-bz-NO2
DTPAe
d
parameter
Bz-CB-TTDA
CB-TTDA
log KGdL(therm)
log KGdL(cond) (pH = 7.4)
pGd
log KcaL(therm)
log KCaHL
log KCaL(cond) (pH = 7.4)
pCa
log KZaL(therm)
log KZnHL
log KZnL(cond) (pH = 7.4)
pZn
log KCuL(therm)
log KCuHL
log KCuL(cond) (pH = 7.4)
pCu
log Ksel(Gd/Ca)
log Ksel(Gd/Zn)
log Ksel(Gd/Cu)
20.09 (0.04)
14.77
15.77
9.32 (0.04)
6.31 (0.05)
4.00
20.28 (0.03)
14.82
15.82
9.52 (0.02)
6.24 (0.02)
4.06
18.96
14.25
15.25
9.13
8.20
4.42
6.28
16.03
7.80
11.32
12.86
16.77
5.69
22.06 (0.02)
17.06
19.94
10.45 (0.06)
5.66 (0.02)
5.45
19.20
14.22
15.3
22.46
18.14
19.14
10.75
6.11
6.43
7.46
18.70
5.60
14.38
15.39
21.38
4.81
17.00
18.06
11.71
3.76
9.38
4.40
5.03
5.09
8.34
15.97 (0.06)
8.29 (0.02)
10.65
16.06 (0.03)
8.13 (0.04)
10.60
17.82 (0.09)
5.60 (0.02)
12.82
16.01
11.03
18.47
13.49
12.58
12.41
15.71
17.26 (0.04)
7.20 (0.06)
11.94
13.15
10.77
4.12
2.83
8.28
17.71 (0.02)
5.29 (0.03)
12.25
13.26
10.76
4.22
2.57
8.24
21.3 (0.3)
5.16 (0.08)
16.30
19.18
11.61
4.24
0.76
6.75
12.06
13.06
9.83
2.93
2.19
9.82
3.19
0.73
6.65
1.08
7.06
0
log Ksel
7.18
a Uncertainty (σ) in log K values are given in parentheses. b Data were obtained from ref 33. c Data were obtained from ref 30. d Data were obtained
from ref 44. e Data were obtained from ref 45.
cyclobutyl residue on the carbon backbone of TTDA
may increase the steric rigidity.48 This could facilitate the
CB-TTDA and the Bz-CB-TTDA to constrain into a
chelating conformation, consequently escalating their
coordinating ability toward metal ion. A fraction of this
increase in stability is also due to higher basicity49 of CB-
TTDA and Bz-CB-TTDA than those of TTDA and
EPTPA-bz-NO2.
Conditional Stability and Selectivity Constant. Several
studies have emphasized the importance of the conditional
stability constant for evaluating the stability of Gd(III)
complex under physiological pH 7.4. The conditional
stability constant can be calculated by using eq 2
ꢀ
2
KMLðthermÞ ¼ KMLðcondÞ 1 þ K1½Hþꢁ þ K1K2½Hþꢁ
ꢁ
n
þ
þ Kn½Hþꢁ
ð2Þ
3 3 3
(49) Kumar, K.; Tweedle, M. F. Inorg. Chem. 1993, 32, 4193–4199.

Article
Inorganic Chemistry, Vol. 50, No. 4, 2011 1281
where K1, K2, K3 ,..., Kn are the stepwise protonation
constant of the ligand. The conditional stability constants
of CB-TTDA, Bz-CB-TTDA, TTDA, EPTPA-bz-NO2, and
DTPA complexed with different metals ions (Gd(III),
Zn(II), Cu(II)) are given in Table 2. It is smaller than that
of thermodynamic stability constant due to the competi-
tion between Hþ and the metal ion for the ligand at
physiological pH. Moreover, very significant information
on the stability of the metal complex and sequestering
ability of the ligand at physiological condition comes
from the pM value (pM = -log[Mf], where [Mf] is the
equibilirum concentration of the free aqua metal ion
present at pH=7.4). The pM values reflect the influence
of the ligand basicity and the protonation of the complex.
The larger the pM value is, the higher is the affinity of the
ligand for the metal ion under the given condition; that is,
the larger the pM is, the lower the concentration of free
metal ion present.44,50 The pM values of CB-TTDA, Bz-
CB-TTDA, TTDA, EPTPA-bz-NO2, and DTPA with
different metal ions (Gd(III), Ca(II), Cu(II), and Zn(II))
are summarized in Table 2. The pGd values of [Gd(CB-
TTDA)(H2O)]2- and [Gd(Bz-CB-TTDA)(H2O)]2- are
higher than those of their pCa, pZn, and pCu values.
There is only a small difference between the pGd and the
pCu values of CB-TTDA and Bz-CB-TTDA; hence, Cu-
(II) can compete with Gd(III) for these ligands from a
thermodynamic viewpoint. However, Cu(II) exists in very
low concentrations in vivo. The metal ions, Zn(II) and
Ca(II), have lower affinity for CB-TTDA and Bz-CB-
TTDA than that of Gd(III) at pH 7.4. The pGd values of
[Gd(CB-TTDA)(H2O)]2- and [Gd(Bz-CB-TTDA)(H2O)]2-
are slightly higher than those of [Gd(TTDA)(H2O)]2-
and [Gd(EPTPA-bz-NO2)(H2O)]2- due to elevated basi-
city and rigidity endowed by cyclobutyl groups. However,
both complexes, [Gd(CB-TTDA)(H2O)]2- and [Gd(Bz-
CB-TTDA)(H2O)]2-, have lower pGd values than those
of [Gd(DTPA)H2O]2- and MS-325. Furthermore, as the
blood pool agent retains in the bloodstream for longer
time, reactions of CAs with various endogenous sub-
stances (metal ions and biological ligands) may occur.
Evaluation of competition between Gd(III) and the en-
dogenous metal ions must be taken into account for
toxicological reasons. The logarithmic selectivity con-
stants (defined as log(KGdL/KML), M = Zn(II), Ca(II),
and Cu(II)) of CB-TTDA, Bz-CB-TTDA, TTDA, EPT-
PA-bz-NO2, and DTPA for Gd(III) over Ca(II), Zn(II),
and Cu(II) are summarized in Table 2. The selectivity
of both CB-TTDA and Bz-CB-TTDA for Gd(III) against
Zn(II) is higher than those of TTDA, EPTPA-bz-NO2,
and DTPA, but is smaller than that of MS-325. Higher
selectivity toward Gd(III) against Zn(II) indicates that
endogenously available Zn(II) ion, which is one of the
most copious metal ions in blood plasma, is less able
where R is a side reaction coefficient (eqs 1S-4S in the
0
Supporting Information). log Ksel is a thermodynamic
stability constant corrected for the competition between
the endogenously available metal ions and Hþ. The
0
calculated log Ksel values for the Gd(III) complexes are
summarized in Table 2 and follow the order CB-TTDA
(8.24) ∼ Bz-CB-TTDA (8.28) > TTDA (7.18) ∼ DTPA
(7.06) > MS-325 (6.75) > EPTPA-bz-NO2 (6.65). These
results imply that [Gd(CB-TTDA)(H2O)]2- and [Gd(Bz-
CB-TTDA)(H2O)]2- are fairly stable under physiological
conditions.
