Critical in vitro evaluation of responsive MRI contrast agents for calcium and zinc

Article abstract of DOI:10.1002/chem.201001548

The synthesis of two gadolinium(III) complexes that exhibit an increase in proton relaxivity in the presence of added Ca²⁺ or Zn²⁺ ions is reported. The complexes increase their hydration state from zero to one following metal-ion bi

Full text of DOI:10.1002/chem.201001548

FULL PAPER
DOI: 10.1002/chem.201001548
Critical In Vitro Evaluation of Responsive MRI Contrast Agents for Calcium
and Zinc**
Anurag Mishra,*[a, b] Nikos K. Logothetis,[b] and David Parker*[a]
Abstract: The synthesis of two
gadolinium(III) complexes that exhibit
1.4 T and 310 K, modulation of relaxiv-
ity of the order of 30–40% was ob-
served in mouse serum in each case.
The dissociation constants for Ca2+
and Zn2+ binding were sensitive to the
presence of added bicarbonate, and
were 450 mm (Ca2+) and 200 mm (Zn2+
)
G
an increase in proton relaxivity in the
presence of added Ca2+ or Zn2+ ions is
reported. The complexes increase their
hydration state from zero to one fol-
lowing metal-ion binding, confirmed by
spectral measurements on the corre-
sponding EuIII complexes. At a field of
in serum. Such systems may, therefore,
be considered for use as magnetic reso-
nance imaging (MRI) contrast agents
to track the restoration of changes in
metal-ion concentration in the cerebro-
spinal fluid of the brain, following
neural stimulation.
Keywords: calcium · gadolinium ·
imaging agents · lanthanides · zinc
Introduction
methods suffers from drawbacks of invasiveness and limited
depth penetration, stimulating a search for alternative meth-
ods to track changes with appropriate spatial and temporal
resolution. Amongst these, the development of responsive
magnetic resonance imaging (MRI) contrast agents shows
Calcium plays a pre-eminent role as a signalling ion in biol-
ogy, with an important function as a second messenger in
the brain. The concentration of calcium within cells falls
within the range of 20–100 nm, whereas in extracellular
fluids the concentration is of the order of 1.5 mm. Levels of
extracellular calcium decrease by between 15 and 30%
during normal brain function, whereas reductions of as
much as 90% in the cerebrospinal fluid can occur during
major trauma, epilepsy, or anoxia. This variation in extracel-
lular [Ca2+]0 in the brain is strongly linked to the action of
the parathyroid hormone, and is a significant determinant of
neural function in healthy and diseased states.[1,2] Observa-
tion of changes in ionisable [Ca2+]0 are typically undertaken
by using specific ion-selective microelectrodes or with the
aid of fluorescent calcium-selective probes. Each of these
considerable promise,[3,4] and systems responding selectively
[6,7]
to changes in [Ca2+],[3,4] copper,[5] and [Zn2+
devised.
]
have been
A key aspect of probe design is to ensure that the dissoci-
ation constant of the responsive contrast agent matches the
resting concentration of the target species in the local bio-
logical medium. This issue inhibits the pragmatic develop-
ment of calcium-sensitive MRI probes for intracellular
use.[3b] A similar inherent problem exists with copper-sensi-
tive probes, as the concentration of “kinetically labile” Cu2+
in vivo is likely to be extremely low (i.e., sub-nanomolar).[5]
It has been postulated that concentrations of zinc in cere-
brospinal fluid may rise from 10 nm to values as high as
250 mm, following stress-induced release of zinc ions from
neuronal synaptic vesicles, which suggests a possible applica-
tion for zinc-activated MRI probes.[6]
In this work, we have set out to create and evaluate MRI
contrast agents for Ca2+ and Zn2+ that exhibit an increase
in relaxivity on metal-ion binding to the probe. Thus, as the
local concentration of, for example, Ca2+ increases with
time upon returning to the normal higher equilibrium value,
the relaxivity of the probe must also increase to signal this
change. Obviously, the timescale for image acquisition must
be fast with respect to the time taken for the [Ca2+] to
return to its equilibrium value. A convenient way of engi-
[a] Dr. A. Mishra, Prof. D. Parker
Department of Chemistry, Durham University
South Road, Durham, DH1 3LE (UK)
Fax : (+44)1913342051
E-mail: anurag.mishra@dur.ac.uk
david.parker@dur.ac.uk
[b] Dr. A. Mishra, Prof. N. K. Logothetis
Max Planck Institute for Biological Cybernetics
Department of Physiology of Cognitive Processes
Spemannstrasse 38, Tubingen, 72076 (Germany)
[**] MRI=magnetic resonance imaging.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201001548.
Chem. Eur. J. 2011, 17, 1529 – 1537
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1529

of complexes with and without a bound water were identi-
fied in this work, and the hydration change was signalled
clearly by a reduction in the emission intensity ratios of the
DJ=2/DJ=1 bands, from 2:1 (q=1) to 3:1 (q=0).[11] Euro-
pium emission spectra are particularly sensitive to changes
in the coordination environment of the metal ion, especially
if the axial donor is varied.[9]
Scheme 1. Schematic representation of metal-ion responsive CAs.
Results and Discussion
neering this response is to develop a system in which the
contrast agent increases its hydration state from zero to one
following reversible metal-ion binding, (Scheme 1). In this
hypothetical case, the amide carbonyl oxygen binds to the
Gd3+ ion in the unactivated state to give a 7-ring chelate.
However, it preferentially coordinates to the added metal
ion (Ca2+ or Zn2+), liberating space for a water molecule to
occupy the vacated site at Gd, enhancing relaxivity propor-
tionately. The enhancement of relaxivity associated with
local hydration changes has been reported previously in the
context of pH-responsive systems[7] and metal-ion activated
systems.[4,6]
Ligand and complex synthesis: Ligand L1 was synthesized in
10 steps according to Scheme 3. Stepwise alkylation of 2-
(benzyloxy)phenol with tert-butyl 2-bromoethylcarbamate in
acetonitrile afforded compound 1, which was debenzylated
by hydrogenation, using Pd/C as the catalyst, to give 2 in
80% yield. This phenol was alkylated with methylbromoace-
tonitrile in MeCN to give compound 3, from which the
amine 4 was obtained by acid cleavage of the Boc groups.
