A calcium-sensitive magnetic resonance imaging contrast agent [18]

Full text of DOI:10.1021/ja983702l

J. Am. Chem. Soc. 1999, 121, 1413-1414
1413
A Calcium-Sensitive Magnetic Resonance Imaging
Contrast Agent
Wen-hong Li, Scott E. Fraser, and Thomas J. Meade*
DiVision of Biology and the Beckman Institute
California Institute of Technology
Pasadena, California 91125
ReceiVed October 23, 1998
ReVised Manuscript ReceiVed January 4, 1999
Ca2+ is an important intracellular secondary messenger of signal
transduction.1 Changes in the cytosolic concentration of Ca2+ trig-
ger changes in cellular metabolism and are responsible for cell
signaling and regulation.2 Advances in optical microscopy tech-
niques and improvements in fluorescent dyes capable of measur-
ing Ca2+ concentration have added greatly to the understanding
of the critical role this ion plays in cell and neurobiology.3 How-
ever, a fundamental limitation of light-based microscope imaging
techniques employing dyes or fluorochromes is that they produce
toxic photobleaching byproducts and are limited by light scattering
to those cells within 100 µm of the surface. Magnetic resonance
imaging (MRI) of biological structures provides an alternative to
light-based microscopy that can circumvent these limitations.
Recent work in this area has demonstrated the feasibility of true
three-dimensional MR imaging at cellular resolution (∼10 µm).4
As part of our efforts to study cell signaling and regulation in
intact animals, we are developing MRI contrast agents that provide
information about physiological signals and biochemical events.
To this end, we have prepared and tested a class of enzymatically
activated MRI contrast agents that conditionally enhance image
intensity.5 Here, we report the first MRI contrast agent, DOPTA-
Gd (Figure 1), whose relaxivity is selectively modulated by Ca2+
concentration.6 DOPTA-Gd has a Ca2+ dissociation constant of
0.96 µM, and the relaxivity of the complex increases ap-
proximately 80% when Ca2+ is added to a Ca2+-free solution.
The most abundant molecular species in biological tissues is
water. It is the quantum mechanical “spin” of the water proton
nuclei that ultimately gives rise to the signal in all imaging
experiments. MRI agents enhance the intrinsic differences in the
T1 (spin-lattice) and T2 (spin-spin) relaxation rates. The class
of contrast agents referred to as T1 agents accelerate the T1
relaxation rate, increasing the signal from nearby water protons
and making a voxel appear “brighter” in the resulting image. The
increase in relaxation rate is due, in part, to the direct interaction
of water molecules (inner sphere) with the unpaired electrons of
Figure 1. Schematic of DOPTA-Gd representing the proposed conform-
ational dependence of the structure in the presence and absence of Ca2+
.
a paramagnetic metal ion. The lanthanide ion, Gd3+, is frequently
chosen for MRI contrast agents because it has a very high
2
magnetic moment (µ2 ) 63 µB ), and a symmetric electronic
8
ground state, S7/2. The Gd3+ aqua ion is toxic and hence is
chelated to a ligand in order to reduce toxicity. Typically, eight
of the nine available Gd3+ coordination sites are bound by the
chelate, leaving one site available for an inner sphere water
molecule.5,7
This new class of MRI agents modulates access of water to a
chelated Gd3+ ion in the presence and absence of Ca2+. The design
of the agent is based on the synthesis and characterization of
several model systems that ultimately led to the macrocyclic dimer
shown in Figure 1. 1,2-Bis(o-aminophenoxy)ethane-N,N,N′,N′-
tetraacetic acid (BAPTA) binds Ca2+ with a 105-fold selectivity
versus the divalent metal ion Mg2+ and is relatively insensitive
to pH fluctuations at physiological conditions (>6.8).8 1,4,7-Tris-
(carboxymethyl)-1,4,7,10-tetraazacyclododecane (DO3A) chelates
lanthanides with high affinity to form a thermodynamically stable
and kinetically inert complex.9 DOPTA-Gd (Figure 1) was
designed to possess two limiting conformational states with
respect to calcium concentration ([Ca2+]). In the absence of Ca2+
,
the aromatic iminoacetates of BAPTA interact with Gd3+ through
ionic attractions. In the presence of Ca2+, the aromatic iminoac-
etates of BAPTA will rearrange to bind Ca2+, thereby allowing
water to bind directly to Gd3+
.
On the basis of complexes synthesized as part of model studies,
a propyl linker was chosen to covalently connect DO3A to BAP-
TA.10 The propyl linker places the aromatic iminoacetates of
BAPTA in the proximity of the chelated Gd3+. Therefore, the
aromatic iminoacetates shield the Gd3+ ion from water when the
Ca2+ concentration is low. Upon binding Ca2+, the complex
undergoes reorganization that exposes the Gd3+ ion to bulk water,
thereby changing the relaxivity from weak to strong.
