Comparative in vivo behavior studies of cyclen-based copper-64 complexes: Regioselective synthesis, X-ray structure, radiochemistry, log P, and biodistribution

Article abstract of DOI:10.1021/jm0496990

The in vivo behavior of copper(II)-cyclen complexes was modified via substitution of the parent ligand with two different substituents, 4-tert-butylbenzyl and acetate. This was achieved by using same synthetic strategy (regioselective protection/first alk

Full text of DOI:10.1021/jm0496990

J. Med. Chem. 2004, 47, 6625-6637
6625
Comparative in Vivo Behavior Studies of Cyclen-Based Copper-64 Complexes:
Regioselective Synthesis, X-ray Structure, Radiochemistry, log P, and
Biodistribution
Jeongsoo Yoo, David E. Reichert, and Michael J. Welch*
Mallinckrodt Institute of Radiology, Washington University School of Medicine, Campus Box 8225,
510 South Kingshighway Boulevard, St. Louis, Missouri 63110
Received April 21, 2004
The in vivo behavior of copper(II)-cyclen complexes was modified via substitution of the parent
ligand with two different substituents, 4-tert-butylbenzyl and acetate. This was achieved by
using same synthetic strategy (regioselective protection/first alkylation/deprotection/second
alkylation) to give nine cyclen derivatives. The X-ray structure of [Cu(2c)Cl]+ (2c ) 1-(4-tert-
butylbenzyl)-1,4,7,10-tetraazacyclododecane) showed that the chlorine ion from the reaction
mixture occupied the remaining apical position of a square pyramidal coordination environment
of these Cu-cyclen complexes. Eight out of nine compounds were labeled with 64Cu in high
radiochemical purity. log P measurements showed that the lipophilicities of the copper
complexes were increased dramatically by attaching hydrophobic substituents on the nitrogen
atoms of cyclen. Conversely, as the number of acetate groups increased, the lipophilicity was
decreased. The biodistribution of Cu-cyclen complexes was found to be influenced mostly by
the overall charge of the complexes rather than their lipophilicity. Positively charged (+2)
complexes showed high blood retention at early time points with sluggish clearance from liver
by 24 h. The attachment of even one acetate group onto cyclen accelerated blood and liver
clearance dramatically compared to +2 charged Cu(II) complexes. Neutral trans-substituted
Cu-4 showed the best clearance and lowest retention of doses from all organs most time,
followed by -1 charged complex Cu-2. Trans-substituted complexes structure isomers Cu-3
and Cu-4 showed better clearance and lower retention from all organs than their cis-
counterparts Cu-5 and Cu-6.
Introduction
metal complexes and bifunctional chelating agents. For
example, alkylation of cyclen with acetate groups is an
effective route to prepare complexes for attachment to
biomolecules such as peptides and antibodies or to
neutralize the charge of the coordinated metal. In many
applications it would be advantageous to have two
different types of substituents, such as charged and
lipophilic groups. Mixing substituent groups with dif-
ferent properties has potential for fine-tuning the
properties of the metal ion complex,11 enabling us to do
systematic studies of the relationship between structure
and physiological properties.11,12
One of the most important ligands in medicinal
inorganic chemistry is 1,4,7,10-tetraazacyclododecane
(cyclen). Two derivatives of this macrocycle, DOTA and
DO3A, both of which are functionalized with acetate
groups, are used to form stable metal complexes with
gadolinium, copper, indium, gallium, yttrium, and other
metals.1-6 These complexes are used heavily in the
fields of medical imaging and radiotherapy.7 In addition
to these uses, tetraazamacrocycles have recently found
applications as both antitumor drugs and anti-HIV
agents.8,9
To utilize these compounds in medicine, derivatization
of the parent macrocycle is required in order to impart
specific physical properties and to meet specific biologi-
cal requirements, such as localization to a specific organ
or receptor. Of particular importance is the regioselec-
tive introduction of one or more acetate groups on the
N atoms, as these will typically determine the formal
charge and core structure of the final metal complex,
affecting both the kinetic inertness and thermodynamic
stability under physiological conditions.10 Systematic
studies of the biodistribution of radiometal complexes
can aid in the design of new radiopharmaceuticals and
bifunctional chelates.
We are particularly interested in understanding the
in vivo behavior of 64Cu-labeled complexes. Copper-64
has favorable properties as a radionuclide for use in both
positron emission tomography (PET) imaging and tar-
geted radiotherapy due to its medium-lived half-life (t1/2
) 12.7 h) and two different decay modes (â+ 19%, â-
39%).1,7,13-17
In a previous communication,18 we presented a facile
synthetic strategy involving various protecting groups
leading to all possible cyclen substitution patterns
involving two different alkylating agents, including at
least one acetate group. We reported methods for the
regioselective protection of cyclen leading to four dif-
ferent substitution patterns mono, cis-di, trans-di, and
tri. The monoprotected cyclen compound, mono-N-Cbz-
cyclen was also prepared in a very effective way using
two protecting groups consecutively.
Regioselective substitution of the four nitrogen atoms
of cyclen is a crucial step for new cyclen-based radio-
* Corresponding author. Tel: 1 314 362 8436. Fax: 1 314 362 8399.
E-mail: welchm@wustl.edu.
10.1021/jm0496990 CCC: $27.50 © 2004 American Chemical Society
Published on Web 11/20/2004

6626 Journal of Medicinal Chemistry, 2004, Vol. 47, No. 26
Yoo et al.
Figure 1. Nine newly synthesized cyclen-based ligands.
Scheme 1a
Jones-Wilson et al. reported the in vivo behavior of
64Cu-labeled azamacrocyclic complexes.19 Biodistribu-
tion studies of Cu(II) complexes of several tetraaza
macrocycles such as cyclen, DO2A, DOTA, cyclam, et-
cyclam, and TETA provided a rough idea about the
relationship between structure and biological behavior.
It was demonstrated that the biodistribution of copper-
64 complexes correlates with differences in the size of
the macrocycle backbone and the formal charge of the
complex. The +2 charged complexes were shown to have
significantly higher uptake in liver and kidneys than
neutral and -2 charged complexes.
Herein, we report in detail synthetic procedures
leading to the regioselective N-alkylation of cyclen using
different protecting groups and all possible isomers with
two different substituents. We also report the X-ray
crystal structure of Cu(II) complex (2c). Eight out of
nine compounds were labeled with copper-64 in high
yield and were subject to log P measurements in order
to determine the relationship between structures of
metal complexes and their lipophilicity. Biodistribution
studies of the radiolabeled complexes revealed a strong
relationship between in vivo distribution and the overall
charge of the copper complexes.
a Reagents and conditions: (a) 4-tert-butylbenzyl bromide, (i-
Pr)2NEt, MeCN, 60 °C.
compounds is synthetically challenging and more com-
plicated.20 To achieve selective N-functionalization,
various approaches, such as direct functionalization,
cyclization of N-alkylated precursors, and regioselective
N-protection, have been devised.20 Herein, we utilized
a strategy of regioselective protection/first alkylation/
deprotection/second alkylation to obtain a total of eight
new regioselectively N-substituted cyclen derivatives.18
In this strategy, the proper choice of protecting group
for regioselective blocking of four identical nitrogen
atoms of cyclen is the most critical step. The protecting
groups are required to be introduced regioselectively in
high yield, stable to a wide range of reaction conditions,
and easily cleaved under mild reaction conditions
without attacking other functional groups.21
In this work varying numbers of nitrogen atoms of
cyclen were protected selectively and the remaining
available nitrogen atoms were alkylated with hydro-
phobic 4-tert-butylbenzyl groups and then the protecting
groups were cleaved to give mono-, trans-di- and cis-
di-, and trialkylated cyclen compounds (2c, 3, 5, 7). A
second alkylating agent, tert-butyl bromoacetate, was
introduced on nitrogen atoms that were previously
blocked by protecting groups. Cleavage of the tert-butyl
groups gave four hetero-substituted cyclen derivatives
having at least one acetic acid group (2, 4, 6, 8) (Schemes
2-5).
A triprotected cyclen compound, 1,4,7-triformylcyclen
(2a), was synthesized by modification of the reported
procedure.22 A mixture of cyclen and 6 equiv of chloral
hydrate in ethanol was stirred for 4 h at 60 °C. The
solution was concentrated in vacuo, and the crude
product was isolated via chromatography. After intro-
duction of a 4-tert-butylbenzyl group using the general
Results and Discussion
Synthesis. To compare the in vivo behavior of copper
complexes with different overall charges and lipophi-
licities, we synthesized nine new N-substituted cyclen
compounds with various numbers and substitution
patterns of 4-tert-butylbenzyl, which is very hydropho-
bic, and acetate (CH2CO2H). The nine newly synthesized
cyclen-based ligands are shown in Figure 1.
1,4,7,10-Tetra(4-tert-butylbenzyl)-1,4,7,10-tetraaza-
cyclododecane (1) was prepared by reacting cyclen with
4 equiv of 4-tert-butylbenzyl bromide in the presence of
N,N-diisopropylethylamine in good yield. (Scheme 1)
This general alkylation method was utilized to substi-
tute the nitrogen atoms of cyclen to prepare all other
N-substituted cyclen compounds.
While the synthesis of tetra-N-homo-substituted de-
rivatives such as 1 is straightforward, the preparation
of partially substituted or hetero-substituted cyclen