N,N-Dialkylation of 4 with tert-butylbromoacetate in anhy-
drous DMF proceeded in 68% yield and the mono-acid 6
was obtained by selective deprotection of the methyl group
with LiOH. Alkylation of tris-tert-butyl-DO3A 7 with bro-
moacetonitrile gave the nitrile 8, from which the corre-
sponding amine 9 was obtained by hydrogenation over
Raney nickel (7m NH3/MeOH, 208C).[12,13] The penta-ester
10 was synthesized by coupling the amine 9 and acid 6 by
The metal-ion binding site in the responsive agent needs
to be designed with the coordination number, donor prefer-
ence, and desired overall affinity borne in mind. In this
work, we utilised six and eight-coordinate pyro-EGTA-
based ion-binding moieties for Zn2+ and Ca2+, which are
variants of structurally related analogues,[8] respectively.
Thus, we have set out to examine the behaviour of
using
N’-(3-dimethylaminopropyl)-N-ethylcarbodiimide
(EDC), 1-hydroxybenzotriazole (HOBt) and N-methylmor-
pholine (NMM) in DMF, from which L1 was obtained by hy-
drolysis with neat trifluoroacetic acid (TFA; 51%).
gadoliniumACHTUNGTRENNUNG
(III) complexes of L1 for Zn2+ and L2 for Ca2+
(Scheme 2). In addition, we have examined the emission
spectral properties of the EuIII analogues. Complexes of
The related ligand L2 was synthesized in 9 steps as shown
in Scheme 4. Alkylation of the phenol 2 with benzyl-2-bro-
moethylcarbamate gave the carbamate 11, which was further
treated with mild acid, cleaving the Boc group to afford the
amine 12. Reductive amination (benzene, 4 ꢁ molecular
sieves) of the amine 12 with benzaldehyde yielded the sec-
ondary amine 13, from which the methyl ester 14 was ob-
tained by alkylation with methylbromoacetate. The benzyl-
carbamate was hydrogenated over Pd/C to yield the diamine
15. N,N,N’-Trialkylation with tert-butylbromoacetate (DMF,
K2CO3) proceeded in 74% yield. Following basic hydrolysis
of 16, the resultant acid 17 was coupled with the macrocyclic
primary amine 9 to give the hexa-ester 18. Finally L2 was
obtained by acid hydrolysis using TFA. Ligands, L1 and L2
were purified by reverse-phase (RP)-HPLC and loaded with
Ln3+ (Gd3+ or Eu3+) using LnCl3·6H2O in water at pH 6.5.
The final concentration of Ln3+ was determined by induc-
tively coupled plasma optical emission spectrophotometry
(ICP-OES).
europiumACHTUNGTRENNUNG(III) exhibit an emission spectral profile that is
sensitive to the europium coordination environment.[7,9,10] In
this context, the spectral behaviour of structurally related
DO3A-monoamide europiumACTHNURGTNE(UNG III) complexes, with similar
CH2CH2NHCOR substituents, provides a useful reference
point for comparative emission spectral analyses. Examples
Modulation of proton longitudinal relaxivity with calcium
and zinc: The proton longitudinal relaxivity (r1p) of [Gd·L1]
and [Gd·L2] were measured to be 3.80 and 3.29 mMꢀ1 sꢀ1 in
buffered aqueous solution (310 K, 1.4 T, pH 7.4, 0.1m 3-mor-
pholinopropanesulfonic acid (MOPS)). These values slightly
change (up to 5%) over the pH range 3.5 to 8, and mea-
Scheme 2. Structures of ligands.
1530
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 1529 – 1537

MRI Contrast Agents for Calcium and Zinc
FULL PAPER
(73%
enhancement)
and
[Gd·L2] selectively responds to
calcium (67% enhancement).
Neither exhibit
a significant
change (<25%) in the presence
of added Na+, K+, Mg2+, Cu2+
and Zn2+or Ca2+. The relative-
ly low change in relaxivity ob-
served in the presence of
copper (II) may be tentatively
linked to its preference to bind
to the more remote amino-diac-
etate moiety in [Gd·L1] and
[Gd·L2]. Since copper(II) pre-
fers 4-coordinate square-planar
geometry, it is unable to bind
cooperatively to the amide car-
bonyl group that Zn2+or Ca2+
will ligate. The relaxivity values
measured for [Gd·L1] and
[Gd·L2] (310 K, 1.4 T) in the ab-
sence of added metal ions were
somewhat higher than might be
expected for a q=0 complex,
unless
a
significant second
Scheme 3. Reagents and conditions: i) tert-butyl 2-bromoethylcarbamate, K2CO3, anhydrous MeCN, 72%;
ii) Pd-C (10%), H2, MeOH, 40 psi, 80%; iii) methylbromoacetate, K2CO3, anhydrous MeCN, 88%; iv) 10%
TFA in CH2Cl2, 92%; v) tert-butylbromoacetate, K2CO3, anhydrous DMF, 68%; vi) LiOH, THF/MeOH/H2O
(3:2:2); vii) bromoacetonitrile, K2CO3, anhydrous MeCN, 78%; viii) 7m NH3·MeOH, Raney nickel, H2,
MeOH, 72%; ix) NMM, EDC, HOBt, anhydrous DMF, 28%; x) neat TFA, 51%.
sphere contribution to the over-
all relaxivity is present. The im-
portance of second-sphere con-
tributions to relaxivity has been
emphasised[14] previously. On
the other hand, the limiting re-
sured relaxation rates varied linearly with complex concen-
tration over the range 0.5 to 5 mm. Taken together, this sug-
gests that neither inter- nor intramolecular ligation by a
remote ligand carboxylate was occurring. These relaxivity
values were also measured in solutions containing various
added cations (K+, Na+, Mg2+, Ca2+, Zn2+ and Cu2+) at
concentrations similar to, or greater than, those encountered
in extracellular biological fluids (Figure 1). To a first approx-
imation, it is evident that [Gd·L1] is sensitive to added zinc
laxivity values measured in the presence of added zinc and
calcium (6.3 and 5.8 mmꢀ1 sꢀ1, respectively) are reasonable
values for mono-aqua Gd complexes with proximate hydro-
philic moieties that are likely to create a strong second
sphere of hydration. The origin of the observed modulation
was considered first to involve a hydration change at the Gd
centre. Accordingly, the hydration state (q) of the analogous
pair of Eu complexes was measured. This process involved
determining the radiative lifetime of the Eu 5D0 excited
state in H2O and D2O, in the presence and absence of added
Zn2+ or Ca2+, Table 1.