Assessment of the Hydration State. Luminescence life-
time measurements have been used to quantify the number
of inner-sphere water molecules51 in [Gd(Bz-CB-TTDA)-
(H2O)]2- and [Gd(CB-TTDA)(H2O)]2-. Eu(III) is com-
monly applied for lifetime measurements because it emits
in the visible region and shows longer emission lifetime.52
In the lanthanide series of the periodic table, Gd(III)
(Z = 64) just comes after Eu(III) (Z = 63), and hence,
Gd(III) has nearly the same ratio of charge to ionic radius
as Eu(III) and correspondingly similar coordination
chemistry. Therefore, the number of inner-sphere water
molecules determined for [Eu(Bz-CB-TTDA)(H2O)]2-
and [Eu(CB-TTDA)(H2O)]2- should correspond to the
hydration status of Gd(III) complexes of these ligands.
The emission lifetimes (τ) of the Eu(5D0) excited level
have been measured in 2 mM solutions of the [Eu(Bz-CB-
TTDA)(H2O)]2- and [Eu(CB-TTDA)(H2O)]2- in D2O and
H2O, and were used to calculate the number of coordi-
nated water molecules q (Table 3) using eqs 4 and 553
h
i
- 1
- 1
q ¼ 1:05 τH O þ τD O
ð4Þ
ð5Þ
2
2
hꢀ
ꢂ
i
- 1
- 1
q ¼ 1:2 τH O þ τD O
- 0:25
2
2
where q is the number of water molecules bound to metal
ions, τH O is the luminescence half-life in H2O solution,
2
and τD O is the luminescence half-life in D2O solution.
2
Equation 5 takes into account the vibration of the N-H
bonds. Consequently, eq 5 is a little more precise than
eq 4. The q values obtained for [Eu(Bz-CB-TTDA)-
(H2O)]2- and [Eu(CB-TTDA)(H2O)]2- fall within the usual
range that corresponds to single hydration. For instance,
the q values of [Eu(Bz-CB-TTDA)(H2O)]2- and [Eu(CB-
TTDA)(H2O)]2- are similar to those of MS-325,16 [Eu(BO-
PTA)(H2O)]2- (BOPTA = 4-carboxy-5,8,11-tris(carboxy-
methyl)-1-phenyl-2-oxa-5,8,11-triazatridecan-13-oicacid),54
and [Eu(DTPA)(H2O)]2-55 (Chart 1). Hence, a general
(50) (a) Bannochie, C. J.; Martell, A. E. Inorg. Chem. 1991, 30, 1385–1392.
(b) Tweedle, M. F.; Hahan, J. J.; Kumar, K.; Mantha, S.; Chang, C. A. Magn.
Reson. Imaging 1991, 9, 409–415.
(51) Horrocks, W. D., Jr.; Sudnick, D. R. Acc. Chem. Res. 1981, 14, 384–
392.
to displace Gd(III) in the chelate. Additionally, the
0
modified selectivity constant (log Ksel ) given by eq 3
provides more precise information about the selectivity
for Gd(III) over other endogenous metal ions and Hþ for
a ligand
46
ꢀ
ꢀ
ꢀ
(52) Mato-Iglesias, M.; Roca-Sabio, A.; Palinkas, Z.; Esteban-Gomez,
ꢀ
D.; Platas-Iglesias, C.; Toth, E.; de Blas, A.; Rodrıguez-Blas, T. Inorg. Chem.
´
2008, 47, 7840–7851.
(53) (a) Beeby, A.; Clarkson, I. M.; Dickins, R. S.; Faulkner, S.; Parker,
D.; Royle, L.; Sousa, A. S. d.; Williams, J. A. G.; Woods, M. J. Chem. Soc.,
Perkin Trans. 1999, 2, 493–504. (b) Horrocks, W. D.; Sudnick, D. R. J. Am.
Chem. Soc. 1979, 101, 334–340.
ꢀ
ꢂ
- 1
- 1
- 1
- 1
- 1
0
Ksel ¼ KMLðthermÞ RH þRCaL þRCuL þRZnL
(54) Anelli, P. L.; Balzani, V.; Prodi, L.; Uggeri, F. Gazz. Chim. Ital. 1991,
121, 359–364.
(55) Bryden, C. C.; Reilley, C. N. Anal. Chem. 1982, 54, 610–615.
ð3Þ

1282 Inorganic Chemistry, Vol. 50, No. 4, 2011
Chang et al.
Table 3. Luminescence Lifetimes of [Eu(CB-TTDA)(H2O)]2-, [Eu(Bz-CB-TTDA)(H2O)]2-, MS-325, [Eu(BOPTA)(H2O)]2-, and [Eu(DTPA)(H2O)]2- Complexes and
Number of Inner-Sphere Water Molecules (q)
luminescence lifetimes
H O (ms)
q
Eu(III) complex
τ
τ
D O (ms)
eq 4
eq 5
2
2
[Eu(CB-TTDA)(H2O)]2-a
[Eu(Bz-CB-TTDA)(H2O)]2-a
MS-325b
0.64
0.62
2.44
2.49
1.21 ( 0.01
1.26 ( 0.02
1.09 ( 0.02
1.41 ( 0.03
1.02
[Eu(BOPTA)(H2O)]2-c
[Eu(DTPA)(H2O)]2-d
1.2
1.2
a The value was determined with eqs 4 and 5. b Data were obtained from ref 16. c Data were obtained from ref 54. d Data were obtained from ref 55.
Reported errors are statistical errors, but the methodological errors are much larger.
Table 4. Relaxivity r1 of [Gd(CB-TTDA)(H2O)]2-, [Gd(Bz-CB-TTDA)(H2O)]2-
,
formula of [Eu(L)(H2O)]2- can be proposed for these
complexes in solution.