DOPTA-Gd was synthesized from nitroresorcinol in 8 steps
(Figure 2). The monoalkylated nitroresorcinol 1 was dimerized
with 1,2-dibromoethane. The free hydroxyls of 4 were converted
into dibromides 5. Excess cyclen was used to minimize the
formation of the bisalkylated product during the synthesis of 6.
(1) Tsien, R. W.; Tsien, R. Y. Annu. ReV. Cell Biol. 1990, 6, 715-760.
(2) (a) Li, W. H.; Llopis, J.; Whitney, M.; Zlokarnik, G.; Tsien, R. Y. Nature
1998, 392, 936-941. (b) Meldolesi, J. Nature 1998, 392, 863. (c) Berridge,
M. J. Neuron 1998, 21, 13-26.
(3) (a) Tsien, R. Y.; Waggoner, A. In Handbook of Biological Confocal
Microscopy; 2nd ed.; Pawley, J. B., Ed.; Plenum Press: New York, 1995; pp
267-280. (b) Tsien, R. Y. Chem. Eng. News 1994, 72, 34-44. (c) Czarnik,
A. W., Ed. ACS Symposium Series 538; American Chemical Society:
Washington, DC, 1993.
(4) (a) Huber, M. M.; Staubli, A. B.; Kustedjo, K.; Gray, M. H. B.; Shih,
J.; Fraser, S. E.; Jacobs, R. E.; Meade, T. J. Bioconjugate Chem. 1998, 9,
242-249. (b) Bowtell, R. W.; Sharp, J. C.; Mansfield, P.; Hsu, E. W.; Aiken,
N.; Horsman, A. Magn. Reson. Med. 1995, 33, 790-794. (c) Kayyem, J. F.;
Kumar, R. M.; Fraser, S. E.; Meade, T. J. Chem. Biol. 1995, 2, 615-620. (d)
Liang, P.; Lauterbur, P. C. IEEE Trans. Med. Imaging 1994, 13, 677-686.
(e) Jacobs, R. E.; Fraser, S. E. Science 1994, 263, 681-684. (f) Mellin, A.
F.; Cofer, G. P.; Smith, B. R.; Suddarth, S. A.; Hedlund, L. W.; Johnson, G.
A. Magn. Reson. Med. 1994, 13, 677-686. (g) Kuhn, W. Angew. Chem., Int.
Ed. Engl. 1990, 29, 1-19.
(7) (a) Aime, S.; Botta, M.; Fasano, M.; Terreno, E. Chem. Soc. ReV. 1998,
27, 19-29. (b) Alexander, V. Chem. ReV. 1995, 95, 273-342.
(8) (a) Grynkiewicz, G.; Poenie, M.; Tsien, R. Y. J. Biol. Chem. 1985,
260, 3440-3450. (b) Tsien, R. Y. Biochemistry 1980, 19, 2396-2404.
(9) Kumar, K.; Chang, C. A.; Tweedle, M. F. Inorg. Chem. 1993, 32, 587-
593.
(10) Model complexes with varying linker lengths were prepared to
optimize the modulation of water access to the Gd3+ ion based on data from
T1 and luminescence lifetime measurements.
(5) Moats, R. A.; Fraser, S. E.; Meade, T. J. Angew. Chem., Int. Ed. Engl.
1997, 36, 726-728.
(6) Fluorinated BAPTA derivatives have been described. The chemical
shifts of 19F signals of these compounds are affected by Ca2+ binding and
NMR spectroscopy provides Ca2+ concentration. (a) Clarke, S. D.; Metcalfe,
J. C.; Smith, G. A. J. Chem. Soc., Perkin Trans. 2 1993, 1187-1194. (b)
Marban, E.; Kitakaze, M.; Koretsune, Y.; Yue, D. T.; Chacko, V. P.; Pike,
M. M. Circ. Res. 1990, 66, 1255-1267.
10.1021/ja983702l CCC: $18.00 © 1999 American Chemical Society
Published on Web 01/29/1999

1414 J. Am. Chem. Soc., Vol. 121, No. 6, 1999
Communications to the Editor
Figure 2. Synthetic scheme of DOPTA-Gd. (a) NaOH, DMF, 3-bromopropanol, 42%. (b) K2CO3, DMF, 1,2-dibromoethane, 55%. (c) Pd/C (10%),
EtOH/EtOAc, H2, 1 atm. (d) Bromoethyl acetate, DIEA, CH3CN, 61% for 2 steps. (e) CBr4, PPh3, Et2O, 60%. (f) Cyclen (5 equiv), CHCl3, 80%. (g)
1. Bromoacetate, NaOH, pH > 10, 82%; 2. GdCl3, NaOH, pH 5-6, 81%.
insensitive to [Mg2+] change. Indeed, increasing Mg2+ concentra-
tion from 0 to 10 mM changed the relaxivity of DOPTA-Gd
less than 8%.12 Intracellular [Mg2+] is approximately 1 mM, and
its fluctuation is less dynamic than [Ca2+]; therefore, interference
on the [Ca2+] measurements by DOPTA-Gd should be minimal.