Table 1. Rate constants (k) defining the decay of the Eu 5D0 excited
state (298 K, H2O or D2O, 0.1m MOPS, pH 7.4, 3 mm complex).
Complex
[Eu·L2]
[Eu·L2]+Ca2+
[Eu·L2]+Ca2+ (2 mm)
[Eu·L2]+Ca2+ (3 mm)
[Eu·L1]
[Eu·L1]+Zn2+ (0.5 mm)
[Eu·L1]+Zn2+ (1 mm)
[Eu·L1]+Zn2+ (3 mm)
kH O [msꢀ1
]
kD O [msꢀ1
]
q[a,b,c]
2
2
U
1.49
1.57
1.68
1.83
1.46
1.53
1.72
2.06
0.87
0.77
0.70
0.69
0.80
0.77
0.87
0.93
0.35
0.57
0.88
0.97
0.40
0.52
0.63
1.00
E
ACHTUNGTRNE(NUNG 0.5 mm)
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
[a] q values possess an error of 20%; [b] no changes in q (>10%) were
measured in the presence of added MgCl2 and CaCl2 or ZnCl2 (1 equiv);
[c] q values in the absence of added metal ions were the same at pH 7.4
and 3.5 and the Eu spectral form was the same in each case also.
Figure 1. Relaxivity values for aqueous solutions containing [Gd·L1]
(dark grey) and [Gd·L2] (light grey) in the presence of the stated salts
(310 K, 1.4 T, pH 7.4, 0.1m MOPS, 0.5 mm complex).
Chem. Eur. J. 2011, 17, 1529 – 1537
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1531

A. Mishra, D. Parker and N. K. Logothetis
fitted to a 1:1 binding isotherm,
notwithstanding the evidence
for “ML2” formation (i.e., two
lanthanide complexes per Ca2+
ion) at low added Ca2+ values.
This gave an apparent associa-
tion
constant,
logK=3.59
(ꢂ0.03) (310 K, pH 7.4, 0.1m
MOPS). Subsequent addition of
ethylenediaminetetraacetic acid
(EDTA) to this solution re-
versed the observed increases
and, in the limit, restored the
initial
relaxivity
value
(Figure 2). Such behaviour is
consistent with preferential
binding of Ca2+ to EDTA, and
accords with the reversibility of
Ca2+ binding to [Gd·L2].
Changes to the lanthanideACHTUNGTRENNUNG(III)
ionic environment with added
calcium could also be moni-
tored by observing the changes
to the form of the [Eu·L2] emis-
sion spectrum, (Figure 3). In
the presence of added CaCl2,
the ratio of the DJ=2/DJ=1
band intensity (centred at 615/
592
3:1 to 2:1 and was reversed
upon addition of EDTA.
strikingly similar spectral
G
Scheme 4. Reagents and conditions: i) benzyl-2-bromoethylcarbamate, K2CO3, anhydrous MeCN, 66%;
ii) 10% TFA in CH2Cl2, 72%; iii) benzaldehyde, benzene; MeOH, NaBH4, 58%; iv) methylbromoacetate,
K2CO3, anhydrous MeCN, 82%; v) Pd/C (10%), H2, MeOH, 50 psi, 80%; vi) tert-butylbromoacetate, K2CO3,
anhydrous DMF, 74%; vii) LiOH, THF/MeOH/H2O (3:2:2); viii) compound 9, NMM, EDC, HOBt, anhydrous
DMF, 33%; ix) neat TFA, 55%.
change had been noted earlier
with structurally analogous eu-
ropium complexes and was at-
tributed to the change in hydra-
The empirical relationship that accounts for vibrational
deactivation of the metal excited state by directly coordinat-
ed, closely diffusing water molecules and proximate NH os-
cillators was used.[15] Hence, q values were estimated. This
data confirmed that each complex underwent a change from
a low q value (qꢁ0.4) to q=1, in the presence of more than
one equivalent of Ca2+ [Gd·L2], or Zn2+ [Gd·L1]. This hy-
dration switch accounts for the observed relaxivity increase
that is expected for Gd complexes of this molecular volume.
The overall relaxivity will also contain a significant second-
sphere contribution that is likely to vary from the free to
the metal-bound system. It is also possible that the free Gd
complexes exist in solution as a mixture of q=0 and q=1
complexes in differing ratio and hence give rise to the non-
integral hydration state (Eu) and the slightly elevated relax-
ivity values.
tion,[11] from zero to one bound water molecule.
Human serum albumin (HSA) is the major endogenous
protein in most extracellular fluids and its effect on the mea-
sured relaxivity of [Gd·L2] was assessed in the absence and
presence of 1 equivalent of CaCl2. In the absence of CaCl2,
addition of HSA led to a gentle increase in r1p of up to 33%
(4.36–5.80 mmꢀ1 sꢀ1, at 0.6 mm added protein), compared
with a rise of 14% (6.92–7.94 mmꢀ1 sꢀ1) in the presence of an
equivalent of CaCl2. No significant spectral changes were
observed in parallel series of experiments with [Eu·L2], con-
sistent with no variation in the Eu coordination environment
associated with the presence of added protein.
Anions in extracellular fluids can bind reversibly to lan-
thanide ion centres that are coordinatively unsaturated.[16]
The most abundant anion is bicarbonate and it is commonly
found at concentrations of the order of 25 mm, compared
with 2.3 mm for lactate and 1 mm for various phospho-
anions.[16,17] It has been shown that carbonate can chelate to
a LnIII centre, displacing any bound water in q=1 or q=2
complexes with heptadentate macrocyclic ligands.[17,18] This
is the case for the Ca2+ ternary adduct of [Gd·L2],
Calcium activation of [Gd·L2] and modulation of [Eu·L2]:
Incremental addition of CaCl2 to [Gd·L2] resulted in increas-
es in the proton relaxivity, up to a 1:1 stoichiometry,
(Figure 2). The observed variation in r1p with [Ca2+] was
1532
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 1529 – 1537

MRI Contrast Agents for Calcium and Zinc
FULL PAPER
Figure 2. Variation of relaxivity with added Ca2+ and subsequent EDTA addition for an aqueous solution containing [Gd·L2] (pH 7.4, 0.1m MOPS,
310 K, 1.4 T, 1 mm complex).
Figure 4. Variation of relaxivity in a solution containing [Gd·L2] in the
presence of 0.5 mm human serum albumin (square), 25 mm NaHCO3
Figure 3. Changes in the europium emission spectrum of [Eu·L2] with
added CaCl2, also showing (inset) the concomitant changes in the hydra-
tion state, q (3 mm complex, 0.1m MOPS, pH 7.4, lexc 397 nm).