MS-325 [Gd(TTDA)(H2O)]2-, and [Gd(DTPA)(H2O)]2- at 37.0 ( 0.1ꢀC and 20
MHz
Relaxometric Studies of Gd(III) Complexes. The long-
itudinal water proton relaxivity (r1) that results from the
contribution of the water molecules in the inner and outer
coordination spheres is given by eq 6.56
complex
pH
relaxivity r1 (mM-1 s-1
)
[Gd(CB-TTDA)(H2O)]2-
[Gd(Bz-CB-TTDA)(H2O)]2-
MS-325a
7.4 ( 0.1
7.4 ( 0.1
7.4 ( 0.1
7.5 ( 0.1
7.6 ( 0.1
4.12 ( 0.05
4.29 ( 0.03
6.84 ( 0.48
3.85 ( 0.03
3.89 ( 0.03
r1 ¼ rI1S þ rO1 S
ð6Þ
[Gd(TTDA)(H2O)]2-b
[Gd(DTPA)(H2O)]2-c
The superscripts “IS” and “OS” refer to the inner and
outer sphere, respectively. It is exceedingly difficult to
alter the outer-sphere contribution.57 Furthermore, for
the Gd(III) complexes of similar size, it can be assumed
that the outer sphere makes a similar contribution at high
magnetic flux density (>0.1 T). Consequently, for the
new generation contrast agent of higher efficiency, the
inner-sphere contribution becomes much more important
and is expressed by eq 757-59
a Data were obtained from ref 59. b Data were obtained from ref 9.
c Data were obtained from ref 60
of the low molecular weight Gd(III) complexes is
limited by rotational correlation time rather than
water exchange rate. As a consequence, the difference
in water exchange rate in the two Gd(III) complexes has
practically no influence on the relaxivity which is restricted
by fast rotation.
Water-Exchange Rate and Rotational Correlation Time
Studies of Gd(III) Complexes. To probe the microscopic
parameters that determine the proton relaxivity of a
Gd(III) complex, for instance, the water exchange rate
(kex), the rotational correlation time (τR), and the electron
spin relaxation (1/T1,2e), and to better understand the
consequence of structural modification of ligand on the
water exchange and the rotational dynamics, we carried
out a combined NMRD and variable temperature 17O
NMR study. The former technique measures the excess
longitudinal proton relaxation caused by the presence of
the Gd(III) complex as a function of magnetic field.61 The
method is a direct measure of relaxation rate and is
theoretically sensitive to all the parameters that influence
relaxation rate. However, there are many parameters that
are ill-defined by the NMRD curve alone.12 Combination
of 17O NMR is helpful to investigate the number of
parameters that influence proton relaxivity, for instance,
the number of inner-sphere water molecules, the rota-
tional correlation time, and the longitudinal electronic
relaxation rate of the complexes.62-64 In addition, it is
an efficient technique to study water exchange rate and
its mechanism of reaction which can be elucidated by
qC
1
rI1S
¼
ð7Þ
55:6T1M þ τM
where C represents the molar concentration of the Gd-
(III) complex, q is the number of water molecules bound
to metal ions, T1M is the longitudinal relaxation time of
the inner-sphere water protons, and τΜ is the residence
lifetime of the bound water. Thus, rI1S is maximized when
T1M > τM (fast exchange conditions) and T1M is as short
as possible.59 The longitudinal relaxivity (r1) values of
[Gd(Bz-CB-TTDA)(H2O)]2- and [Gd(CB-TTDA)(H2O)]2-
determined at 20 MHz and 37.0 ( 0.1 ꢀC are shown in
Table 4 and Figure 8S in the Supporting Information
along with those of [Gd(TTDA)(H2O)]2-9 and [Gd(DT-
PA)(H2O)]2- 60
The relaxivity of [Gd(Bz-CB-TTDA)-
.
(H2O)]2- is similar to that of [Gd(CB-TTDA)(H2O)]2- and
slightly higher than those of [Gd(TTDA)(H2O)]2- and
[Gd(DTPA)(H2O)]2- owing to higher molecular weight.
Although the water exchange rate on [Gd(TTDA)(H2O)]2-
is much higher than that of [Gd(DTPA)(H2O)]2-, the
relaxivity profile shown by the two complexes is
similar.9,33 This is due to the fact that the relaxivity
ꢀ
(56) Toth, E., Helm, L., Merbach, A. E., Eds. The Chemistry of Contrast
(61) Koenig, S. H.; Brown, R. D., III. Prog. NMR Spectrosc. 1990, 22,
487–567.
Agents in Medical Magnetic Resonance Imaging; John Wiley and Sons: New
York, 2001; pp 45-119.
ꢀ
(62) Gonzalez, G.; Powell, D. H.; Tissieres, V.; Merbach, A. E. J. Phys.
(57) Swift, T. J.; Connick, R. E. J. Chem. Phys. 1962, 37, 307–320.
(58) Luz, Z.; Meiboom, S. J. Chem. Phys. 1964, 40, 1058–1066.
(59) (a) Caravan, P.; Parigi, G.; Chasse, J. M.; Cloutier, N. J.; Ellison,
J. J.; Lauffer, R. B.; Luchinat, C.; McDermid, S. A.; Spiller, M.; McMurry,
T. J. Inorg. Chem. 2007, 46, 6632–6639. (b) Aime, S.; Botta, M.; Fasano, M.;
Terreno, E. Acc. Chem. Res. 1999, 32, 941–949.
Chem. 1994, 98, 53–59.
€
(63) Micskei, K.; Helm, L.; Brucher, E.; Merbach, A. E. Inorg. Chem.
1993, 32, 3844–3850.
€
(64) Micskei, K.; Powell, D. H.; Helm, L.; Brucher, E.; Merbach, A. E.
Magn. Reson. Chem. 1993, 31, 1011–1020.
(65) Frey, U.; Merbach, A. E.; Powell, D. H. In Solvent Exchange on
Metal Ions: A Variable NMR Approach; Delpuech, J.-J., Ed.; John Wiley &
Sons: Chichester, U.K., 1995; p 263.
(60) Weinmann, H. J.; Brasch, R. C.; Press, W. R.; Wesbey, G. E. Am. J.
Roentgenol. 1984, 142, 619–624.