Cyclen-Based Copper-64 Complexes
Journal of Medicinal Chemistry, 2004, Vol. 47, No. 26 6627
Scheme 2a
tive positions of tertiary nitrogen atoms) cyclen deriva-
tives, respectively (3 and 5). The current synthesis of 3
is much higher yielding than the previously reported
method,26 which employed ethyl carbamate instead of
Cbz as the protecting group. Ethyl carbamate groups
were deprotected under very harsh condition (KOH,
hydrazine, reflux in ethylene glycol) to give product 3
in a yield of 10%, which is much lower than the current
96% yield. Two structural isomers can be clearly as-
signed by 13C NMR spectroscopy. D2h symmetry of trans-
disubstitued cyclen (3) gave only two signals (46.0, 51.9
ppm) for eight cyclen ring carbons in the 13C NMR
spectrum, while cis-isomer 5 showed four peaks (46.6,
46.9, 50.5, 50.8 ppm) for ring CH2 atoms due to the C2v
symmetry of 5 in solution. The second alkylation and
subsequent hydrolysis of the tert-butyl groups used the
same methods as previously described and gave two
DO2A derivatives (4 and 6) as the hydrochloric acid
salts, which were further recrystallized from diethyl
ether.
The ligands 7 and 8 were synthesized as shown in
Scheme 5. A monoprotected cyclen, mono-N-Cbz-cyclen
(7b) was prepared in high yield by employing two
consecutive protections of all four nitrogen atoms in
cyclen with formyl and Cbz groups followed by selective
removal of the three formyl groups and purification by
simple recrystallization.18 In brief, it is important to
keep the pH basic (up to 10) and to use excess benzyl
chloroformate (4.5 equiv) during the second protection
step. Selective deprotection of the formyl groups using
mild acid (1 M HCl, 50 °C) afforded the monoprotected
cyclen compound 7b as the 3HCl salt. Our method for
7b has several advantages as compared to the Kimura
group’s procedure.12 In their paper, less than 3 equiv of
di-tert-butyl dicarbonate compared to cyclen was added
slowly to the reaction mixture and the product was
purified from less protected and overprotected side
products to give 3Boc-cyclen.20,27 After a second protec-
tion of cyclen with Cbz, the product was purified again
by column chromatography. Their overall yield from
cyclen is 50%, which is much lower than the 80% overall
yield achieved with our methodology.12 Typical alkyl-
ation on the three available nitrogen atoms followed by
deprotection of the Cbz group by hydrogenation over
Pd/C gave trialkylated cyclen 7. The tert-butyl groups
of 8a were cleaved by TFA in quantitative yield to afford
8.
a Reagents and conditions: (a) Cl3CCH(OH)2, EtOH, 60 °C; (b)
4-tert-butylbenzyl bromide, (i-Pr)2NEt, MeCN, 60 °C; (c) 2 M HCl,
60 °C; (d) tert-butyl bromoacetate, (i-Pr)2NEt, MeCN, 60 °C; (e) 6
M HCl, reflux.
alkylation procedure, the formyl groups were easily
removed using 2 M HCl solution to give monoalkylated
cyclen compound (2c‚4HCl). Further alkylation with
tert-butyl bromoacetate on the deprotected amine ni-
trogens was carried out. A large excess of N,N-diiso-
proylethylamine (10 equiv) was used to scavenge HCl
salt of reactant 2c and byproduct HBr. After column
purification of 2d, the tert-butyl group was easily
hydrolyzed by 6 M HCl to give the DO3A analogue 2 in
hydrochloric acid form (Scheme 2). Alternatively, the
tert-butyl groups can also be easily cleaved by treatment
with TFA to give neutral product.23
Schemes 3 and 4 illustrate the syntheses of regio-
selective disubstituted cyclen derivatives. Two possible
regioselective protections of two amine nitrogens in
cyclen were achieved by treating cyclen with benzyl
chloroformate or diethyl oxalate to give 1,7-dibenzyl-
oxycarbonylcyclen24 and cyclenoxamide,25 respectively.
After the initial alkylation, the Cbz and oxalate groups
were deprotected by hydrogenation over Pd/C and
strong base treatment (10 M NaOH) to give trans- and
cis-disubstituted intermediate (with respect to the rela-
Scheme 3a
a Reagents and conditions: (a) benzyl chloroformate, pH 2-3, H2O; (b) 4-tert-butylbenzyl bromide, (i-Pr)2NEt, MeCN, 60 °C; (c) H2,
10% Pd/C, EtOH; (d) tert-butyl bromoacetate, (i-Pr)2NEt, MeCN, 60 °C; (e) 6 M HCl, reflux.

6628 Journal of Medicinal Chemistry, 2004, Vol. 47, No. 26
Yoo et al.
Scheme 4a
a Reagents and conditions: (a) diethyl oxalate, EtOH; (b) 4-tert-butylbenzyl bromide, (i-Pr)2NEt, MeCN, 60 °C; (c) NaOH, EtOH, reflux;
(d) tert-butyl bromoacetate, (i-Pr)2NEt, MeCN, 60 °C; (e) 6 M HCl, reflux.
Scheme 5a
a Reagents and conditions: (a) (1) Cl3CCH(OH)2, EtOH, 60 °C, (2) benzyl chloroformate, pH 4-10, H2O; (b) 1 M HCl, 50 °C; (c) 4-tert-
butylbenzyl bromide, (i-Pr)2NEt, MeCN, 60 °C; (d) H2, 10% Pd/C, EtOH; (e) tert-butyl bromoacetate, (i-Pr)2NEt, MeCN, 60 °C; (f) TFA.
Table 1. Electronic Absorption Data of [Cu(2c)Cl]+(ClO4)-
λmax
(nm)
ꢀmax
(M-1 cm-1
)
solvent
H2O
MeNO2
MeCN
594
705
710
312
303
356
tion28,29,32 strongly implies that 64Cu-labeled 1, 3, 5, and
7 complexes also have square pyramidal coordination
environment. An X-ray structure of [Cu(3)Cl]+Cl-,33
prepared from reaction of CuCl2 and ligand 3, also
supports this preference of a square pyramidal coordi-
nation geometry with Cl- ion in an apical position. The
cyclen ring backbone is bent away from the plane of the
nitrogen atoms in a direction opposite to that of the
copper ion. The 4-tert-butylbenzyl group attached to the
one N atom of cyclen ring is directed away from the
cyclen ring with the benzyl group somewhat above the
copper ion and the C5-C6 bond in benzyl group roughly
parallel to the Cu1-N1 bond (Figure 2).
Figure 2. X-ray structure of [Cu(2c)Cl]+(ClO4)-. Hydrogen
atoms and perchlorate ion have been omitted for clarity.
Crystallographic atom numbering is used. Displacement el-
lipsoids are scaled to the 30% probability level.
X-ray Structure Determination and Visible Ab-
sorption Spectra. The reaction of 2c‚4HCl with cop-
per(II) perchlorate gave [Cu(2c)Cl]+(ClO4)-, which was
then subjected to X-ray diffraction analysis. The crystal
structure of [Cu(2c)Cl]+ is given in Figure 2. One
chloride ion was found to coordinate to copper and one
perchlorate anion counterbalanced the charge of the
cationic copper complex. The Cu(II) ion is coordinated
to four N atoms, and a Cl- ion occupies the apical
position in a square pyramidal coordination structure.
The bond distances and angles around the Cu ion
indicate geometry close to an ideal square pyramid.28-32
The preference of square pyramidal coordination geom-
etry of cyclen-based copper complex with axial liga-
Visible absorption spectra of [Cu(2c)Cl]+(ClO4)- were
measured in three different solvents, and the result is
shown in Table 1. When Cu(II) complexes of cyclen
derivatives having no other possible Cu(II) coordination
substituent on N atoms are dissolved in aqueous solu-
tion, the apical ligand is replaced with aqua ligand,19,34,35
and this replacement can be detected by a shift of the
absorption band.36,37 The visible absorption maxima λmax