(circle) and 0.5 mm HSA plus 25 mm NaHCO3 (triangle) (pH 7.4, 0.1m
MOPS, 310 K, 1.4 T, 0.5 mm complex).
(Figure 4). Incremental addition of NaHCO3 to [Gd·L2] was
undertaken at pH 7.4 in buffered solution in the absence
and presence of one equivalent of CaCl2. In the absence of
added Ca2+, no significant variation in proton relaxivity was
observed for aqueous solutions containing 0.5 mm [Gd·L2]
adding up to 50 mm bicarbonate. In contrast, a 25% (6.72–
5.34 mmꢀ1 sꢀ1) reduction in r1p of Ca2+ bound [Gd·L2] was
measured over the same range of added bicarbonate. Paral-
lel series of experiments were undertaken with [Eu·L2]
(3 mm complex, lexc =397 nm) observing the emission spec-
tral profile. No change in spectral form was observed in the
absence of added Ca2+, but in its presence the DJ=2/DJ=1
intensity ratio increased to 4:1, accompanied by a 30% re-
duction in the intensity of the DJ=0 transition at 579 nm.
These spectral changes are typical of the presence in solu-
tion of a carbonate-chelated species, and have been ob-
served on numerous occasions for carbonate adducts with
EuIII complexes of N-alkyl DO3A complexes.[18]
The variation of the proton relaxivity with added CaCl2
was monitored for three sets of aqueous solutions containing
0.5 mm HSA, 25 mm NaHCO3 and 0.5 mm HSA plus 25 mm
NaHCO3 (Figure 4). In the first two cases, added Ca2+
caused
a
47% (5.10–7.51 mmꢀ1 sꢀ1) and 43% (3.86–
5.50 mmꢀ1 sꢀ1) change in r1p, respectively, over the range of
Chem. Eur. J. 2011, 17, 1529 – 1537
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1533

A. Mishra, D. Parker and N. K. Logothetis
Figure 5. Proton relaxivity changes observed in mouse serum in the presence of [Gd·L2] as a function of added CaCl2 and subsequent (right) addition of
EDTA (pH 7.4, 0.1m MOPS, 310 K, 1.4 T, 0.5 mm complex).
zero to one added equivalent. In the latter case, this dimin-
ished only slightly to 40% (4.38–6.16 mmꢀ1 sꢀ1). Further-
more, the apparent association constant was estimated to be
therms indicated a logK value of 3.90
buffer (pH 7.4, 310 K, 0.1m MOPS) and of 3.50
mouse serum.
G
(ꢂ0.02) in
logK=3.35
G
Incremental addition of ZnCl2 to [Eu·L1] gave rise to
changes in the spectral form and the europium emission life-
time, that were also reversed by adding EDTA (Figure 7).
The form of the three allowed transitions in the magnetic-
dipole-allowed DJ=1 manifold (centred around 590 nm)
changed upon addition of ZnCl2. More striking was the var-
iation in the intensity of the DJ=2/DJ=1 manifolds, that re-
duced from 3:1 to 2:1 following addition of one equivalent
of ZnCl2. Such behaviour closely resembles that observed
with [Gd·L2] in the presence of Ca2+, and is consistent with
a reversible change in the europium coordination environ-
ment (Figure 6) associated with the change in metal hydra-
tion state.
of 25 mm NaHCO3 and in the protein bicarbonate medium
(pH 7.4, 0.1m MOPS, 310 K).
Finally, relaxivity changes as a function of added Ca2+
were assessed in mouse serum. Studies had not been report-
ed earlier in this medium by workers examining calcium ac-
tivated contrast agents.[3–6,13] In this case, there is an initial
concentration of CaCl2 in the serum. This was measured to
be 0.64ACHTUNGTRENNUNG
electrode (ISE). This method quantifies only ionised Ca2+ in
mouse serum. A titrimetric analysis study was undertaken
measuring r1p in mouse serum containing [Gd·L2], as a func-
tion of added CaCl2 followed by incremental addition of
EDTA (Figure 5). The observed relaxivity changes were
fitted by iterative least-squares analysis to give an affinity
constant for Ca2+ binding, logK=3.40
consistent with the idea that it is competitive binding of bi-
carbonate to the Ca-bound adduct that most significantly
perturbs the equilibrium between [Gd·L2] and [Gd·L2-
G
Conclusion
The experiments reported herein, serve to establish the
modulation of relaxivity induced by the selective, reversible
binding of CaCl2 to [Gd·L2] and ZnCl2 to [Gd·L1]. In each
case, the variation can be attributed to a change in the
ACHTUNGTRENNUNG(H2O)Ca]. It may also be noted that addition of excess
EDTA to the mouse serum mops up both the initial and the
added Ca2+, so that r1p goes to a lower final value than the
initial one.
lanthanideACTHNUTRGNEUG(N III) hydration state, enhancing relaxivity in the
metal-ion-bound form. A judicious estimate can be made to
define the theoretical relaxivity changes that may occur
Zinc modulation of [Gd·L1] and [Eu·L1]: Parallel series of
experiments were undertaken with [Gd·L1], exploring the
variation of measured relaxivity in buffer solution and in
mouse serum (Figure 6). The percentage change in relaxivity
over the range zero to one added equivalent of ZnCl2 was
73% (3.73–6.40 mmꢀ1 sꢀ1) in the former case and 40% (4.00-
5.61 mmꢀ1 sꢀ1) in the latter. Addition of EDTA reversed the
relaxivity increases and the r1p value returned to the original
value in the limit. This behaviour is consistent with the re-
versibility of the binding of Zn2+ to the putative hexaden-
tate binding moiety in L1. Analysis of the zinc binding iso-
when [Ca2+ or local [Zn2+] vary. In the case of a Zn2+
]
0
burst in the cerebrospinal fluid of the brain, the transient
concentration may increase from near 0 to 250 mm locally.