Article
Inorganic Chemistry, Vol. 50, No. 4, 2011 1283
variable pressure measurement.65,66 The main advantage
of the 17O NMR technique is that the outer-sphere
contributions to both transverse and longitudinal 17O
relaxation rates are negligible; this is due to the fact that
the oxygen nucleus is closer to the paramagnetic center
when it is bound in the inner sphere.63 Since the oxygen is
directly coordinated to Gd(III), the scalar contribution is
the most significant in the case of the 17O transverse
relaxation.63 Furthermore, the longitudinal relaxation
rates in Gd(III) complex solutions are dominated by
dipole-dipole and quadrupolar interactions and give
direct access to the rotational correlation time of the
Gd-O vector (τR).67 The 17O NMR and NMRD data
for [Gd(CB-TTDA)(H2O)]2- and [Gd(Bz-CB-TTDA)-
(H2O)]2- have been analyzed simultaneously (all the relevant
equations are given in the Supporting Information) and
are shown together with fitted curves in Figure 1. In order
to reduce the number of variables in the fit, some of the
parameters were set to fixed values or adopted from the
data published for [Gd(TTDA)(H2O)]2-derivatives.33,34,44
The parameters affecting water exchange and proton
relaxivity of [Gd(CB-TTDA)(H2O)]2- and [Gd(Bz-CB-
TTDA(H2O)]2- are summarized and compared with
other Gd(III) complexes in Table 5.
The kex
values for [Gd(CB-TTDA)(H2O)]2- and
298
[Gd(Bz-CB-TTDA)(H2O)]2- are significantly higher than
those of clinically used nine-coordinate Gd(III) complexes,
for instance, [Gd(DTPA)(H2O)]2-,[Gd(DTPA-BMA)(H2O)],
and [Gd(DOTA)(H2O)]-. The high water exchange rate
on [Gd(CB-TTDA)(H2O)]2- and [Gd(Bz-CB-TTDA)-
(H2O)]2- can be explained in terms of carbon backbone
length. The longer carbon backbone of CB-TTDA and
Bz-CB-TTDA compared to that of DTPA facilitates
better wrapping of the Gd(III) ion and is pulled tightly
into the first coordination sphere.44 This leads to crowd-
ing on the water binding site, which in turn facilitates in
squeezing out the bonded water molecules. The water
exchange rate is further accelerated due to the rigidity
endowed by the presence of the cyclobutyl group49 on the
carbon backbone of [Gd(CB-TTDA)(H2O)]2- and [Gd-
(Bz-CB-TTDA)(H2O)]2-. The water exchange on [Gd(CB-
TTDA)(H2O)]2- and [Gd(Bz-CB-TTDA)(H2O)]2- most
probably takes place via a limiting dissociative D me-
chanism based on the structural analogy to [Gd(EPT-
PA)(H2O)]2- (ΕPTPA=3,6,10-tri(carboxymethyl)-3,6,
10-triazadodecanedioic acid).44
The rotational correlation time (τR) is a vital parameter
that influences proton relaxivity. NMRD technique has
been frequently employed for the determination of τR
values. A negligible scalar relaxation contribution simplifies
the treatment of NMRD data. The τR value of a Gd(III)
complex can also be determined from the longitudional
17O relaxation (transverse relaxation contains no infor-
mation on rotational motion of the system). A major
drawback of both techniques is reliability on distance
The kex value can be directly computed from variable
temperature transverse 17O relaxation rates measured for
Gd(III) complex solutions. The reduced relaxation rates
and the chemical shift, 1/T1r, 1/T2r, and Δωr can be
calculated from measured 17O NMR relaxation rates
and angular frequency of the paramagnetic solution 1/T1,
1/T2, and ω, and the acidified water reference 1/T1A
,
1
1/T2A, and ωA. The data fitted simultaneously according
to eqs 5S-13S68-70 can be found in the Supporting
Information.
estimation (Gd-H distance for H-NMRD and Gd-O
distance for 17O NMR). Furthermore, discrepancy in the
τR value obtained from NMRD and 17O NMR has been
observed, and the τR value obtained by NMRD is gen-
erally lower than that of 17O NMR. Due to the practical
importance of NMRD data for MRI, contrast agents are
generally characterized by the τR value obtained from
NMRD.72 The τR298 values for [Gd(CB-TTDA)(H2O)]2-
The maxima on the reduced transverse 17O relaxation
rate (1/T2r) curve and the inflection point on the reduced
chemical shift (Δωr) curve indicates the change over from
the fast exchange domain to the slow exchange.56 How-
ever, in the case of [Gd(CB-TTDA)(H2O)]2- and [Gd(Bz-
CB-TTDA)(H2O)]2- over the whole temperature range,
we observed neither maxima nor inflection on 1/T2r and
Δωr curves; that is, the 1/ τm side of the curve is not
observed (Figure 1).71 These results indicate that the
and [Gd(Bz-CB-TTDA)(H2O)]2- obtained by fitting 17
O
NMR and NMRD data are shown in Table 5. The
298
order of τR values obtained by fitting 17O NMR and
NMRD data is [Gd(Bz-CB-TTDA)(H2O)]2->MS-325 >
[Gd(CB-TTDA)(H2O)]2- > [Gd(DTPA-BMA)(H2O)] >
[Gd(DTPA)(H2O)]2-. The 17O 1/T1 relaxation rate curve
for [Gd(Bz-CB-TTDA)(H2O)]2- is higher than that of
[Gd(CB-TTDA)(H2O)]2-, as shown in Figure 1B. This
result implies that the benzyl group on the carbon back-
bone of [Gd(Bz-CB-TTDA)(H2O)]2- is contributing to
system is in the fast exchange domain (kex . 1/T2m
)
and the slow exchange region does not exist over the
studied temperature range.71 Consequently, 1/T2r is
determined by the relaxation rate of the coordinated
water molecule (1/T2m), which is influenced by the
water residence time (τm = 1/kex), the longitudinal
electronic relaxation rate (1/T1e), and the scalar cou-
its τR value.3 For the small-molecular weight Gd(III)
complexes, the order of τR values is an index for proton
relaxivity.
Ultrafiltration Studies of Gd(III) Chelates to HSA. The
binding affinity of [Gd(Bz-CB-TTDA)(H2O)]2- to HSA
was evaluated by an ultrafiltration study across a mem-
brane with a 30 kDa MW cutoff. It was assumed that the
filtrate accurately represents the unbound [Gd(Bz-CB-
TTDA)(H2O)]2- in each sample. Concentrations of total
[Gd(Bz-CB-TTDA)(H2O)]2- in the initial solutions con-
taining protein and in the filtrates were determined by
298
pling constant (A/p).34
298
The kex
values for [Gd(CB-TTDA)(H2O)]2- and
[Gd(Bz-CB-TTDA)(H2O)]2- along with other structurally
related Gd(III) complexes are summarized in Table 5.
(66) Lincoln, S. F.; Merbach, A. E. Adv. Inorg. Chem. 1995, 42, 1–88.