Cyclen-Based Copper-64 Complexes
Journal of Medicinal Chemistry, 2004, Vol. 47, No. 26 6629
Table 2. Labeling Conditions and Radiochemical Purities for
Complexing 64Cu to ligands 1 through 8
labeled compounds were determined, to investigate the
effect lipophilicity has on the in vivo behavior. Average
log P values of three measurements of three consecutive
back-extractions are shown in Figure 3, along with the
overall formal charges. The overall formal charge of the
Cu(II) complexes were calculated on the basis of the
number of acetate groups attached to the cyclen ring.19
rxn
solvent (µL)
temp
(°C)
time
radiochem
ligand (ethanol:buffera)
(min) purity (%)
Rf
1
2
3
4
5
6
7
8
300:100
0:300
60
60
40
60
rt
60
30
30
60
60
95
90
0.69c
0.79c
0.15e
0.70c
0.35g
0.39c
0.45g
0.35g
100:100d
100:300
50:80f
60
97
Four complexes (1, 3, 5, 7) in the first row of Figure
3 are all assigned to be positively charged (+2). As
confirmed in the visible absorption study, in aqueous
solution cyclen-based Cu(II) complexes without acetate
groups have a square pyramidal coordination geometry
with a water molecule in the apical position. Therefore,
the overall charge of copper complexes is +2. Previously
this overall charge assignment was also measured
directly by electrophoresis in the case of [Cu(cyclen)-
30
420
60
30
15
93
98
100:270
100:160
50:100
95
98
100
a 0.1 M ammonium acetate, pH 6.4. b Silica gel, MeOH:ethyl
acetate ) 1:2. c Silica gel, H2O:MeOH ) 1:2. d 0.1 M ammonium
citrate, pH 6.5. e Silica gel, H2O:MeOH ) 1:100. f H2O. g RP-8,
H2O:MeCN ) 1:9.
(H2O)]2+ 19
In these four cases, the lipophilicity of the
.
for [Cu(2c)Cl]+(ClO4)- in nitromethane and acetonitrile
was observed at a higher wavelength, 705 and 710 nm,
respectively, than that in water, indicating the presence
of the axial chloride ion coordinated to copper ion. In
aqueous solution, the λmax was observed at 594 nm,
which is comparable with the reported values of
[Cu(cyclen)(H2O)]+ at 59934 and 600 nm.36 This confirms
the displacement of the Cl- ion by H2O molecule in
aqueous solution.37
copper complexes are expected to be determined solely
by the number of hydrophobic 4-tert-butylbenzyl groups
rather than the charge, because all four complexes have
the same positive overall charge. As the number of tert-
butylbenzyl substituents is increased from two (trans-
di) to three and to four, log P values decrease, which
indicates that the lipophilicity of the complexes was
decreased. The log P value of Cu-5, which has two
substituents in cis nitrogen position of cyclen, is slightly
lower than Cu-7, which has three substituents (2.62
vs 2.73). Complex Cu-1, which has four substituents,
is most hydrophilic, while Cu-3, which has two sub-
stituents, is most hydrophobic among the four +2
compounds. This unexpected result could be explained
by considering the charge distribution over the whole
structure of Cu(II) complexes.19,38 X-ray crystal struc-
tures of [Cu(2c)Cl]+ and [Cu(3)Cl]+ 33 and literature
structures32 showed that substituents on the ring ni-
trogens are directed away from the 12-membered ring
and are all oriented on the same side of the plane
containing the four nitrogens. The four ethylene groups
of the cyclen ring are all oriented opposite the coordi-
nated Cl ion, relative to the nitrogen plane. As the
number of substituents on the cyclen ring increase, the
dipole moment of this series of Cu(II) complexes seems
to increase, making the compound more hydrophilic.
Complexes Cu-3 and Cu-5 are structural isomers in
which only the substituent position is different. How-
ever, the log P value of trans-substituted isomer Cu-3
(3.19) is much higher than that of cis-substituted isomer
Cu-5 (2.62). Two pendant arms on one side of cyclen
might make dipole moments of cis-substituted isomer
bigger than that of the trans-isomer. The symmetry
within complexes seems to play a great role in deter-
mining their lipophilicity.
Radiochemistry. All eight compounds were success-
fully labeled in more than 90% radiochemical purity,
most in greater than 95% yield. The labeling conditions
were optimized to give the best labeling yield depending
on each ligand. The labeling solvents, temperatures,
incubation times, radiochemical purities, Rf values, and
TLC developing solvents are summarized in Table 2.
Ethanol/buffer mixtures were used for radiolabeling,
except for ligand 2, due to the low solubility of cyclen
derivatives and their copper complexes in pure aqueous
solution. Four different TLC systems were used to check
radiochemical yield. The developing solvent was chosen
on the basis of the polarity of 64Cu-labeled complexes
and therefore their Rf values. Interestingly, 64Cu com-
plexes of ligands 5, 7, and 8 were found not to be stable
on a normal phase silica gel TLC plate during develop-
ment. Reverse phase C8 TLC plates were used instead
for 5, 7, and 8. Ligand 5 was mixed at room temperature
with 64CuCl2 in ethanol/water mixture solvent and
allowed to stand for 7 h at room temperature to give
the best labeling yield. The free authentic 64CuCl2 and
64Cu-acetate moved further, Rf ) 0.66 and 0.64, re-
spectively, under this reverse-phase C8 TLC condition.
The 64Cu-acetate remained on silica gel TLC at the
origin, Rf ) 0.0, under all other developing solvent
conditions.
Ligand 2c proved refractory to radiolabeling with
64Cu. Varying the labeling conditions by changing the
buffer, solvent, temperature, reaction time, and TLC
conditions all failed to give a satisfactory radiochemical
yield. The best radiochemical yield of 83% was achieved
by adding 1 equiv of cold CuCl2 relative to the ligand
and shaking the mixture for 30 min at 30 °C in 0.1 M
ammonium acetate solution (pH 6.4) [Rf ) 0.24; RP-8,
MeCN:H2O (9:1)]. No further study was carried out with
this labeled compound due to relatively low radiochemi-
cal yield.
In the case of the parent Cu-cyclen complex, octanol-
water partition coefficients could not be determined
because the 64Cu-labled cyclen could not be back-
extracted into the octanol layer.19
Overall charges of the other four complexes (2, 4, 6,
8) shown in the second row of Figure 3 were assigned
depending on the number of acetate groups present. Up
to two acetate groups per cyclen ring can coordinate to
the copper to give a five- or six-coordinated metal
complex.39,40 Charge assignment of Cu-cyclen com-
plexes based on the number of acetate groups was
experimentally confirmed using the electrophoresis
Lipophilicity of Complexes. The log P (octanol/
water partition coefficient) values of the eight radio-

6630 Journal of Medicinal Chemistry, 2004, Vol. 47, No. 26
Yoo et al.
Figure 3. log P values and overall charges of 64Cu-labeled complexes.
migration method.19 Complex Cu-8 has one acetate
group and therefore was assigned a +1 charge. The log
P value of Cu-8 is exactly the same as that of its
counterpart, Cu-7, which does not have an acetate
group. However, as the overall charge of the complexes
is changed from positive to neutral and negative, the
lipophilicity is decreased significantly. Compounds Cu-4
and Cu-6, which are assigned as neutral, have much
lower log P values than their counterparts Cu-3 and
Cu-5, which do not have acetate groups. The log P
values of Cu-4 and Cu-6 were decreased by 29% and
32%, respectively, from their counterparts, Cu-3 and
Cu-5. The same trend, in which the cis-isomer has a
lower log P value than the trans-isomer, was observed
with structural isomers Cu-4 and Cu-6. Trans-substi-
tuted isomer Cu-4 has a higher log P value than cis-
substituted isomer Cu-6. Negatively charged complex
Cu-2 has the lowest log P value, 1.27. As the overall
charge was changed from neutral to -1, the log P was
decreased by a similar amount as from +1 to neutral.
The influence of the overall charge of the copper complex
on the log P value is clearly seen in these complexes.
Hydrophilicity is increased and the log P value de-
creased as the number of acetate groups is increased
up to three, changing the charge from +1 to neutral and
-1. The addition of one acetate group on a cyclen
nitrogen atom was found to have a marginal influence
on the lipophilicity of copper(II) cyclen complexes.
Symmetry within the complex also plays an important
role in determining the lipophilicity of structural iso-
mers.
Biodistribution Studies. To elucidate the relation-
ship between structure and in vivo behavior, eight 64Cu-
labeled compounds were subjected to biodistribution
studies. Labeled compounds were injected into female
Sprague-Dawley rats, and blood, liver, kidney, muscle
and heart were harvested at 15 min and 4 and 24 h
postinjection. Biodistribution results are summarized
in Table 3. All eight compounds show different uptake
and clearance patterns; a comparison of blood, liver, and
kidney clearance patterns for 64Cu-labeled 1-8 is shown
in Figure 4.
From the blood clearance comparison, a dependence
on the overall charge is clearly seen. All four +2
complexes, Cu-1, Cu-3, Cu-5, and Cu-7, show much
higher blood uptake than other four compounds at early
time points (15 min). However, the activities cleared
dramatically by 4 h to render comparable amounts of
activity to those of the other four compounds. In the
cases of Cu-3, Cu-5, and Cu-7, blood activities at 24
Table 3. Biodistributiona of 64Cu-Labeled 1-8 in Mature
Female Sprague-Dawley Rats
organ
Cu-1
Cu-2
15 min
Cu-3
Cu-4
blood
liver
3.92 ( 0.611 0.37 ( 0.043 2.75 ( 0.290 0.16 ( 0.002
6.60 ( 1.061 2.28 ( 0.226 4.73 ( 0.834 1.26 ( 0.001
kidney 1.77 ( 0.120 2.64 ( 0.307 1.55 ( 0.155 3.66 ( 0.389
muscle 0.11 ( 0.035 0.12 ( 0.011 0.08 ( 0.009 0.06 ( 0.003
heart
0.64 ( 0.062 0.15 ( 0.021 0.55 ( 0.050 0.18 ( 0.008
4 h
blood
liver
0.96 ( 0.089 0.21 ( 0.032 0.10 ( 0.006 0.15 ( 0.012
4.93 ( 0.620 0.91 ( 0.037 6.96 ( 0.370 0.69 ( 0.007
kidney 5.31 ( 0.629 3.42 ( 0.390 2.98 ( 0.386 2.17 ( 0.644
muscle 0.14 ( 0.022 0.04 ( 0.004 0.03 ( 0.003 0.04 ( 0.009
heart
0.49 ( 0.058 0.13 ( 0.017 0.11 ( 0.026 0.15 ( 0.009
24 h
blood
liver
0.55 ( 0.048 0.19 ( 0.021 0.18 ( 0.006 0.12 ( 0.008
2.66 ( 0.487 0.80 ( 0.046 5.82 ( 0.425 0.43 ( 0.055
kidney 2.62 ( 0.681 1.63 ( 0.326 2.60 ( 0.142 0.79 ( 0.194
muscle 0.10 ( 0.008 0.03 ( 0.002 0.05 ( 0.008 0.03 ( 0.001
heart
0.44 ( 0.022 0.16 ( 0.005 0.16 ( 0.012 0.16 ( 0.013
Cu-5 Cu-6 Cu-7 Cu-8
15 min
organ
blood
liver
2.34 ( 0.345 0.54 ( 0.028 6.56 ( 0.621 0.49 ( 0.060
5.89 ( 1.246 4.43 ( 0.731 6.58 ( 0.931 5.19 ( 0.540
kidney 3.25 ( 0.382 4.55 ( 0.423 1.54 ( 0.118 1.82 ( 0.142
muscle 0.16 ( 0.025 0.15 ( 0.002 0.08 ( 0.015 0.08 ( 0.007
heart
1.70 ( 0.267 0.30 ( 0.007 1.15 ( 0.074 0.32 ( 0.035
4 h
blood
liver
0.22 ( 0.040 0.51 ( 0.051 0.64 ( 0.148 0.38 ( 0.024
7.16 ( 1.323 2.15 ( 0.113 5.65 ( 0.798 2.80 ( 0.483
kidney 5.92 ( 1.377 6.11 ( 1.126 9.59 ( 0.809 3.93 ( 0.449
muscle 0.09 ( 0.013 0.09 ( 0.006 0.11 ( 0.004 0.07 ( 0.003
heart
0.30 ( 0.037 0.34 ( 0.037 0.44 ( 0.054 0.27 ( 0.008
24 h
blood
liver
0.31 ( 0.051 0.42 ( 0.032 0.65 ( 0.114 0.33 ( 0.063
3.64 ( 0.371 1.80 ( 0.208 3.24 ( 0.538 1.83 ( 0.451
kidney 3.27 ( 1.026 2.91 ( 0.424 3.92 ( 1.339 2.09 ( 0.628
muscle 0.09 ( 0.007 0.09 ( 0.004 0.13 ( 0.006 0.07 ( 0.008
heart
0.36 ( 0.031 0.42 ( 0.039 0.57 ( 0.052 0.31 ( 0.039
a Data are expressed as the %ID/g ( SD with four animals per
data point.
h are slightly higher compared to those at 4 h. This may
be due to metabolic degradation of the 64Cu-labeled
complexes19,41 with subsequent incorporation of the 64Cu
into copper-containing proteins such as superoxide
dismutase or cerruloplasmin. Blood levels of the other
four compounds, which have at least one acetate group,
were cleared within the first 15 min, which implies that
the complexes do not dissociate in the blood.19,42,43 But
blood levels did not lower much further out to 24 h. Even
though the differences in blood levels of Cu-2, Cu-4,
Cu-6, and Cu-8 are very small, neutral complex Cu-4