Such a change would be signalled by a 12% increase in re-
laxivity that would diminish with time as the zinc concentra-
tion returned to equilibrium. Modulation of ionisable [Ca2+
in cerebrospinal fluid, based on microelectrode studies, is
hypothesised to involve a rapid reduction in [Ca2+]0 of 30%
under maximal stimulation, and up to 90% for a traumatic
event. The variation of r1p with time might allow tracking of
the restoration of the [Ca2+
]
to its equilibrium value by
0
1534
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 1529 – 1537

MRI Contrast Agents for Calcium and Zinc
FULL PAPER
Figure 6. Proton relaxivity changes in aqueous buffer (square) and mouse serum (circle) in the presence of [Gd·L1], as a function of added ZnCl2 with
subsequent (right) addition of EDTA (0.1m MOPS, pH 7.4, 310 K, 1.4 T, 0.5 mm complex).
scale. Finally, it is appropriate to note that these relaxivity
changes have been measured at 1.4 T and 310 K. At higher
fields, the modulation of r1p will differ, although this change
is not expected to be very significant at 3 or 7 T. Measure-
ments of relaxivity made at 7 T for [Gd·L2] gave values that
were only 10% lower than those measured at 1.4 T. Indeed,
the NMRD profiles for such low MW contrast agents that
show little protein association exhibit a fairly flat-field de-
pendence of relaxivity after this field range.[19]
Experimental Section
Figure 7. Europium emission spectral changes for [Eu·L1] in the presence
of added ZnCl2, also showing (inset) the accompanying variation in Eu
hydration state, q (3 mm complex, lexc 397 nm, pH 7.4, 0.1m MOPS).
General: The chemicals were purchased from commercial suppliers
(Acros, Aldrich, Fluka, Merck, Strem and VWR) and were used without
further purification unless otherwise stated. Solvents were dried using an
appropriate drying agent when required (CH3CN over CaH2, CH3OH
over MgACHTUNTGRNE(NUG OMe)2 and THF over Na/benzophenone). Unless otherwise
MRI of the brain. The two cited cases would be character-
ised by 4 and 22% modulations of r1p (310 K, 1.4 T). These
calculations are only valid if the timescale for MRI data ac-
quisition is faster than the time taken for the metal-ion con-
centration to return to equilibrium. Modern imaging tech-
niques allow the acquisition of a single 2 mm thick volume
element within less than 50 msec (typically you take more
than 20 slices per second), and it is believed that the relaxa-
tion time characterising re-establishment of ionic equilibri-
um is of the order of several seconds.[1,2] Furthermore, over
each period the local concentration of the contrast agent in
the volume element examined must be assumed not to
change, which is a reasonable assumption for this short time-
mentioned, all the reactions were carried out under a nitrogen atmos-
phere and the reaction flasks were pre-dried with a heat gun under
vacuum. All the chemicals, which were air or water sensitive, were stored
under an inert atmosphere. Ultra pure deionised water (18 MWcmꢀ1) was
used throughout. All glassware was washed with a mixed acid solution
and thoroughly rinsed with deionised, distilled water.
Chromatography: Flash column chromatography was performed by using
flash silica gel 60 (70–230 mesh) from Merck. Thin layer chromatography
(TLC) was performed on aluminium-backed silica gel plates with 0.2 mm
thick silica gel 60 F254 (E. Merck) using different mobile phases. The com-
pounds were visualized by UV irradiation (254 nm) or iodine staining.
HPLC was performed at room temperature on a Varian PrepStar Instru-
ment, equipped with PrepStar SD-1 pump heads. UV absorbance was
measured using a ProStar 335 photodiode array detector at 254 nm. The
detector was equipped with a dual-path length flow cell which enables
Chem. Eur. J. 2011, 17, 1529 – 1537
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1535

A. Mishra, D. Parker and N. K. Logothetis
measurement of absorption of analytical and semi-preparative samples
without changing the flow cell. Reversed-phase (RP) analytical HPLC
was performed in a stainless steel Chromsep (length 250 mm, internal di-
ameter 4.6 mm, outside diameter 3/8 inch and particle size 8 mm) C18
column and semi-preparative RP-HPLC was performed on a stainless
steel Chromsep (length 250 mm, internal diameter 21.2 mm and particle
size 5 mm) C18 column (Varian, Advanced Chromatographic Solutions).
The compounds were purified using the following method: the gradient
was used with the mobile phase starting from 90% solvent A (0.1% TFA
in water) and 10% of solvent B (0.1% TFA in MeCN) to 70% B in
10 min, 100% B in 14 min, 100% B isocratic till 16 min and decreased to
thanide ion, another correction is made to obtain q’ [Eq. (2)], in which n
refers to the number of coordinated amide NH oscillators.
Relaxivity measurements: Relaxivity measurements were carried out at
378C, 60 MHz (1.4 T) on a Bruker Minispec mq60 instrument and at
238C, 300 MHz on a vertical 7 T/60 cm MRI Biospec system (Bruker
Biospin). The mean value of three separate measurements was recorded
and averaged. The relaxivities of the compounds were calculated as the
slope of the function shown in Equation (3):
1=T1,obs ¼ 1=T1,d þ r1 ꢄ ½GdLnꢃ
ð3Þ
10%
B in 20 min. The flow rate used for analytical HPLC was
in which T1,obs is the measured T1, T1,d is the diamagnetic contribution of
the solvent (calculated to be 4000 ms) and [GdLn] is the concentration in
mM of the appropriate GdIII complex (n=1 and 2). The errors for all re-
1 mLminꢀ1 and for preparative HPLC was 15 mLminꢀ1. All the solvents
for HPLC were filtered through a nylon-66 Millipore filter (0.45 mm)
prior to use.
laxivity values were less than (0.6) mMꢀ1 sꢀ1
.
Spectroscopy: 1H and 13C NMR spectra were recorded on a Bruker
300 MHz spectrometer (1H; internal reference CDCl3 at 7.27 ppm or
D2O at 4.75 ppm); 75 MHz spectrometer (13C; internal reference CDCl3
at 77.0 ppm). All experiments were performed at 238C.