The effect of H+ on the observed relaxivity of DOPTA-Gd
was also tested. Changing the pH from 6.80 to 7.40 (in 0.2 pH
unit steps) changed the measured T1 of DOPTA-Gd by less than
3% (in the presence or absence of Ca2+).13 Therefore, within
physiological pH ranges, H+ should not interfere with the
relaxivity of the complex with respect to [Ca2+].
In summary, we have synthesized a MRI contrast agent where
the relaxivity of the complex is controlled by the presence or
absence of the divalent ion Ca2+. By structurally modulating inner-
sphere access of water to a chelated Gd3+ ion we observe a
substantial change in T1 upon the addition of Ca2+. Importantly,
Figure 3. Relaxivity measurement of DOPTA-Gd as a function of free
the agent is selective for binding Ca2+ ions versus Mg2+ and H+.
[Ca2+].11 The fitted curve corresponds to the apparent dissociation constant
An immediate application of this agent is to study the change in
cellular calcium activity during embryogenesis. The agent can
be conveniently microinjected inside cells at an early develop-
mental stage, and subsequent cell movements and calcium
fluctuations during the development can be monitored over long
periods of time. These investigations may help to resolve current
uncertainties concerning cellular Ca2+ activities of interior cell
layers not accessible to light microscopy. Finally, this new class
of contrast agent offers a large and reliable change in relaxivity
upon exposure to Ca2+. These results serve as an ideal template
for the future design of biochemically activated MRI contrast
agents.
) 0.96 µM and Hill coefficient ) 0.92.
The remaining secondary amines were alkylated under basic
conditions that concomitantly hydrolyzed the aromatic imino ethyl
diacetates. The final Gd3+ complex was formed in good yield
under weakly acidic conditions.
The effect of [Ca2+] on the relaxivity of DOPTA-Gd was
assessed by T1 measurements.11 The relaxivity of DOPTA-Gd
in Ca2+-free buffer was 3.26 mM-1 sec-1 and increased with
increasing [Ca2+]. The change in relaxivity is most striking in
the [Ca2+] range of 0.1 to 10 µM and levels off at higher levels
of [Ca2+], reaching a maximum of 5.76 mM-1 sec-1 (Figure 3).
Hill-plot analysis of the measured relaxivities at varying [Ca2+
]
results in a dissociation constant of 0.96 µM (Hill coefficient of
Acknowledgment. This work was supported by the Biological
Imaging Center of the Beckman Institute and the Human Brain Project
with contributions from the National Institute on Drug Abuse, the National
Institute of Mental Health, and the National Science Foundation.
0.92). These results reveal a 1:1 stoichiometry of binding.8
The increase in observed relaxivity of DOPTA-Gd (∼80%)
that is induced by an increase in [Ca2+] corresponds to a 80%
relaxivity change of each Gd3+ unit and is significantly higher
than our previously reported enzyme-reporter class of agents (E-
Gad: 20% change of T1 at a [Gd3+] of 2 mM after enzyme cleav-
age).5 Since BAPTA binds Ca2+ with a much higher selectivity
than Mg2+, the relaxivity of DOPTA-Gd should be relatively
JA983702L
(12) The longitudinal relaxation rate of DOPTA-Gd (0.3 mM) was
measured in a buffer containing 130 mM KCl, 10 mM MOPS and 1 mM
EGTA, pH 7.20, 25 °C. Mg2+ concentration was adjusted to 2, 5, and 10 mM
using the reported procedures.8a The relaxivity (3.23 mM-1 sec-1, 0 Mg2+
;
(11) [Ca2+] was controlled by Ca2+/EGTA and Ca2+/HEEDTA systems
below 10-5 M free Ca2+ and unbuffered Ca2+ above.8 The buffers contained
100 mM KCl, 10 mM KMOPS, pH 7.20, 20 mM of EGTA or HEEDTA. The
T1 was measured by using the standard inversion-recovery procedure (Bruker
AMX 500, 25 °C).5 The relaxivity of DOPTA-Gd was determined from the
slope of the plot of 1/T1 vs [DOPTA-Gd] (0, 0.2, 0.3, 0.4 mM).
3.30 mM-1 sec-1, 2 mM Mg2+; 3.39 mM-1 sec-1, 5 mM Mg2+; 3.47 mM-1
sec-1, 10 mM Mg2+) was calculated from the T1 after subtracting the
longitudinal relaxation rate in the absence of DOPTA-Gd.
(13) The T1 was measured in buffers containing 100 mM KCl, 5 mM MO-
PS, 0.3 mM DOPTA-Gd with either 2 mM EGTA or 1 mM CaCl2 at 25 °C.
The pH was varied from 6.80 to 7.40 in 0.2 pH unit steps with HCl or KOH.