Cyclen-Based Copper-64 Complexes
Journal of Medicinal Chemistry, 2004, Vol. 47, No. 26 6631
those four +2 charged complexes, disubstituted cyclen
complexes Cu-3 and Cu-5 continued to accumulate in
liver over time and showed the highest uptake by 4 h
postinjection. However, the most lipophilic complex
Cu-3 showed severe retardation of liver clearance at
24 h. The liver activity of Cu-3 decreased by 16%
between 4 and 24 h, while Cu-5 decreased by 49% over
the same time span. In contrast, Cu-1 and Cu-7
cleared out from liver continuously by 24 h with similar
clearance rates. Cu-1, which has the lowest log P value
among these four complexes, showed the best liver
clearance at 4 and 24 h time points. High liver ac-
cumulation at early time points followed by slow clear-
ance of +2 charged azamacrocyclic complexes were
consistent with other literature reports.19,41,44
Negatively charged complex Cu-2 and trans-substi-
tuted neutral complex Cu-4 showed very fast clearance
through liver by 15 min. Liver uptakes of both com-
pounds are less than 0.91% injected dose (ID)/g at 4 h.
Cu-4 had the fastest liver clearance at all time points.
Cis-substituted neutral complex Cu-6 and +1 charged
Cu-8 cleared rapidly from the blood with an initial high
liver uptake followed by rapid washout. From 46 to 51%
of the 15 min dose cleared from liver in 4 h, but further
washout was retarded to give 1.8% ID/g for both
compounds at 24 h postinjection. In comparison, trans-
substituted isomer Cu-4 showed much faster liver
clearance than cis-substituted isomer Cu-6 at the same
time points, as seen with the blood clearance.
Kidney uptake and clearance of cyclen-based copper
complexes showed a more complex pattern than found
in the blood and liver, but they did not show a clear
charge dependence. All compounds except Cu-4 slowly
accumulated in kidney over time and showed highest
uptake at 4 h and then slowly cleared out by 24 h. In
particular, +2 charged 64Cu-labeled 1, 3, and 7 showed
very low initial kidney uptake at 15 min (e4% ID/g)
with continuous accumulation of activity in kidney
achieving a maximum at 4 h. Cu-7 showed a 6-fold
increase in the kidney level by 4 h compared to 15 min
dose, consistent with the observed decrease in the blood
level. Cu-3 that has highest log P value showed only
13% further clearance of 4 h dose in 24 h.
Cis-substituted neutral complex Cu-6 and +1 charged
Cu-8 complexes also showed slow initial uptake with
continuous accumulation of dose up to 4 h followed by
a slow clearance out to 24 h. As in the blood and liver,
negatively charged Cu-2 and especially trans-substi-
tuted neutral complex Cu-4 showed faster kidney
clearance than the other six compounds. Despite a
preference for hepatic clearance, neutral compound
Cu-4 showed good kidney clearance without retarda-
tion over time to end up with 0.79% ID/g at 24 h
postinjection.
Figure 4. Comparison of blood, liver, and kidney clearance
for 64Cu-labeled complexes.
Muscle and heart uptake for all complexes were
minimal and no significant uptake in the myocardium
was found for any of the radiometal complexes. Only
Cu-4 and Cu-5 showed a heart/blood ratio >1 at 4 and
24 h, but their percent ID/g values were less than 0.3.
Other organs, such as spleen, lung, fat, bone, and brain,
were harvested in some cases to see whether any tissue
specificity existed. No significant uptake was found
outside of the clearance organs. The brain/blood ratios
of complexes Cu-2 and Cu-6 were less than 0.2 at all
and -1 charged Cu-2 showed the best blood clearance.
Trans-substituted isomers Cu-3 and Cu-4 showed
better blood clearance than their cis-substituted isomers
Cu-5 and Cu-6, respectively.
Comparison of liver clearance of 64Cu-labeled 1-8 also
clearly showed a charge dependence for the copper
complexes. All four +2 charged compounds Cu-1, Cu-
3, Cu-5, and Cu-7 showed high liver uptakes at early
time points and very slow clearance by 24 h. Among

6632 Journal of Medicinal Chemistry, 2004, Vol. 47, No. 26
Yoo et al.
Table 4. Biodistributiona of 64Cu-Labeled Cyclen, DO2A, and
DOTA in Mature Female Sprague-Dawley Rats19
in the cases of Cu-4 and Cu-6, which is much smaller
than 92% of Cu-DO2A.
15 min
2 h
24 h
Conclusion
Cyclen
blood
liver
0.58 ( 0.065
1.24 ( 0.166
7.69 ( 1.058
0.17 ( 0.038
0.85 ( 0.274
5.91 ( 1.748
0.14 ( 0.012
0.65 ( 0.020
1.62 ( 0.550
In summary, a total of nine cyclen derivatives were
synthesized successfully using the same synthetic strat-
egy. Regioselective alkylation to afford mono, cis- and
trans-di-, tri-, and tetra-substituted cyclen compounds
were achieved through thoughtful choice of protecting
group. A second alkylation with acetate groups gave all
possible isomers with two different substituents such
as DO1A, cis-, and trans-DO2A, and DO3A derivatives
in reasonable yield. X-ray structure of copper(II) com-
plex with mono-substituted cyclen 2c, [Cu(2c)Cl]+,
showed that chloride ion from the reaction mixture
occupied the remaining fifth position to give a square
pyramidal coordination environment with four nitrogen
atoms and chloride ion in the apical position.
kidney
DO2A
blood
liver
kidney
0.51 ( 0.062
0.24 ( 0.028
2.23 ( 0.182
0.04 ( 0.008
0.15 ( 0.021
0.67 ( 0.050
0.04 ( 0.008
0.11 ( 0.017
0.24 ( 0.076
DOTA
blood
liver
kidney
0.52 ( 0.110
0.24 ( 0.079
1.77 ( 0.272
0.06 ( 0.020
0.15 ( 0.013
0.78 ( 0.050
0.04 ( 0.015
0.15 ( 0.025
0.36 ( 0.054
a Data are expressed as the %ID/g ( SD with four or five
animals per data point.
time points, which indicates that these compounds did
not cross the blood-brain barrier.
Eight out of nine cyclen ligands were labeled with
64Cu in high radiochemical yield. To give best labeling
yield, labeling conditions were optimized. Partition
coefficients (log P) as measured by octanol/ammonium
acetate buffer were determined to range from 1.27 to
3.19. Lipophilicity of Cu(II) complexes was increased by
attaching hydrophobic substituents on cyclen N atoms.
However, as the number of substituents was increased,
lipophilicity was decreased instead. Charge distribution
and symmetry within the complex play important roles
in determining log P in a series of complex with the
same charge. Trans-alkylated isomers showed much
higher log P value than their cis-alkylated counterparts.
As the number of acetate groups was increased, the
overall charge was changed from positive to neutral and
negative, and lipophilicity was decreased. However, the
change from +2 charged to +1 charged complex by
adding one acetate group is insignificant.
The biodistribution of Cu-cyclen derivatives was
significantly altered by the number of lipophilic sub-
stituents and overall charge of complexes. All eight
hydrophobic compounds were excreted mostly through
the liver rather than kidneys. Plus-two-charged com-
plexes showed high blood retention at early time points
with sluggish clearance from liver by 24 h. The attach-
ment of even one acetate group on cyclen N atoms
accelerated blood and liver clearance dramatically com-
pared to +2 charged Cu(II) complexes. Kidney ac-
cumulation was continued up to 4 h for most complexes
except Cu-4 and slowly cleared by 24 h. Neutral trans-
substituted DO2A derivative showed best clearance and
lowest retention of doses from all organs most time,
followed by negatively charged complex. Trans-substi-
tuted complexes showed better clearance and lower
retention from all organs than their cis-counterparts.
The attachment of a substituent except acetate group
on cyclen rendered higher retention of activity in organs
at 24 h compared to those of Cu-cyclen and Cu-DO2A.
The effect on in vivo behavior by attaching hydropho-
bic pendants on N atoms of cyclen ring is clearly seen
by comparison with cyclen-based copper complexes that
have only acetate group as substituents. Jones-Wilson
et al. reported the in vivo behavior of 64Cu-labeled
azamacrocyclic complexes.19 Biodistribution data of
Cu(II) complexes of cyclen, DO2A, and DOTA are re-
expressed as percentage of injected dose per gram in
Table 4 for direct comparison with current data. All
eight compounds reported here have much higher log
P values than those of Cu(II) complexes of cyclen, DO2A,
and DOTA. The radiolabeled cyclen, DO2A, and DOTA
even could not be back-extracted into the octanol layer
from the aqueous layer.19
The copper complex of cyclen showed fast kidney
uptake at early time points with rapid clearance through
the kidneys. 79% of kidney dose at 15 min cleared in
24 h. The percentage of injected dose per gram in kidney
is 6 times as high as that of the liver at 15 min. In
contrast, Cu complexes of 1, 3, 5, and 7, which have at
least two lipophilic substituents showed much higher
liver uptake (%ID/g) compared to kidney at early time
points. The values of 4.7-6.6% ID/g for liver were 1.8-
4.3 times higher than those of kidney at 15 min.
Furthermore, kidney uptake was increased over time
up to 4 h, with more activity in kidney found at 24 h
rather than 15 min for copper complexes of 1, 3, 5, and
7. Blood clearance is also changed dramatically by
attaching lipophilic substituents on cyclen N atoms.
Cu-cyclen complex cleared out from blood as fast as
Cu-DO2A and Cu-DOTA. However, all four +2 charged
complexes, which have 4-tert-butylbenzyl groups as
substituents, showed high blood levels at 15 min (2.3-
6.6% ID/g) with fast clearance by 4 h.
DO2A and DOTA complexes with Cu(II) showed
faster clearance and lower retention from all organs at
all time points compared to Cu-cyclen. DO2A analogues
Cu-4 and Cu-6 cleared much faster from blood circu-
lation and liver at early time points and showed much
lower retention in liver at all times than copper com-
plexes of 1, 3, 5, and 7. Complexes Cu-4 and Cu-6
showed much higher retention in liver and kidney than
Cu-DO2A at all times points. Five and 18 times higher
activities of Cu-4 and Cu-6, respectively, were found
in liver compared to that of Cu-DO2A at 15 min. Less
than 25% of the dose of blood at 15 min cleared by 24 h
Experimental Section
General Methods and Materials. 1,4,7,10-Tetraaza-
cyclododecane (cyclen) was purchased from Strem Chemicals
(Newburyport, MA), and all other solvents and reagents were
obtained from Aldrich and used as received without any
further purification. Water was distilled and then deionized
(18 MΩ/cm2) by passing through a Milli-Q water filtration
system (Millipore Corp., Bedford, MA). 1H and 13C NMR