Inductively coupled plasma optical emission spectrometry (ICP-OES) for
[Gd and Eu] analyses was performed using a Jobin–Yvon Ultima 2 spec-
trometer.
The apparent binding constant of the selected cation was calculated using
Equation (4)[20]
:
Electrospray mass spectra (ESI-MS) were recorded on SL 1100 system
(Agilent, Germany) with ion-trap detection in positive and negative ion
mode. HRMS were measured on a Thermo Finnigan LQT.
ꢀ
ꢁ
2
ðRꢀR0
Þ
ðRꢀR0
Þ
ðRꢀR0
Þ
ðR1 ꢀR0
Þ
þ ½GdLꢃ ꢄ
Þ ꢀ½GdLꢃ ꢄ
Emission spectra were measured on an ISA Joblin-Yvon Spex Fluorolog-
3 luminescence spectrometer (using DataMax v2.20 software), whereas
lifetimes were measured on a Perkin–Elmer LS55 luminescence spec-
trometer (using FL Winlab software). All samples were contained in
quartz cuvettes with a path length of 1 cm and polished base. Measure-
ments were obtained relative to a reference of pure solvent contained in
a matched cell. Luminescent titrations were carried out by normalising
the emission spectra with the absorption spectra in each point, to account
for the decrease in the sample concentration caused by addition of
cation/anion stock solution, where appropriate. All measurements were
carried in 0.1m MOPS buffer at pH 7.4. To avoid the undesirable evolu-
tion of carbon dioxide in buffer, argon was bubbled through the solution
for 30 min before proceeding to titrations.
K
ðR1 ꢀR0
ðR1 ꢀR0Þ
ð4Þ
ð5Þ
½Xꢃ ¼
K ¼
ðRꢀR0
Þ
1ꢀ
ðR1 ꢀR0
Þ
½GdXꢃ
½Xfꢃ½Gdfꢃ
in which [X] is the total concentration of cation in the solution; [GdL]
the total concentration of the complex; K the binding constant; R the re-
laxation rate of a given concentration of X; R0 the initial relaxation rate;
R1 the final relaxation rate; [GdX]: the concentration of the cation-coor-
dinated complex; [Xf] the concentration of free cation in the mixture;
[Gdf] the concentration of the free complex.
Synthesis of compound 10: A solution of 2-(2-(2-(bis(2-tert-butoxy-2-ox-
oethyl)amino)ethoxy)phenoxy)acetic acid (0.6 g, 1.37 mmol), compound
9 (0.69 g, 1.23 mmol), NMM (0.38 mL, 2.73 mmol) and HOBt (0.2 g,
1.5 mmol) in anhydrous DMF (5 mL) was stirred at 0–58C for 15 min
and then N’-(3-dimethylaminopropyl)-N-ethyl-carbodiimide (EDC)
(0.29 g, 1.5 mmol) was added. The reaction mixture was stirred for 12 h
at room temperature. The completion of the reaction was verified by
ESI-MS then the solution was poured into water (40 mL) and extracted
with EtOAc (3ꢂ50 mL). The combined organic layers were dried over
anhydrous Na2SO4, filtered and the filtrate evaporated under reduced
pressure. The residue was purified by column chromatography (silica gel,
10% MeOH in CH2Cl2, Rf =0.15) to give 10 as a dark yellow gum
(0.34 g, 28%). 1H NMR (300 MHz, CDCl3): d=1.46 (s, 18H; CH3), 1.47
(s, 27H; CH3), 2.05–3.00 (brm, 16H; NCH2, CH2 of ring), 3.03–3.40
(brm, 8H; NCH2CO and CH2 of ring), 3.44–3.73 (brm, 6H, NCH2CO),
3.81 (s, 2H; CH2NH), 4.15–4.21 (m, 2H; CH2O), 4.69 (s, 2H;
OCH2CONH), 6.83–7.01 (m, 4H; ArH), 7.51 ppm (brs, 1H; NH);
13C NMR (75 MHz, CDCl3): d=27.8, 27.9, 28.1, 37.9, 52.0, 53.3, 53.5,
55.6, 55.7, 55.8, 56.9, 58.0, 58.2, 67.0, 68.3, 80.9, 81.1, 81.8, 114.1, 115.9,
121.0, 122.8, 147.7, 149.2, 169.3, 170.7, 172.7, 172.9 ppm; HRMS (ESI+):
m/z: calcd for C50H87N6O13: 979.6331 [M+H]+; found 979.6307.
Synthesis of L1: [4,7-Bis-butoxycarbonylmethyl-10-(2-(2-(2-(2-(bis(2-tert-
butoxy-2-oxoethyl)amino)ethoxy)phenoxy)acetamido)ethyl)-1,4,7,10-tet-
raaza-cyclododec-1-yl]-acetic acid tert-butyl ester (0.3 g, 0.3 mmol) was
dissolved in neat TFA (10 mL) and stirred overnight. The completion of
the reaction was verified by ESI-MS and the solvent was evaporated
under reduced pressure. The residue was then purified by preparative
HPLC (tR =4.8 min). After lyophilization, ligand L1 was obtained as a
light yellow sticky solid (0.011 g, 51%). 1H NMR (300 MHz, D2O): d=
2.95–3.29 (brm, 12H; NCH2 and CH2 of ring), 3.30–3.8 (brm, 12H;
NCH2CO and CH2 of ring), 3.81–4.10 (m, 4H; CH2N and CH2NH), 4.24–
4.33 (m, 2H; NCH2CO), 4.37 (s, 2H; NCH2CO), 4.40–4.51 (m, 2H;
CH2O), 4.80 (s, 2H; OCH2CONH), 6.94–7.11 ppm (m, 4H; ArH);
13C NMR (75 MHz, D2O): d=34.6, 49.0, 49.7, 52.1, 53.6, 54.3, 55.9, 56.4,
Lifetime measurements: Lifetimes of europium complexes were mea-
sured by direct excitation of EuIII ion using a short pulse of light
(397 nm) followed by monitoring the integrated intensity of light (for eu-
ropium 615 nm) emitted during a fixed gate time (tg) after a delay time
(td). At least 20 delay times were used covering 3 or more lifetimes. A
gate time of 0.1 ms was used, and the excitation and emission slits were
set to 10 and 2.5 nm band-pass respectively. The obtained exponential
decay curves were fitted to the equation below, using Origin 6.0 software
(Data Analysis & Technical Graphics):
I ¼ I0 þ A1eðꢀktÞ
ð1Þ
in which: I=intensity at time t after the flash, I0 =intensity after the
decay has finished, A1 =pre-exponential factor and k=rate constant for
decay of the excited state. The excited state lifetime (t) is the inverse of
the radiative rate constant (k).