ꢀ
(67) Toth, E.; Helm, L.; Merbach, A. E.; Hedinger, R.; Hegetschweiler,
K.; Janossy, A. Inorg. Chem. 1998, 37, 4104–4113.
€
ꢀ
(68) Muller, R. N.; Raduchel, B.; Laurent, S.; Platzek, J.; Pierart, C.;
Mareski, P.; Vander Elst, L. Eur. J. Inorg. Chem. 1999, 1949–1955.
(69) Brittain, H. G.; Desreux, J. F. Inorg. Chem. 1984, 23, 4459–4466.
(70) Abragam, A. The Principles of Nuclear Magnetism; Oxford University
Press: London, 1961.
ꢀ
ꢀ
(71) Caravan, P.; Ellison, J. J.; McMurry, T. J.; Lauffer, R. B. Chem. Rev.
1999, 99, 2293–2352.
(72) Toth, E.; Connac, F.; Helm, L.; Adzamli, K.; Merbach, A. E. Eur. J.
Inorg. Chem. 1998, 2017–2021.

1284 Inorganic Chemistry, Vol. 50, No. 4, 2011
Chang et al.
Figure 1. Temperature dependence of (A) transverse and (B) longitudinal 17O relaxation rates; (C) 17O chemical shifts at B = 9.4 T for [Gd(CB-
TTDA)(H2O)]2- (left) and [Gd(Bz-CB-TTDA)(H2O)]2- (right). The lines represent simultaneous least-squares fits to all data points displayed. (D) NMRD
profile (bottom) for [Gd(CB-TTDA)(H2O)]2- (left) and [Gd(Bz-CB-TTDA)(H2O)]2- (right).
measuring Gd(III) using ICP-MS. The extent of bind-
ing of [Gd(Bz-CB-TTDA)(H2O)]2- to HSA is shown in
Figure 2, and it represents the plot of the ratio of the
bound [Gd(Bz-CB-TTDA)(H2O)]2- per HSA molecule
(n) versus the concentration of unbound [Gd(Bz-CB-
TTDA)(H2O)]2-. The value of n increases with the in-
crease in molar concentration of unbound [Gd(Bz-CB-
TTDA)(H2O)]2-. A complete binding curve is indicated
by saturation plateau. The value at which n reaches the
saturation plateau would represent the total number of
occupied sites on HSA when it is saturated with GdL.16
The saturation-like profile has not been observed at 1.3
equiv of bound [Gd(Bz-CB-TTDA)(H2O)]2- per HSA.
However, previous studies suggest that it is possible to fit
the data to a stoichiometric model to estimate the step-
wise binding constant.73 The first three stepwise binding
constants of Gd(Bz-CB-TTDA)(H2O)]2- are shown
in Table 6 together with [Gd(TTDA-BOM)(H2O)]2-
(TTDA-BOM = 6-carboxymethyl-benzyloxymethyl-3,
10-di(carboxymethyl)-3,6,10-triazadodecanedioic acid)
(Chart 1) and MS-325. From Table 6, we conclude that
the affinity of [Gd(Bz-CB-TTDA)(H2O)]2- to HSA is
higher than that of [Gd(TTDA-BOM)(H2O)]2- but is
slightly lower than that of MS-325. Hence, the intro-
duction of the benzyl and cyclobutyl groups into
TTDA results in a strong binding of this GdL chelate
to HSA.
Lipophilicity Study. Lipophilicity is well-known as a
prime physicochemical descriptor of drug potency. As
examples, drug absorption, plasma protein binding, and
hydrophobic drug-receptor interactions are all corre-
lated with lipophilicity. The lipophilic residue facilitates
noncovalent interaction of MS-325 to the lipid binding
(73) Klotz, I. M. Ligand Receptor Energetics-A Guide for the Perplexed;
John Wiley & Sons: New York, 1987.

Article
Inorganic Chemistry, Vol. 50, No. 4, 2011 1285
Table 5. Kinetic and NMR Parameters of [Gd(Bz-CB-TTDA)(H2O)]2-, [Gd(CB-TTDA)(H2O)]2-, [Gd(TTDA)(H2O)]2-, [Gd(EPTPA-bz-NO2)(H2O)]2-, [Gd(DTPA-
BMA)(H2O)], MS-325, and [Gd(DTPA)(H2O)]2-a
c
parameter
Bz-CB-TTDA CB-TTDA
TTDAb
EPTPA-bz-NO2
DTPA-BMAd
MS-325e, f
DTPAd
3.3 ( 0.2
51.6 ( 1.4
53.0 ( 4.7
-3.8 ( 0.2
kex298 (106 s-1
ΔH‡ (kJ mol-1
)
271 ( 13
232 ( 14
146 ( 17
23.1 ( 0.5
-11.1 ( 3.1 -9.1 ( 5.0
-3.2 ( 0.3
150 ( 40
22.1 ( 1.8
0.45 ( 0.01
47.6 ( 1.1
22.9 ( 3.6
-3.8 ( 0.2
6.1
)
23.1 ( 0.1
-5.9 ( 0.3
-3.5 ( 0.2
23.5 ( 0.4
-5.8 ( 1.0
-3.7 ( 0.1
53.7
65
-4.46
ΔS‡ (J mol-1 k-1
)
A/p (106 rad s-1
τR298 (ps)
)
-3.2 ( 0.2g
122 ( 12
2.2 ( 1
3.10
124 (151 ( 3) 89 (112 ( 2) 104 ( 12
5.73
3.10
66 ( 11 (167 ( 5) 110h or 142h (188) 58 ( 11 (103 ( 10)
τv (10-11 s)
rGd-H (A)
4.91
3.10
2.5 ( 1
3.10
41
0.11 ( 0.04
1
2.2
3.17
7.1
0.23
1
2.5 ( 1
3.10
46
0.18 ( 0.04
1
3.10
Δ2 (1018 s-2
Cos
)
8.31
0
1.14
0
40
0
1
q
1
1
1
ER (kJ mol-1
method
)
27.2 ( 1.1
13.6 ( 0.8
24.8 ( 1.5
19.0 ( 1.7
21.9 ( 0.5
31.5
17O, EPR, NMRD 17O,EPR, NMRD 17O, NMRD
17.3 ( 0.8
17O, NMRD
17O, NMRD 17O
17O, NMRD
a As obtained from the simultaneous fitting of 17O NMR and NMRD data. b Data were obtained from ref 33. c Data were obtained from ref 44. d Data
were obtained from ref 12. e Data were obtained from ref 16. f Data were obtained from ref 68. g The empirical constant for the outer-sphere contribution
to the chemical shift was fixed at Cos = 0.1. h The first values correspond to the results of the fitting performed with r set to 0.295 nm, the second ones
correspond to r = 0.31 nm. The τR298 values obtained by fitting only 17O NMR data are given in parentheses. Bold values are fixed in the fit.