Cyclen-Based Copper-64 Complexes
Journal of Medicinal Chemistry, 2004, Vol. 47, No. 26 6633
59.9 (CH2Ph), 129.1, 133.3, 133.5, 155.6; HRMS (FAB) m/z
spectra were measured using a Varian Gemini 300 instrument,
and chemical shifts are reported in ppm on the δ scale relative
to TMS or solvent peak. Proton chemical shifts are annotated
as follows: ppm (multiplicity, coupling constant (Hz), integral).
Elemental microanalyses were performed by Galbraith Labo-
ratories, Knoxville, TN. Mass spectra were obtained from
Washington University Mass Spectrometry Resource. A Per-
kin-Elmer Lambda 25 UV/Vis spectrometer was used to record
absorption spectra.
Copper-64 was prepared on the Washington University
School of Medicine Cyclotron CS-15 cyclotron by the
64Ni(p,n)64Cu nuclear reaction at a specific activity range of
50-200 mCi/µg as previously described.17 EM Science silica
gel 60 F254 and Whatman C18 silica gel TLC plates were
purchased from Fisher Scientific (Pittsburgh, PA), and EM
Science C8 TLC plates RP-8 were purchased from Alltech
(Deerfield, IL). Radio-TLC was accomplished using a Bioscan
200 imaging scanner (Bioscan, Inc., Washington, DC). Radio-
activity was counted with a Beckman Gamma 8000 counter
containing a NaI crystal (Beckman Instruments, Inc., Irvine,
CA)
319.2868 ([M + H]+
, C19H35N4, calcd 319.2862). Anal. Calcd
for C19H34N4‚4HCl: C, 49.15; H, 8.25; N, 12.07. Found: C,
48.11; H, 8.43; N, 11.33.
1,4,7-Tris(tert-butoxycarbonylmethyl)-10-(4-tert-butyl-
benzyl)-1,4,7,10-tetraazacyclododecane (2d). To a solution
of 1-(4-tert-butylbenzyl)-1,4,7,10-tetraazacyclododecane‚4HCl
(499 mg, 1.07 mmol) in 30 mL of acetonitrile were added N,N-
diisopropylethylamine (1497 µL, 8.60 mmol) and tert-butyl
bromoacetate (634 µL, 4.27 mmol). The reaction mixture was
slowly heated to 60 °C and allowed to stir for 23 h. After
evaporation of solvents in vacuo, the residue was dissolved in
30 mL of ethyl acetate and washed by H2O and brine and then
dried by MgSO4. Crude product was washed through an
alumina column using methylene chloride and then eluted
using MeOH:CH2Cl2 (1:20) to give a pale oil (579 mg, 74%):
1H NMR (CDCl3) δ 1.11 (s, 9H), 1.23 (s, 18H), 1.25 (s, 9H),
2.63 (br s, 6H), 2.87-2.94 (br m, 8H), 3.23 (s, 2H), 3.40 (br s,
4H), 3.47 (br s, 2H), 4.47 (s, 2H), 7.23 (d, J ) 8.1 Hz, 2H), 7.30
(d, J ) 8.1 Hz, 2H); 13C NMR (CDCl3) δ 27.8 (OCCH3), 27.9
(OCCH3), 30.9 (PhCCH3), 34.4 (PhCCH3), 47.2, 50.4, 50.9, 52.1,
53.4 (NCH2Ph), 55.3 (CH2CO2), 55.5 (CH2CO2), 55.7 (CH2CO2),
81.3 (OCCH3), 81.5 (OCCH3), 81.9 (OCCH3), 125.7 (Ph ring
CH), 127.2 ((Ph ring C), 131.0 (Ph ring CH), 152.6 (Ph ring
C), 168.9 (CO2), 169.4 (CO2), 169.9 (CO2).
Ligand Synthesis. 1,4,7-Triformyl-1,4,7,10-tetraazacyclo-
dodecane22 (2a), 1,7-bis(benzyloxycarbonyl)-1,4,7,10-tetraaza-
cyclododecane24 (3a), and 1,4,7,10-tetraazabicyclo[8.2.2]tetra-
decane-11,12-dione)25 (cyclenoxamide, 5a) were prepared by
literature methods.
1,4,7-Tris(carboxymethyl)-10-(4-tert-butylbenzyl)-
1,4,7,10-tetraazacyclododecane (2‚4HCl). HCl solution (6
M, 25 mL) was added to 187 mg (0.28 mmol) of 1,4,7-tri(tert-
butoxycarbonylmethyl)-10-(4-tert-butylbenzyl)-1,4,7,10-tetraaza-
cyclododecane. The flask was slowly heated to 110 °C and
allowed to stir for 10 h. After evaporation of all solvents, the
crude product was recrystallized from ethanol/diethyl ether
to give solid 2 as the salt (4HCl) (145 mg, 80%): 1H NMR (D2O,
300 MHz) δ 1.28 (s, 9H), 3.05-3.16 (br m, 10H), 3.41 (br m,
6H), 3.49 (br s, 4H), 4.06 (br s, 2H), 4.46 (br s, 2H), 7.47 (d, J
) 8.1 Hz, 2H), 7.54 (br d, J ) 8.1 Hz, 2H); 13C NMR (D2O, 75
MHz) δ 33.3 (CCH3), 37.1 (CCH3), 51.0 (br), 51.2 (br), 52.6 (b)
54.4 (br), 56.1 (CH2CO2), 58.0 (CH2CO2), 60.4 (CH2Ph), 129.7
(Ph ring CH), 133.5 (Ph ring CH), 133.8 & 133.9 (Ph ring C),
157.1 (v br, Ph ring C), 176.1 (v br, CO2).
1,7-Bis(benzyloxycarbonyl)-4,10-bis(4-tert-butylbenzyl)-
1,4,7,10-tetraazacyclododecane (3b). To a mixture of
235 mg of 1,7-bis(benzyloxycarbonyl)-1,4,7,10-tetraazacyclo-
dodecane (0.53 mmol) and 929 µL of N,N-diisopropylethyl-
amine (5.33 mmol) in 40 mL of acetonitrile was added 216 µL
of 4-tert-butylbenzyl bromide (1.17 mmol). The resulting
mixture was warmed slowly to 60 °C and allowed to stir
overnight. The solution was concentrated in vacuo to give clear
liquid. Crude product was extracted by 3 × 20 mL of methylene
chloride, washed by 20 mL of brine, and dried by magnesium
sulfate, and the solvent was removed under reduced pressure.
The residue was purified via column chromatography on
alumina, eluting with ethyl acetate/hexane (1:3) to afford a
clear oil (343 mg, 88%): 1H NMR (CDCl3, 300 MHz) δ 1.27 (s,
18H), 2.64 (br d, J ) 21 Hz, 8H), 3.31-3.47 (br m, 8H), 3.55
(br s, 4H), 4.89 (br s, 4H), 7.18 (br m, 10H), 7.27 (m, 8H); 13C
NMR (CDCl3, 75 MHz) δ 31.3 (CCH3), 34.3 (CCH3), 45.0, 46.6,
47.6, 54.4 (NCH2Ph), 59.2, 59.9, 60.2, 66.8 (OCH2Ph), 125.1,
127.8, 128.0, 128.9, 135.8, 136.8, 149.7, 156.4 (NCO2). Anal.
Calcd for C46H60N4O4: C, 75.37; H, 8.25; N, 7.64. Found: C,
75.18; H, 8.53; N, 7.36.
1,4,7,10-Tetra(4-tert-butylbenzyl)-1,4,7,10-tetraaza-
cyclododecane (1). To a solution of 300 mg of cyclen (1.74
mmol) in 30 mL of acetonitrile were added 3033 µL of N,N-
diisopropylethylamine (17.41 mmol) and 1408 µL of 4-tert-
butylbenzyl bromide (7.66 mmol) consecutively. The reaction
mixture was warmed to 60 °C and allowed to stir for 24 h.
The precipitated white solid was filtered, washed with 10 mL
of acetonitrile, and dried in air (1091 mg, 83%): 1H NMR
(CDCl3) δ 1.28 (s, 36H), 2.70 (s, 16H), 3.43 (s, 8H), 7.25 (d, J
) 8.4 Hz, 8H), 7.37 (d, J ) 8.4 Hz, 8H); 13C NMR (CDCl3) δ
31.6 (CCH3), 34.5 (CCH3), 53.5 (cyclen ring CH2), 59.7 (CH2Ph),
125.2, 128.6, 137.4, 149.4. Anal. Calcd for C52H76N4: C, 82.48;
H, 10.12; N, 7.40. Found: C, 82.51; H, 10.19; N, 7.56.
1,4,7-Triformyl-10-(4-tert-butylbenzyl)-1,4,7,10-tetraaza-
cyclododecane (2b). To a solution of 1,4,7-triformyl-1,4,7,10-
tetraazacyclododecane (2624 mg, 10.24 mmol) in 80 mL of
acetonitrile were added N,N-diisopropylethylamine (3567 µL,
20.48 mmol) and 4-tert-butylbenzyl bromide (1881 µL, 10.24
mmol). The reaction was stirred at 60 °C for 2 days. After
evaporation of solvent, the residue was dissolved in 30 mL of
water, extracted by methylene chloride (3 × 20 mL), dried by
MgSO4, and concentrated to give a yellow oil. Crude product
was recrystallized from diethyl ether to give off-white powder
(3157 mg, 77%): 1H NMR (CDCl3, 300 MHz) δ 1.28 (s, 9H),
2.66-2.80 (m, 4H), 3.28-3.73 (m, 12H), 3.78 (s, 2H), 7.11 (dd,
J ) 11.0, 11.1 Hz, 2H), 7.33 (dd, J ) 11.0, 11.1 Hz, 2H), 7.88-
8.18 (m, 3H); 13C NMR (CDCl3, 75 MHz) δ 31.4 (CCH3), 34.6
(CCH3), 42.2, 43.3, 43.7, 44.1, 44.6, 45.4, 46.2, 46.8, 46.9, 48.1,
49.0, 50.3, 52.1, 54.2, 54.4, 54.8, 55.3, 56.5, 56.7, 57.4, 125.4
& 125.7 (Ph ring CH), 130.1 & 130.3 (Ph ring CH), 132.0 &
132.3 (Ph ring C), 151.0 & 151.3 (Ph ring C), 163.3, 163.4,
163.5, 163.6 & 163.8 (NCHO); HRMS (FAB) m/z 409.2789 ([M
+ Li]+, C22H34N4O3Li, calcd 409.2791). Anal. Calcd for
C22H34N4O3‚H2O: C, 62.83; H, 8.63; N, 13.32. Found: C, 63.79;
H, 8.92; N, 12.50.
1-(4-tert-butylbenzyl)-1,4,7,10-tetraazacyclododecane‚
4HCl (2c‚4HCl). 1,4,7-Triformyl-10-(4-tert-butylbenzyl)-1,4,7,10-
tetraazacyclododecane (3039 mg, 7.55 mmol) was dissolved in
30 mL of 2 M HCl and the solution was heated to 60 °C and
allowed to stir for 18 h. Solvent was evaporated under reduced
pressure at 50 °C to give an off-white solid. Ethanol (50 mL)
was added and the solution was agitated vigorously for 4 h.
White power was collected, washed by ethanol and ether, and
dried in air to give pure product as the HCl salt (2217 mg,
63%): 1H NMR (D2O) δ 1.27 (s, 9H), 2.90 (t, J ) 5.1 Hz, 4H),
2.98 (br s, 4H), 3.11 (t, J ) 5.1 Hz, 4H), 3.18 (br s, 4H), 3.83
1,7-Bis(4-tert-butylbenzyl)-1,4,7,10-tetraazacyclo-
dodecane (3). To a solution of 1,7-bis(benzyloxycarbonyl)-
4,10-bis(4-tert-butylbenzyl)-1,4,7,10-tetraazacyclododecane (976
mg, 1.33 mmol) in 50 mL of ethanol was added 10% Pd on
activated carbon (523 mg). The reaction was stirred under
hydrogen gas at room temperature for 10 h. Pd catalyst was
filtered off through a Celite pad and the filtrate was concen-
trated under reduced pressure at 40 °C to give clear oil (529
mg, 96%): 1H NMR (CDCl3) δ 1.34 (s, 18H), 2.70 (s, 16H), 3.67
(s, 4H), 7.31 (d, J ) 8.3 Hz, 4H), 7.41 (d, J ) 8.3 Hz, 4H); 13
C
(s, 2H), 7.31 (d, J ) 8.4 Hz, 2H), 7.52 (d, J ) 8.4 Hz, 2H); 13
NMR (D2O) δ 33.5 (CCH3), 37.1 (CCH3), 45.2, 45.4, 47.0, 51.5,
C
NMR (CDCl3) δ 31.4 (CCH3), 34.5 (CCH3), 46.0, 51.9, 60.1
(CH2Ph), 125.5, 128.8, 136.2, 150.3; HRMS (FAB) m/z 465.3955