To examine the influence of some biologically common cations/anions on
Eu complexes, luminescent titrations were carried out by using different
concentrations of Ca2+, Zn2+ and NaHCO3 in 0.1m MOPS buffer at
pH 7.4. All of these measurements were carried out by adding the select-
ed cations/anions as liquid concentrated stock solutions and the addition
at each point was approx. 0.05–0.5% in volume of the original solution
observed, to avoid significant increase in sample volume. HSA was added
as a solid. Each Eu emission spectrum was corrected for dilution.
Inner-sphere hydration number (q’) determination: Inner-sphere hydra-
!
tion numbers (q’) were determined after excitation at the 5L6 7F0 band
(397 nm) and emission at 615 nm. 3 mm solutions of the Eu-complexes
were prepared in 0.1m MOPS buffer at pH 7.4 in H2O and in D2O. Hy-
dration numbers (q’) were calculated according to Equation (2):
q0Eu ¼ 1:2½ðkH OꢀkD Oꢀ0:25Þꢀ0:075nꢃ
ð2Þ
2
2
In the case of coordinated amide NH oscillators present close to the lan-
1536
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 1529 – 1537

MRI Contrast Agents for Calcium and Zinc
FULL PAPER
56.6, 63.8, 65.4, 113.8, 114.4, 118.2, 122.7, 146.5, 146.9, 166.0, 168.5, 171.6,
173.0 ppm; MS (ESI+): m/z: calcd for C30H47N6O13: 699.3 [M+H]+; found
699.4.
Acknowledgements
We thank ESF COST Action D38, the Max Planck Society and the
Hertie Foundation for support and the EC for a Marie Curie Intra-Euro-
pean Fellowship (PIEF-GA-2009-237253).
Synthesis of compound 18: A solution of 2-((2-(2-(2-(bis(2-tert-butoxy-2-
oxoethyl)amino)ethoxy)phenoxy)ethyl)(2-tert-butoxy-2-oxoethyl)amino)-
acetic acid (0.29 g, 0.48 mmol), compound 9 (0.24 g, 0.43 mmol), NMM
(0.1 g, 0.96 mmol) and HOBt (0.07 g, 0.53 mmol) in anhydrous DMF
(5 mL) was stirred at 0–58C for 15 min and then EDCI (0.1 g, 0.53 mmol)
was added. The reaction mixture was stirred overnight at room tempera-
ture. After ensuring completion of the reaction, the reaction mixture was
poured into water (40 mL) and extracted with EtOAc (3ꢂ50 mL). The
combined organic layers were dried over anhydrous Na2SO4 and filtered.
The filtrate was evaporated under reduced pressure and the residue puri-
fied by column chromatography (silica gel, 10% MeOH in CH2Cl2, Rf =
[1] G. Breitwieser, Int. J. Biochem. Cell Biol. 2008, 40, 1467; S. Hrabto-
va, D. Masri, L. Tao, F. Xiao, C. Nicholson, J. Physiol. 2009, 587,
4029.
[2] S. Zanotti, C. Andrew, J. Neurochem. 1997, 69, 594.
[3] a) J. L. Major, T. J. Meade, Acc. Chem. Res. 2009, 42, 893; b) W-H.
Li, S. E. Fraser, T. J. Meade, J. Am. Chem. Soc. 1999, 121, 1413;
c) E. L. Que, C. J. Chang, Chem. Soc. Rev. 2010, 39, 51.
[4] a) K. Dhingra, M. E. Maier, M. Beyerlin, G. Angelovski, N. K. Log-
othetis, Chem. Commun. 2008, 3444; b) K. Dhingra, P. Fouskovꢃ, G.
Angelovski, M. E. Maier, N. K. Logothetis and E. Toth, J. Biol.
Inorg. Chem. 2007, 13, 35; c) C. S. Bonnet, E. Toth, Future Med.
Chem. 2010, 2, 367384.
[5] M. Andrews, A. J. Amoroso, C. P. Harding, S. J. A. Pope, Dalton
Trans. 2010, 39, 3407; E. L. Que, E. Giandio, S. L. Baker, S. Aime,
C. J. Chang, Dalton Trans. 2010, 39, 469; Y. You, E. Tomat, K.
Hwang, T. Atanasijevic, W. Nam, A. P. Jasanoff, S. J. Lippard, Chem.
Commun. 2010, 46, 4139.
1
0.15) to give 18 as a pale yellow gum (0.16 g, 33%). H NMR (300 MHz,
CDCl3): d=1.36 (s, 9H; CH3), 1.37 (s, 18H; CH3), 1.40 (s, 18H; CH3),
1.41 (s, 9H; CH3), 2.49–2.74 (m, 16H; NCH2 and CH2 of ring), 2.95–3.24
(m, 18H; CH2 of ring, NCH2CO and CH2NH), 3.40–3.60 (m, 4H;
NCH2CO), 3.83–4.14 (m, 4H; CH2O), 6.68–6.89 ppm (m, 4H; ArH);
13C NMR (75 MHz, CDCl3): d=27.4, 27.6, 27.7; 53.1, 54.2, 54.5, 55.3,
55.4, 56.1, 56.4, 56.7, 57.1, 58.7, 59.2, 61.8, 67.0, 67.6, 80.6, 82.0, 82.3, 82.4,
114.4, 120.9, 148.5, 148.7, 170.2, 171.2, 172.0, 172.2, 173 ppm; HRMS
(ESI+): m/z: calcd for C58H102N7O15
:
1136.7428 [M+H]+; found
1136.7420.
[6] J. L. Major, R. M. Boiteau, T. J. Meade, Inorg. Chem. 2008, 47,
10788; J. L. Major, G. Porigi, C. Luchinat, T. J. Meade, Proc. Natl.
Acad. Sci. USA 2007, 104, 13881.
[7] M. P. Lowe, D. Parker, Chem. Commun. 2000, 707; M. P. Lowe, D.