Table 6. Association Constants for the Binding of [Gd(Bz-CB-TTDA)(H2O)]2-
,
[Gd(TTDA-BOM)(H2O)]2-, and MS-325 to HSA in Phosphate Buffer (pH 7.4) at
37.0 ( 0.1 ꢀC
[Gd(Bz-
CB-TTDA)
[Gd(TTDA-
BOM)
binding
constant
(H2O)]2- (M-1
)
(H2O)]2- (M-1 a
)
MS-325 (M-1 b
)
Ka1
Ka2
Ka3
Ka4
(1.3 ( 0.9) ꢀ 103
(1.0 ( 0.7) ꢀ 102
(0.3 ( 0.1) ꢀ 102
(5.9 ( 1.6) ꢀ 102
1.1 ꢀ 104
8.4 ꢀ 102
2.6 ꢀ 102
4.3 ꢀ 102
(0.9 ( 0.3) ꢀ 102
a Data were obtained from ref 34. b Data were obtained from ref 16.
HPLC method is an indirect way to estimate log P values
in the range of 0-6.78 However, the HPLC chromato-
gramsof[Gd(CB-TTDA)(H2O)]2- and[Gd(Bz-CB-TTDA)-
(H2O)]2- in the octanol phase could not be detected (data
not shown). These results indicate that [Gd(CB-TTDA)-
(H2O)]2- and [Gd(Bz-CB-TTDA)(H2O)]2- have low
lipophilicity.75 Recently, several computer programs
have been used to predict the lipophilicity of molecules.
These programs are based on atom/fragment contribu-
tions, structural parameters, atom-type electrotopologi-
cal-state indices, and topological structure descriptors.79
We used the miLogP 1.2 online program80 to calculate the
log P values of TTDA, CB-TTDA, Bz-TTDA (4-benzyl-3,
6,10-tri(carboxymethyl)-3,6,10-triazadodecanedioic acid),37
Bz-CB-TTDA, and MS-325 ligand, and the values are
shown in Table 7. The software calculates log P values
as a sum of group contributions and correction factors.
The sequence of increase in lipophilicity is TTDA < CB-
TTDA < Bz-TTDA < Bz-CB-TTDA < MS-325 ligand.
The trend in the increase of lipophilicity indicates that
both cyclobutyl and benzyl residues contribute to the
lipophilicity of ligands, and their cumulative effect on the
lipophilicity can be observed in Bz-CB-TTDA. Further-
more, the partition coefficient of the [Gd(Bz-CB-
TTDA)(H2O)]2- complex was determined following a
previously reported method.41 The log P value of
[Gd(Bz-CB-TTDA)(H2O)]2- is -2.34 ( 0.15 which is
Figure 2. Plot of the ratio of bound [Gd(Bz-CB-TTDA)(H2O)]2- per
HSA versus the concentration of free [Gd(Bz-CB-TTDA)(H2O)]2- (37.0
( 0.1 ꢀC, phosphate buffer, pH 7.4).
sites of HSA and significantly augments its plasma half-
life. Hydrophilic-lipophilic balance (HLB) of the Gd(III)
complex is a vital factor controlling the plasma half-life
and elimination from the organism.74 To augment the
lipophilicity of the [Gd(TTDA)(H2O)]2-, we introduce
lipophilic residues such as cyclobutyl and benzyl groups
on the carbon backbone. The octanol-water distribution
ratio, KO/W, is the accepted physicochemical property
measuring the hydrophobicity of Gd(III) complexes.75,76
The log P (logarithm of partition coefficient in n-octanol/
water) values can estimate the lipophilicity of a compound
in a biological environment,77 and thus, itis directly related
to the binding affinity of Gd(III) complexes to HSA. The
(74) Hifumi, H.; Tanimoto, A.; Honda, A.; Citterio, D.; Suzuki, K. Anal.
Sci. 2007, 23, 1159–1165.
(75) Jastrzebska, B.; Lebel, R.; Therriault, H.; McIntyre, J. O.; Escher, E.;
ꢀ
Guerin, B.; Paquette, B.; Neugebauer, W. A.; Lepage, M. J. Med. Chem.
2009, 52, 1576–1581.
(78) Chen, Z.; Weber, S. G. Anal. Chem. 2007, 79, 1043–1049.
ꢀ ꢀ
(79) Medic-Saric, M.; Mornar, A.; Jasprica, I. Acta Pharm. 2004, 54,
(76) Sangster, J. Octanol-Water Partition Coefficients, Fundamentals and
Physical Chemistry; John Wiley & Sons: Chichester, U.K., 1997.
(77) Van de Waterbeemd, H.; Gifford, E. Nat. Rev. Drug Discovery 2003,
2, 192–204.
91–101.
(80) Molinspiration Cheminformatics Calculation of Molecular Proper-
ties and Drug-likeness. http://www.molinspiration.com (accessed 2009).

1286 Inorganic Chemistry, Vol. 50, No. 4, 2011
Chang et al.
Table 7. log p Values for TTDA, CB-TTDA, Bz-TTDA, Bz-CB-TTDA, and MS-
325 Ligand, Computed by miLogP 1.2 Computer Software
ligand
TTDA
CB-TTDA
Bz-TTDA
Bz-CB-TTDA
MS-325 ligand
miLogP
-5.539
-5.432
-5.055
-4.875
-4.575
slightly lower than that of MS-325 (-2.11 ( 0.06). This
is consistent with ultrafiltration study which provides
an accurate estimation of [Gd(Bz-CB-TTDA)(H2O)]2-
binding constants to HSA with a lower affinity than
that of MS-325. This result implies that [Gd(Bz-CB-
TTDA)(H2O)]2- also has the optimum lipophilicity required
for a blood pool agent. Excess lipophilic groups on the
carbon backbone of [Gd(TTDA)(H2O)]2- would have
accelerated hepatobiliary excretion from the liver. The
relevance of carbon backbone modification by appropri-
ate lipophilic residues to impart optimum lipophilicity to
Gd(III) complexes is clearly supported by the current
findings.
Figure 3. E-titration of a 0.1 mM solution of [Gd(Bz-CB-TTDA)-
(H2O)]2- (2) and [Gd(CB-TTDA)(H2O)]2- ([) with HSA (20 MHz,
25.0 ( 0.1 C, 50 mMphosphate buffer, pH7.4). The solid curves represent
the best fit.