6634 Journal of Medicinal Chemistry, 2004, Vol. 47, No. 26
Yoo et al.
([M + H]+, C30H49N4, calcd 465.3957). Anal. Calcd for C30H48N4‚
2H2O: C, 71.95; H, 10.47; N, 11.19. Found: C, 70.51; H, 10.44;
N, 11.56.
Calcd for C30H48N4‚4H2O: C, 67.13; H, 10.52; N, 10.44.
Found: C, 65.27; H, 9.34; N, 9.69.
1,4-Bis(tert-butoxycarbonylmethyl)-7,10-bis(4-tert-
butylbenzyl)-1,4,7,10-tetraazacyclododecane (6a). To a
solution of 287 mg of 1,4-bis(4-tert-butylbenzyl)-1,4,7,10-tetra-
azacyclododecane (0.62 mmol) in 30 mL of acetonitrile were
added 645 µL of N,N-diisopropylethylamine (3.70 mmol) and
201 µL of tert-butyl bromoacetate (1.36 mmol). The reaction
mixture was slowly heated to 60 °C with stirring for 24 h. After
evaporation of solvent in vacuo, 30 mL of water was added.
The reaction mixture was extracted with methylene chloride
(4 × 20 mL). The organics were dried over MgSO4, concen-
trated, and then purified by column chromatography on
alumina using ethyl acetate/ethanol (10:1) as eluent to give
pale yellow oil (207 mg, 48%): 1H NMR (CDCl3) δ 1.31 (s, 18H),
1.45 (s, 18H), 2.63 (br s, 8H), 2.91 (br s, 8H), 3.26 (s, 4H), 3.46
(s, 4H), 7.28 (s, 8H); 13C NMR (CDCl3) δ 28.4 (OCCH3), 31.6
(PhCCH3), 34.5 (PhCCH3), 52.3 (2C), 52.6, 52.7, 56.6 (CH2CO2),
59.8 (CH2Ph), 80.6 (OCCH3), 125.0, 128.8, 137.0, 149.5, 171.4
(CH2CO2).
1,4-Bis(carboxymethyl)-7,10-bis(4-tert-butylbenzyl)-
1,4,7,10-tetraazacyclododecane‚4HCl (6‚4HCl). HCl solu-
tion (6 M, 30 mL) was added to 1,4-bis(tert-butoxycarbonyl-
methyl)-7,10-bis(4-tert-butylbenzyl)-1,4,7,10-tetraazacyclo-
dodecane (207 mg, 0.30 mmol). The flask was refluxed for 12
h. All solvent was evaporated under reduced pressure and
crude product was recrystallized from diethyl ether to afford
white crystalline solid 6 as a salt (4HCl) (179 mg, 83%): 1H
NMR (CD3OD, 300 MHz) δ 1.22 (s, 18H), 3.30-3.50 (br m,
20H), 3.95 (br s, 4H), 7.39 (s, 8H); 13C NMR (CD3OD, 75 MHz)
δ 31.7 (CCH3), 35.6 (CCH3), 50.5 (br, cyclen ring CH2), 51.4
(br, cyclen ring CH2), 51.7 (br, cyclen ring CH2), 54.9 (CH2CO2),
58.5 (CH2Ph), 127.4 (Ph ring C), 127.5 (Ph ring CH), 131.9
(Ph ring CH), 153.8 (Ph ring C); HRMS (FAB) m/z 581.4074
([M + H]+, C34H53N4O4, calcd 581.4067).
1,4,7-Triformyl-10-(benzyloxycarbonyl)-1,4,7,10-tetra-
azacyclododecane (7a). A mixture of cyclen (800 mg, 4.64
mmol) and chloral hydrate (3047 mg, 18.42 mmol) was
dissolved in 30 mL of ethanol. The flask was stirred at 60 °C
for 4 h. The reaction mixture was concentrated under vacuum
to dryness and dissolved in 30 mL of water again (pH 9,
measured using pH paper). Benzyl chloroformate (1 mL) was
added and the reaction was stirred for 1 h. The pH of the
solution was adjusted to pH 10 from pH 4 by using saturated
Na2CO3 solution, and then 1 mL of benzyl chloroformate was
added again. After 1 h of stirring, the pH of the solution was
adjusted to pH 10 again and 1 mL of benzyl chloroformate was
added. The reaction mixture was allowed to stir overnight. The
aqueous solution was extracted by methylene chloride (4 ×
20 mL). The combined organic layer was washed by saturated
NaHCO3, dried over MgSO4, and concentrated under vacuum
to give a clear oil. The crude product was recrystallized from
diethyl ether and dried under reduced pressure to give a
hygroscopic white powder (1782 mg, 98%): 1H NMR (CDCl3,
300 MHz) δ 3.02-3.65 (m, 16H), 5.07 (s, 2H), 7.27 (br s, 5H),
7.88-8.09 (m, 3H); 13C NMR (CDCl3, 75 MHz) δ 43.0, 43.7,
43.8, 44.1, 44.7, 45.0, 45.8, 46.7, 46.9, 47.6, 48.1, 48.8, 49.4,
49.8, 50.1, 50.4, 50.5, 50.8, 52.2, 53.5 (cyclen ring CH2), 67.5,
67.6 (CH2Ph), 127.9, 128.3, 128.6, 129.0 (Ph ring CH), 135.8
(Ph ring C), 155.9, 156.8, 157.5 (CO2Ph), 163.0, 163.4, 163.6,
163.8, 163.9, 164.3, 164.6, 165.1, 165.7, 165.9 (NCHO); HRMS
(ESI) m/z 413.1791 ([M + Na]+, C19H26N4O5Na, calcd 413.1801).
Anal. Calcd for C19H26N4O5‚H2O: C, 55.87; H, 6.91; N, 13.72.
Found: C, 55.31; H, 6.68; N, 13.04.
1,7-Bis(tert-butoxycarbonylmethyl)-4,10-bis(4-tert-
butylbenzyl)-1,4,7,10-tetraazacyclododecane (4a). To a
solution of 1,7-bis(4-tert-butylbenzyl)-1,4,7,10-tetraazacyclo-
dodecane (194 mg, 0.42 mmol) in 50 mL of acetonitrile were
added N,N-diisopropylethylamine (727 µL, 4.17 mmol) and
tert-butyl bromoacetate (154 µL, 1.04 mmol). The reaction
mixture was slowly heated to 60 °C and allowed to stir for 18
h. After evaporation of solvents in vacuo, the residue was
dissolved in 20 mL of Na2CO3 solution, extracted by methylene
chloride (3 × 20 mL), washed by brine, dried over MgSO4, and
concentrated to give a clear oil. Crude product was filtered
through an alumina column using methylene chloride and then
eluted using ethyl acetate to give pure product as a clear oil
(168 mg, 58%): 1H NMR (CDCl3) δ 1.33 (s, 18H), 1.40 (s, 18H),
2.63 (m, 8H), 2.87 (m, 8H), 3.14 (s, 4H), 3.55 (s, 4H), 7.36 (s,
4H), 7.37 (s, 4H); 13C NMR (CDCl3) δ 28.3 (OCCH3), 31.6
(PhCCH3), 34.6 (PhCCH3), 52.5, 52.8, 56.5 (CH2Ph), 59.8
(CH2CO2), 80.7 (OCCH3), 125.2, 128.9, 137.4, 149.8, 171.7
(CH2CO2). Anal. Calcd for C42H68N4O4: C, 72.79; H, 9.89; N,
8.08. Found: C, 71.08; H, 9.80; N, 7.97.
1,7-Bis(carboxymethyl)-4,10-bis(4-tert-butylbenzyl)-
1,4,7,10-tetraazacyclododecane‚4HCl (4‚4HCl). HCl solu-
tion (6 M, 35 mL) was added to 168 mg (0.24 mmol) of 1,7-
bis(tert-butoxycarbonylmethyl)-4,10-bis(4-tert-butylbenzyl)-
1,4,7,10-tetraazacyclododecane. The flask was slowly heated
to 110 °C and allowed to stir for 18 h. After evaporation of all
solvents, the crude product was recrystallized from diethyl
ether to give white crystalline solid 4 as a salt (4HCl) (125
mg, 71%):1H NMR (D2O, 300 MHz) δ 1.28 (s, 18H), 2.82 (s,
4H), 2.91 (br d, J ) 15.6 Hz, 4H), 3.08 (br d, J ) 15.6 Hz, 4H),
3.42 (br s, 8H), 4.45 (s, 4H), 7.51 (d, J ) 8.0 Hz, 4H), 7.58 (d,
J ) 8.0 Hz, 4H); 13C NMR (D2O, 75 MHz) δ 30.0 (CCH3), 33.9
(CCH3), 47.9, 49.9, 53.2 (CH2CO2), 57.8 (CH2Ph), 125.4, 126.6,
130.8, 154.4, 173.9 (CH2CO2); HRMS (FAB) m/z 581.4069 ([M
+ H]+, C34H53N4O4, calcd 581.4067). Anal. Calcd for C34H52N4O4‚
4HCl: C, 56.20; H, 7.77; N, 7.71. Found: C, 56.45; H, 8.32; N,
7.91.
4,7-Bis(4-tert-butylbenzyl)-1,4,7,10-tetraazabicyclo[8.2.2]-
tetradecane-11,12-dione (5b). To a solution of 463 mg of
1,4,7,10-tetraazabicyclo[8.2.2]tetradecane-11,12-dione (cyclen-
oxamide) (2.04 mmol) in 30 mL of acetonitrile were added 3553
µL of N,N-diisopropylethylamine (20.4 mmol) and 825 µL of
4-tert-butylbenzyl bromide (4.49 mmol). The flask was stirred
at 60 °C overnight. All solvent was evaporated in vacuo. The
residue was dissolved again in 30 mL of water, extracted by 3
× 20 mL of methylene chloride, washed by 10 mL of brine,
dried by MgSO4, and concentrated to give pale yellow foam,
which was recrystallized in diethyl ether to give white crystal-
line solid (807 mg, 76%): 1H NMR (CDCl3, 300 MHz) δ 1.26
(s, 18H), 2.25 (m, 2H), 2.42 (br d, J ) 14.4 Hz, 2H), 2.59 (m,
4H), 2.89 (m, 2H), 3.38 (m, 4H), 3.53 (m, 2H), 4.12 (br d, J )
13.8 Hz, 2H), 4.32 (br s, 2H), 7.07 (d, J ) 7.4 Hz, 4H), 7.30 (d,
J ) 7.4 Hz, 4H); 13C NMR (CDCl3, 75 MHz) δ 31.4 (CCH3),
34.5 (CCH3), 47.6, 49.7, 52.5, 55.7, 58.6 (CH2Ph), 125.3, 129.5,
135.6, 150.3, 160.0 (NCO). Anal. Calcd for C32H46N4O2: C,
74.09; H, 8.94; N, 10.80. Found C, 74.42; H, 9.15; N, 10.79.
1,4-Bis(4-tert-butylbenzyl)-1,4,7,10-tetraazacyclo-
dodecane (5). A slurry of 113 mg of 4,7-bis(4-tert-butylbenzyl)-
1,4,7,10-tetraazabicyclo[8.2.2]tetradecane-11,12-dione (0.22
mmol) in a mixed solvent of 15 mL of 10 M NaOH and 10 mL
of ethanol was refluxed for 18 h. After removing the solvent,
50 mL of water was added and the mixture was extracted by
4 × 20 mL of methylene chloride. The organic layers were dried
over MgSO4 and concentrated over a vacuum. The crude
product was purified by filtration on alumina (ethyl acetate,
methanol) to give a clear oil (91 mg, 90%): 1H NMR (CDCl3)
δ 1.28 (s, 18H), 2.66 (s, 4H), 2.81 (m, 4H), 2.88 (m, 4H), 2.99
(s, 4H), 3.62 (s, 4H), 7.09 (d, J ) 8.1 Hz, 4H), 7.31 (d, J ) 8.1
Hz, 4H); 13C NMR (CDCl3) δ 31.4 (CCH3), 34.5 (CCH3), 46.6,
46.9, 50.5, 50.8, 55.9 (CH2Ph), 125.5, 129.7, 134.1, 150.7. Anal.
1-(Benzyloxycarbonyl)-1,4,7,10-tetraazacyclodo-
decane‚3HCl (7b‚3HCl). 1,4,7-Triformyl-10-(benzyloxycar-
bonyl)-1,4,7,10-tetraazacyclododecane (731 mg, 1.87 mmol) was
dissolved in 40 mL of 1 M HCl solution and the solution was
stirred at 50 °C for 5 h. Solvent was evaporated completely
under vacuum at 60 °C to give a white solid. Crude product
was refluxed in 20 mL of ethanol, cooled to room temperature,
filtrated, washed with 5 mL of ether, and dried in air. Excess
ether was added to the ethanol filtrate to see the cloudiness
of the solution. Precipitated white powder was collected,