Parker, O. Reany, S. Aime, M. Botta, G. Castellano, E. Gianolio, J.
Am. Chem. Soc. 2001, 123, 7601.
[8] a) O. Reany, T. Gunnlaugsson, D. Parker, J. Chem. Soc. Perkin
Trans. 2 2000, 1819; b) G. Angelovski, P. Fouskova, I. Mamedov, S.
Canals, E. Toth, N. K. Logothetis, Biochem. Chem. 2008, 9, 1729;
c) G. C. R. Ellis-Davies, R. J. Barsotti, Cell Calcium 2006, 39, 75;
d) L. Tei, Z. Baranyai, M. Botta, L. Piscopo, S. Aime, G. B. Gioven-
zana, Org. Biomol. Chem. 2008, 6, 2361.
[9] R. S. Dickins, J. I. Bruce, D. Parker, D. J. Tozer, Dalton Trans. 2003,
1264; D. Parker, Chem. Soc. Rev. 2004, 33, 156.
[10] R. Pal, D. Parker, Org. Biomol. Chem. 2008, 6, 1020; R. Pal, D.
Parker, L. C. Costello, Org. Biomol. Chem. 2009, 7, 1525.
[11] A. Congreve, D. Parker, E. Gianolio, M. Botta, Dalton Trans. 2004,
1441.
Synthesis of L2: [4,7-Bis-butoxycarbonylmethyl-10-(2-(2-((2-(2-(2-(bis(2-
tert-butoxy-2-oxoethyl)amino)ethoxy)phenoxy)ethyl)(2-tert-butoxy-2-oxo-
ethyl)amino)acetamido)ethyl)-1,4,7,10-tetraaza-cyclododec-1-yl]-acetic acid
tert-butyl ester (0.90 g, 1.3 mmol) was dissolved in neat TFA (10 mL) and
stirred at room temperature for 18 h. The completion of the reaction was
confirmed by ESI-MS. The solvent was evaporated under reduced pres-
sure and the residue was purified by preparative HPLC (tR =4.2 min).
After lyophilization, ligand L2 was obtained as a light yellow sticky solid
(0.062 g, 55%).1H NMR (300 MHz, D2O): d=2.90–3.18 (m, 14H; NCH2
and CH2 of ring), 3.19–3.45 (m, 8H; NCH2CO), 3.46–3.75 (m, 5H;
CH2NH, CH2N, NCH2CO), 3.76–3.97 (m, 5H; CH2N, NCH2CO), 4.27
(brs, 4H; NCH2CO), 4.30 (s, 2H, NCH2CONH), 4.30–4.45 (brm, 4H;
CH2O), 6.91–7.04 (brm, 4H; ArH); 13C NMR (75 MHz, CDCl3): d=42.0,
51.9, 52.5, 54.9, 55.8, 56.5, 57.1, 58.2, 58.3, 58.7, 58.9, 59.3, 59.6, 65.5, 65.8,
121.1, 125.2, 149.4, 149.5, 168.6, 171.2, 171.3, 171.5; MS (ESI+): m/z: calcd
for C34H54N7O15: 800.3 [M+H]+; found 800.4.
General preparation of lanthanide complexes of L1 and L2: Lanthanide
complexes of L1 and L2 were prepared from the corresponding solutions
of the ligands (1 equiv) and solutions of GdCl3·6H2O/EuCl3·6H2O
(1.0 equiv). The reaction mixture was stirred at 608C for 20 h. The pH
was periodically checked and adjusted to 6.5 with solutions of NaOH
(1m) and HCl (1n) as needed. After completion, the reaction mixture
was cooled and passed through chelex-100 to trap free Ln3+ ions, and the
Ln-loaded complexes were eluted. The fractions were dialyzed (500 MW
cutoff; Spectra/Pro biotech cellulose ester dialysis membrane, Spectrum
Laboratories) and lyophilized to obtain off-white solids. The absence of
free Ln3+ was checked with xylenol orange indicator. These complexes
were characterized by ESI-MS in positive and negative mode and the ap-
propriate isotope pattern distribution for Gd3+ and Eu3+ were recorded.
[12] S. Aime, G. Cravotto, S. G. G. Crich, G. B. Ferrari, G. Palmisano, M.
Sisti, Tetrahedron Lett. 2002, 43, 783.
[13] A. Mishra, P. Fouskova, G. Angelovski, E. Balogh, A. K. Mishra,
N. K. Logothetis, E. Toth, Inorg. Chem. 2008, 47, 1370.
[14] a) M. Botta, Eur. J. Inorg. Chem. 2000, 399; b) S. Aime, M. Botta,
D. Parker and J. A. G. Williams, J. Chem. Soc. Dalton Trans. 1996,
17.
[15] A. Beeby, I. M. Clarkson, R. S. Dickins, S. Faulkner, D. Parker, L.
Royle, A. S. de Sousa, J. A. G. Williams, M. Woods, J. Chem. Soc.
Perkin Trans. 2 1999, 493.
[16] R. S. Dickins, J. A. K. Howard, D. Parker, A. S. Batsanov, M. Botta,
J. I. Bruce, C. S. Love, R. D. Peacock, H. Puschmann, J. Am. Chem.
Soc. 2002, 124, 12697.
[17] J. I. Bruce, R. S. Dickins, T. Gunnlaugsson, S. Lopinski, M. P. Lowe,
D. Parker, R. D. Peacock, J. J. B. Perry, S. Aime, M. Botta, J. Am.
Chem. Soc. 2000, 122, 9674.
ACHTUNGTRENNUNG
r
[18] Y. Brettoniꢄre, D. Parker, R. Slater, M. J. Cann, Chem. Commun.
2002, 1930; Y. Brettoniꢄre, M. J. Cann, D. Parker, R. Slater, Org.
Biomol. Chem. 2004, 2, 1624.
ACHTUNGTRENNUNG
2
2
[19] P. Caravan, J. J. Ellison, T. J. McMurry, R. B. Lauffer, Chem. Rev.
1999, 99, 2293.
[20] R. Pal, D. Parker, Chem. Commun. 2007, 474.
ACHTUNGTRENNUNG
=
ACHTUNGTRENNUNG
Received: June 2, 2010
Published online: January 5, 2011
2
2
Chem. Eur. J. 2011, 17, 1529 – 1537
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1537