Relaxivity Studies of Gd(III) Chelates to HSA. Macro-
molecular adducts formed as a result of noncovalent
interaction of Gd(III) complexes to HSA have higher
relaxivity than that of free Gd(III) complexes. Enhance-
ment in the relaxivity upon binding to HSA is primarily
because of remarkable reduction in the rotational diffu-
sion rate of the molecules. In addition, binding of Gd(III)
complexes to HSA slows down the renal excretion rate,
and consequently, the prolonged blood plasma half-life
facilitates the angiographic applications. The advantage
of this concept is that a lesser amount of Gd(III) complex
has to be injected into the patient. Moreover, the blood
pool agent concentration remains relatively stable in
plasma during examination and, consequently, allows
imaging of multiple body regions without repeated
doses. Hydrophobic residues as well as the negative
electric charges are the basic requirement for binding of
Gd(III) complexes to HSA.81 The binding affinity of
[Gd(CB-TTDA)(H2O)]2- and [Gd(Bz-CB-TTDA)(H2O)]2-
to HSA was evaluated by proton relaxation enhancement
method.82 This technique allows us to obtain reliable and
reproducible physical values such as the thermodynamic
binding constant (KA), the number of binding sites (n),
and the bonded relaxivity (rb1) of the macromolecular
adduct. The experimental procedure involves two distinct
titrations, called E- and M-titration83 (see the Supporting
Information). The quantitative analysis of E-titration
provides an accurate estimation of bound relaxivity (rb1)
of Gd(III) complexes to HSA with higher affinity. How-
ever, E-titration is not sensitive to n particularly when KA
is small. The quantitative analysis of M-titration provides
independent evaluation of KA and n.82 The results of
E- and M-titrations for [Gd(CB-TTDA)(H2O)]2- and
[Gd(Bz-CB-TTDA)(H2O)]2- in the presence of HSA are
shown in Figures3and4, respectively. [Gd(Bz-CB-TTDA)-
Figure 4. Scatchard plot of the data obtained from the PRE titration of
a 0.6 mM HSA solution (pH 7.4, HEPES buffer, 25.0 ( 0.1 ꢀC, 20 MHz)
with [Gd(Bz-CB-TTDA)(H2O)]2- (2) and [Gd(CB-TTDA)(H2O)]2-
([); r = [GdL-HSA]/[HSA]T.
(H2O)]2- displays steep enhancement in the longitudinal
water proton relaxation rate upon addition of HSA
owing to the formation of macromolecular adducts. On
the other hand, the relaxation rate enhancement shown
by [Gd(CB-TTDA)(H2O)]2- is much less in the presence
of HSA and increases linearly with concentration, which
reflects weak protein interaction.83 This result implies
that the HSA has higher affinity for [Gd(Bz-CB-TTDA-
(H2O)]2- than[Gd(CB-TTDA)(H2O)]2-. The high affinity
ofHSAtoward[Gd(Bz-CB-TTDA)(H2O)]2- can be explained
in terms of hydrophobicity. The co-presence of benzyl
and cyclobutyl moieties in [Gd(Bz-CB-TTDA)(H2O)]2-
induces much higher hydrophobicity than cyclobutyl
alone in [Gd(CB-TTDA)(H2O)]2-. The effect of hydro-
phobicity of the Gd(III) complex on the affinity toward
HSA can be evaluated by comparing the KA value of the
Gd(III) complex.84 The KA value of [Gd(Bz-CB-TTDA)-
(H2O)]2- is almost 1 order of magnitude higher than that
of [Gd(CB-TTDA)(H2O)]2-. The binding parameters of
[Gd(CB-TTDA)(H2O)]2- and [Gd(Bz-CB-TTDA)(H2O)]2-
calculated for the single binding site (n = 1) on HSA are
presented in Table 8, along with those of other linear
(81) Peters, T. J. All About Albumin: Biochemistry, Genetics, and Medical
Applications; Academic Press: San Diego, CA, 1996.
(82) Dwek, R. A. Nuclear Magnetic Resonance in Biochemistry, Applica-
tion to Enzyme System; Clarendon Press: Oxford, 1973.
(83) Aime, S., Fasano, M., Terreno, E., Botta, M., Eds. The Chemistry of
Contrast Agents in Medical Magnetic Resonance Imaging; John Wiley and
Sons: New York, 2001; pp 193-241.
(84) Aime, S.; Botta, M.; Fasano, M.; Geninatti Crich, S.; Terreno, E.
J. Biol. Inorg. Chem. 1996, 1, 312–319.
(85) Aime, S.; Chiaussa, M.; Digilio, G.; Gianolio, E.; Terreno, E. J. Biol.
Inorg. Chem. 1999, 4, 766–774.

Article
Inorganic Chemistry, Vol. 50, No. 4, 2011 1287
Table 8. Binding Parameters to HSA from PRE Measurements (25.0 ( 0.1 ꢀC, 20 MHz, pH 7.4, 50 mM Phosphate Buffer)a
complexes
KA (M-1
)
n
εb
rf1ree (mM-1 s-1
)
rb1 (mM-1 s-1
)
[Gd(Bz-CB-TTDA)(H2O)]2-
[Gd(CB-TTDA)(H2O)]2-
[Gd(TTDA-BOM)(H2O)]2-b
[Gd(BOPTA)(H2O)]2-c
MS-325 d
(1.5 ( 0.1) ꢀ 103
(1.1 ( 0.1) ꢀ 102
(4.6 ( 0.1) ꢀ 102
(4.0 ( 0.1) ꢀ 102
(3.0 ( 0.2) ꢀ 104
1
1
1
13.1 ( 0.4
6.1 ( 0.2
13.7 ( 0.1
5.1 ( 0.3
4.8 ( 0.1
4.8 ( 0.02c
66.7 ( 2.2
29.3 ( 0.8
65.8 ( 2.7
33
1
6.6 ( 0.4c
47.0 ( 4
a KA = [GdL - HSA]/[GdL]F[nHSA]F, where [nHSA] represents the binding sites concentration, the subscripts T and F refer to “total” and “free”
species, respectively. rb1 = εb ꢀ r1free, where, εb refers to maximum value of the enhancement factor ε*, and r1free and r1b refer to the relaxivity of free and
bound complex, respectively. b Data were obtained from ref 34. c Data were obtained from ref 84. d Data were obtained from ref 85.