Cyclen-Based Copper-64 Complexes
Journal of Medicinal Chemistry, 2004, Vol. 47, No. 26 6635
washed by a small amount of ether, and dried in air as a
second crop (633 mg, 81%): 1H NMR (D2O, 300 MHz) δ 3.19
(br s, 12H), 3.69 (br t, J ) 5.1 Hz, 4H), 5.17 (s, 2H), 7.44 (br
s, 5H); 13C NMR (D2O, 75 MHz) δ 45.8, 47.5, 48.2, 49.5, 71.6
(CH2Ph), 131.4, 131.8, 131.9, 138.7, 161.4 (CO2CH2); HRMS
(FAB) m/z 307.2139 ([M + H]+, C16H27N4O2, calcd 307.2134).
Anal. Calcd for C16H26N4O2‚3HCl: C, 46.22; H, 7.03; N, 13.47;
Cl, 25.58. Found: C, 45.67; H, 7.42; N, 12.94; Cl, 25.34.
149.38 & 149.48 & 149.66 (Ph ring C), 171.47 & 171. 54 (CO2).
Anal. Calcd for C47H72N4O2‚H2O: C, 75.96; H, 10.04; N, 7.54.
Found: C, 74.57; H, 10.06; N, 7.60.
1-(Carboxymethyl)-4,7,10-tris(4-tert-butylbenzyl)-
1,4,7,10-tetraazacyclododecane (8). 1-(tert-Butoxycarbonyl-
methyl)-4,7,10-tris(4-tert-butylbenzyl)-1,4,7,10-tetraazacyclo-
dodecane (36 mg, 0.05 mmol) was dissolved in 25 mL of
trifluoroacetic acid and the solution was stirred at room
temperature for 18 h. After evaporation of solvent, the residue
was dissolved in 10 mL of methylene chloride, washed by water
(10 mL) and brine (10 mL), dried over MgSO4, and concen-
trated to give clear oil (33 mg, 99%): 1H NMR (CDCl3, 300
MHz, no HCl) δ 1.28 (s, 27H), 2.75 (br s, 6H), 2.82 (br s, 6H),
3.12 (br s, 4H), 3.45 (s, 2H), 3.56 (s, 2H), 3.63 (br s, 4H), 6.90
(d, J ) 7.8 Hz, 2H), 7.21-7.30 (m, 6H), 7.35 (d, J ) 7.8 Hz,
4H); 13C NMR (CDCl3, 75 MHz, no HCl) δ 31.5 (CCH3), 34.7
(CCH3), 50.4, 50.6, 51.8, 54.7, 57.4, 57.7, 60.2, 125.5 & 125.7
(Ph ring CH), 129.8 & 130.0 (Ph ring CH), 131.7 & 133.4 (Ph
ring C), 151.0 & 151.2 (Ph ring C), 171.6 (CO2); HRMS (FAB)
m/z 669.5112 ([M + H]+, C43H65N4O2, calcd 669.5108).
Copper(II) Complex of 2c, [Cu(2c)Cl]+(ClO4)-. To a
aqueous solution (10 mL) of Cu(ClO4)2‚6H2O (122 mg, 0.33
mmol) was added a water solution (10 mL) of 2c‚4HCl (139
mg, 0.30 mmol). The mixture was stirred at room temperature
overnight. Blue precipitates were collected by a sintered glass
funnel and washed with small amount of water, ethanol, and
ether, then dried in air (120 mg, 78%). A single crystal for
X-ray structure determination was grown by vapor diffusion
of ether into a DMF solution of 2c: HRMS (FAB) m/z 416.1779
([M - (ClO4)]+, C19H34ClCuN4, calcd 416.1768). Anal. Calcd
for C19H34Cl2CuN4O4: C, 44.14; H, 6.63; N, 10.84. Found: C,
44.14; H, 6.90; N, 10.51.
1-(Benzyloxycarbonyl)-4,7,10-tris(4-tert-butylbenzyl)-
1,4,7,10-tetraazacyclododecane (7c). N,N-diisopropylethyl-
amine (6995 µL, 40.16 mmol) was added to a slurry of
1-(benzyloxycarbonyl)-1,4,7,10-tetraazacyclododecane‚3HCl
(2269 µL, 5.46 mmol) in 40 mL of acetonitrile to make the
mixture dissolved clearly. 4-tert-Butylbenzyl bromide (3229 µL,
17.57 mmol) was added to the solution dropwise at room
temperature. The reaction mixture was slowly heated to 60
°C and allowed to stir for 15 h. Precipitated white powder was
collected, washed by a small amount of acetonitrile and ether,
and dried to give the first crop (394 mg). The yellow filtrate
was concentrated to dryness, and the residue was dissolved
in 5 mL of toluene and then excess hexane was added to
precipitate the yellow oil, which was recrystallized in ether to
1
give a white powder (1218 mg): yield 40%; H NMR (CDCl3,
300 MHz) δ 1.25 (s, 9H), 1.31 (s, 9H), 1.32 (s, 9H), 2.57 (br s,
8H), 2.88 (br t, J ) 5.7 Hz, 2H), 3.09 (br t, J ) 5.7 Hz, 2H),
3.40 (s, 2H), 3.41 (s, 2H), 3.55 (s, 2H), 3.62 (br t, J ) 5.7 Hz,
2H), 3.73 (br t, J ) 5.7 Hz, 2H), 4.93 (s, 2H), 7.10-7.34 (m,
17H); 13C NMR (CDCl3, 75 MHz) δ 31.56 & 31.63 (CCH3), 34.53
& 34.61 (CCH3), 46.9, 48.0, 52.2, 52.5, 52.7, 53.4 & 53.5
(NCH2Ph), 59.0, 59.5, 59.9, 67.0 (OCH2Ph), 125.1 & 125.2
(NCH2Ph ring CH), 127.9, 128.0, 128.5, 128.9 & 129.0 (NCH2Ph
ring CH), 136.6 & 136.7 (OCH2Ph ring CH), 136.9 & 137.2
(NCH2Ph ring C), 149.6 & 149.7 (NCH2Ph ring C), 156.6
(NCO2); HRMS (FAB) m/z 745.5403 ([M + H]+, C43H65N4O2,
calcd 745.5421). Anal. Calcd for C49H68N4O2: C, 78.99; H, 9.20;
N, 7.52. Found: C, 78.47; H, 9.40; N, 7.49.
CAUTION! Perchlorate salts of metal complexes with
organic ligands are potentially explosive.
X-ray Crystallography. Crystals of appropriate dimension
were mounted on glass fibers in random orientation. Prelimi-
nary examination and data collection was performed at -55
°C using a Bruker SMART Charge Coupled Device (CCD)
Detector system single-crystal X-ray diffractometer equipped
with a sealed tube X-ray source using graphite-monochromated
Mo KR radiation (λ ) 0.71073 Å). Preliminary unit cell
constants were determined with a set of 45 narrow frames (0.3°
in $) scans. The data set collected consists of 4028 frames of
intensity data collected with a frame width of 0.3° in $ and
counting time of 30 s/frame at a crystal to detector distance of
4.950 cm. The double pass method of scanning was used to
exclude any noise. The collected frames were integrated using
an orientation matrix determined from the narrow frame
scans. SMART and SAINT software packages (Bruker Ana-
lytical X-ray, Madison, WI, 1998) were used for data collection
and data integration. Analysis of the integrated data did not
show any decay. Final cell constants were determined by a
global refinement of xyz centroids from all data. Collected data
were corrected for systematic errors using SADABS (Blessing,
R. H. Acta Crystallogr. 1995, A51, 33-38) based upon the Laue
symmetry using equivalent reflections. Structure solution and
refinement were carried out using the SHELXTL-PLUS
software package (Sheldrick, G. M., Bruker Analytical X-ray
Division, Madison, WI, 2000). The structures were solved by
direct methods and refined successfully in the space groups
Pnma. Full matrix least-squares refinement was carried out
1,4,7-Tris(4-tert-butylbenzyl)-1,4,7,10-tetraazacyclo-
dodecane (7). To a solution of 1-(benzyloxycarbonyl)-4,7,10-
tris(4-tert-butylbenzyl)-1,4,7,10-tetraazacyclododecane (265 mg,
0.35 mmol) in 30 mL of ethanol was added 10% Pd/C (250 mg).
The reaction mixture was stirred under hydrogen gas at
ambient pressure for 24 h. Pd catalyst was filtered off by a
pad of Celite and the filtrate was concentrated under vacuum
to give a clear oily solid (186 mg, 87%). A microanalysis sample
was prepared by recrystallization from methylene chloride/
ether: 1H NMR (CDCl3) δ 1.26 (s, 9H), 1.31 (s, 18H), 2.78 (br
s, 12H), 2.96 (br s, 6H), 3.47 (s, 2H), 3.74 (s, 4H), 6.64 (d, J )
8.1 Hz, 2H), 7.20 (d, J ) 8.1 Hz, 2H), 7.34 (d, J ) 8.1 Hz, 4H),
7.44 (d, J ) 8.1 Hz, 4H); 13C NMR (CDCl3) δ 31.4 & 31.5
(CCH3), 34.5 & 34.6 (CCH3), 49.2, 49.6, 50.3, 51.3, 62.5
(NCH2Ph), 125.2 & 125.4 (Ph ring CH), 129.48 & 129.52 (Ph
ring CH), 132.3 & 135.5 (Ph ring C), 150.1 & 151.0 (Ph ring
C); HRMS (ESI) m/z 611.5045 ([M + H]+, C43H65N4O2, calcd
611.5053). Anal. Calcd for C41H62N4‚H2O: C, 78.29; H, 10.26;
N, 8.91. Found: C, 75.26; H, 10.78; N, 11.90.
1-(tert-Butoxycarbonylmethyl)-4,7,10-tris(4-tert-butyl-
benzyl)-1,4,7,10-tetraazacyclododecane (8a). To a solution
of 1,4,7-tris(4-tert-butylbenzyl)-1,4,7,10-tetraazacyclododecane
(426 mg, 0.70 mmol) were added N,N-diisopropylethylamine
(242 µL, 1.39 mmol) and tert-butyl bromoacetate (155 µL, 1.05
mmol). The flask was heated slowly to 60 °C and allowed to
stir for 16 h. After evaporation of solvent in vacuo, 20 mL of
water was added and then crude product was extracted by 3
× 20 mL of methylene chloride, dried over MgSO4, and
concentrated to give pale oil, which was further purified by
column chromatography over alumina using ethanol:chloro-
form (1:20) as eluent to give clear oil (268 mg, 53%): 1H NMR
(CDCl3) δ 1.30 (s, 9H), 1.35 & 1.37 (s, 18H), 1.44 & 1.45 (s,
9H), 2.69 (br s, 8H), 2.93 (br m, 6H), 3.17 & 3.21 (s, 2H), 3.42
(s, 2H), 3.51 (s, 4H), 3.58 (s, 2H), 7.22-7.42 (m, 12H); 13C NMR
(CDCl3) δ 28.5 (OCCH3), 31.6 & 31.7 (PhCCH3), 34.56 & 34.61
(PhCCH3), 52.6, 52.87, 52.93, 53.02, 53.10, 53.22, 56.56, 56.68,
59.70, 59.88, 80.62 (OCCH3), 125.04 & 125.12 (Ph ring CH),
128.68 & 128.81 (Ph ring CH), 137.19 & 137.33 (Ph ring C),
2
by minimizing ∑w(Fo - Fc2)2. The non-hydrogen atoms were
refined anisotropically to convergence. The hydrogen atoms
were treated using appropriate riding model (AFIX m3).
Radiochemistry. 64Cu chloride was converted to 64Cu
acetate or citrate by stirring with 30-100 µL of 0.1 M
ammonium acetate, pH 6.4, or 0.1 M ammonium citrate, pH
6.5, except for ligand 5, where 64Cu chloride was simply diluted
by water. 64Cu acetate or citrate or chloride (0.7-2.3 mCi) was
added to a ligand solution containing 0.3-1.7 mg of ligand in
100-300 µL of a mixture of ethanol and buffer (see Table 1
for specific reaction conditions for each ligand). The amount
of ethanol used was determined on the basis of solubilities of
ligand and copper complex of ligand. For ligand 2, only 0.1 M