Table 9. Values of R1m (t = 3 d)/R1m (t = 0) for the Gd(III) Complexes after 3
Days of Transmetalation with Zn(II) at 20 MHz and 37.0 ( 0.1 ꢀC
poly(aminocarboxylate) complexes. The KA value for
[Gd(Bz-CB-TTDA)(H2O)]2- is lower than that of MS-
325 but is higher than those of [Gd(CB-TTDA)(H2O)]2-
,
complex
R1m (t = 3 d)/R1m(t = 0) (%)
[Gd(TTDA-BOM)(H2O)]2-, and [Gd(BOPTA)H2O]2-
(BOPTA=4-carboxy-5,8,11-tris(carboxymethyl)-1-phenyl-
2-oxa-5,8,11-triazatridecan-13-oic acid). The bound
relaxivity (rb1) of [Gd(Bz-CB-TTDA)H2O]2- obtained
by multiplying b and rF1 is significantly higher than that
of MS-325, which can be explained in terms of the water
exchange rate on these two Gd(III) complexes. The water
exchange rate on the Gd(III) complex generally slows
down by a factor of 2-3-fold upon binding to HSA.16 The
decrease in water exchange rate upon binding to HSA
may be more effectively borne by [Gd(Bz-CB-TTDA)-
(H2O)]2- than by MS-325, owing to higher water exchange
rate on [Gd(Bz-CB-TTDA)(H2O)]2- (271 ꢀ 106 s-1) than
that of MS-325 (5.8 ꢀ 106 s-1).
[Gd(Bz-CB-TTDA)(H2O)]2-
[Gd(CB-TTDA)(H2O)]2-
[Gd(TTDA-BOM)(H2O)]2-a
[Gd(DTPA)(H2O)]2-a
46.14
47.40
43.90
49.79
9.00
[Gd(DTPA-BMA)(H2O)]b
a Data were obtained from ref 34. b Data were obtained from ref 43.
respect to transmetalation. Therefore, the kinetic and thermo-
dynamic stabilities of [Gd(Bz-CB-TTDA)(H2O)]2- and
[Gd(CB-TTDA)(H2O)]2- are stable enough to be used as
contrast agents for MRI.
Conclusion
We have synthesized and characterized two TTDA deri-
vatives which possess cyclobutyl and/or benzyl moieties.
Both complexes, [Gd(CB-TTDA)(H2O)]2- and [Gd(Bz-CB-
TTDA)(H2O)]2-, display high stability constants, and CB-
TTDA and Bz-CB-TTDA show superior selectivity of Gd-
(III) against Zn(II) than TTDA and DTPA but inferior to
MS-325. The simultaneous treatment of 17O NMR and
NMRD data is used to obtain several parameters affecting
the proton relaxivity of Gd(III) complexes. It is found that
the introduction of the cyclobutyl group increases the steric
constraint at the water binding site and thereby increases the
water exchange rate of [Gd(Bz-CB-TTDA)(H2O)]2- and
[Gd(CB-TTDA)(H2O)]2-. From the ultrafiltration studies,
[Gd(Bz-CB-TTDA)(H2O)]2- has a higher affinity to HSA
than [Gd(TTDA-BOM)(H2O)]2- but is slightly lower than
MS-325. However, the rb1 value of the [Gd(Bz-CB-TTDA)-
(H2O)]2- complex has a remarkably high value with HSA. The
kinetic stability of [Gd(CB-TTDA)(H2O)]2- and [Gd(Bz-
CB-TTDA)(H2O)]2- complexes toward transmetalation
Transmetalation. The stability of [Gd(Bz-CB-TTDA)-
(H2O)]2- and [Gd(CB-TTDA)(H2O)]2- complexes against
the presence of Zn(II) ions was determined by trans-
metalation experiments carried out in phosphate buffer
(pH 7.4) in the presence of ZnCl2 as the competing ionic
species. Replacement of Gd(III) in the complex with
Zn(II) leads to free gadolinium ions, which precipitate
out in the presence of phosphate as GdPO4, and conse-
quently, the total amount of paramagnetic species in
solution decreases.86 The estimation of the decrease in
relaxivity allows us to track the transmetalation reaction.
The two characteristic values can be used to describe the
behavior of a Gd(III) chelate in a transmetalation experi-
ment: the time to reach R1m/R1,0 = 0.8 (the ratio index)
which gives information about the kinetics of the reaction
and R1m(t)/R1m(0) value at very long time, t = ¥ (the long
time index, considered after 3 days), that reflects the thermo-
m
dynamic aspect of the system (R1 and R1,0 are the
with Zn(II) is similar to that of [Gd(DTPA)(H2O)]2-
.
relaxation rate at time t and at time 0, respectively).42,43
The percentages of Gd(III) complexes left in the solution
after 3 days for [Gd(Bz-CB-TTDA)(H2O)]2- and [Gd-
(CB-TTDA)(H2O)]2- are summarized in Table 9 together
Therefore, [Gd(Bz-CB-TTDA)(H2O)]2- may potentially be
used as a blood pool contrast agent for MRI.
Acknowledgment. Funding from National Science
Council of Taiwan (Grant Nos. NSC 98-2627-M-009-009
and NSC 97-2113-M-009-016-MY3) is gratefully ac-
knowledged.
with [Gd(TTDA-BOM)(H2O)]2-, [Gd(DTPA)(H2O)]2-
,
and [Gd(DTPA-BMA)(H2O)]2-. Figure 7S in the Sup-
porting Information shows R1m(t)/R1m(0) versus time for
[Gd(DTPA)(H2O)]2-, [Gd(CB-TTDA)(H2O)]2-, and [Gd(Bz-
CB-TTDA)(H2O)]2-. The transmetalation reaction rate
Supporting Information Available: Additional experimental
details, figures, and tables. This material is available free of
charge via the Internet at http://pubs.acs.org.
for[Gd(Bz-CB-TTDA)(H2O)]2-, [Gd(CB-TTDA)(H2O)]2-
,
[Gd(TTDA-BOM)(H2O)]2-, and [Gd(DTPA)(H2O)]2- is
similar and is five times slower than that of [Gd(DTPA-
BMA)(H2O)]2-. This result implies that [Gd(Bz-CB-TTDA)-
(H2O)]2- and [Gd(CB-TTDA)(H2O)]2- are fairly stable with
Note Added after ASAP Publication. This paper was
published ASAP on January 19, 2011, with errors in
production that included the absence of Schemes 1 and 2
and part of Figure 1. The corrected version was published
on January 25, 2011.
ꢂ
(86) Moriggi, L.; Cannizzo, C.; Prestinari, C.; Berriere, F.; Helm, L.
Inorg. Chem. 2008, 47, 8357–8366.