6636 Journal of Medicinal Chemistry, 2004, Vol. 47, No. 26
Yoo et al.
(3) Chappell, L. L.; Ma, D.; Milenic, D. E.; Garmestani, K.; Venditto,
V.; et al. Synthesis and evaluation of novel bifunctional chelating
agents based on 1,4,7,10-tetraazacyclododecane-N,N′,N′′,N′′′-
tetraacetic acid for radiolabeling proteins. Nucl. Med. Biol. 2003,
30, 581-595.
ammonium acetate, pH 6.4, was used as labeling solvent. The
reaction mixture was incubated at temperature in the range
from room temperature to 60 °C. Reaction times ranged from
15 min to 7 h. The radiochemical purity was determined by
radio-TLC using silica gel plates (for ligand 1, 2, 3, 4, 6) or
reverse phase C8 plates (for 5, 7, 8). Developing solvents were
chosen depending on the TLC plate and the polarity of 64Cu-
labeled complex.
(4) Chong, H.-s.; Garmestani, K.; Ma, D.; Milenic, D. E.; Overstreet,
T.; et al. Synthesis and Biological Evaluation of Novel Macro-
cyclic Ligands with Pendent Donor Groups as Potential Yttrium
Chelators for Radioimmunotherapy with Improved Complex
Formation Kinetics. J. Med. Chem. 2002, 45, 3458-3464.
(5) Jamar, F.; Barone, R.; Mathieu, I.; Walrand, S.; Labar, D.; et
al. 86Y-DOTA0-d-Phe1-Tyr3-octreotide (SMT487)sA phase 1
clinical study: Pharmacokinetics, biodistribution and renal
protective effect of different regimens of amino acid co-infusion.
Eur. J. Nucl. Med. Mol. Imaging 2003, 30, 510-518.
(6) Breeman, W. A. P.; de Jong, M.; Visser, T. J.; Erion, J. L.;
Krenning, E. P. Optimising conditions for radiolabeling of
DOTA-peptides with 90Y, 111In and 177Lu at high specific
activities. Eur. J. Nucl. Med. Mol. Imaging 2003, 30, 917-920.
(7) Reichert, D. E.; Lewis, J.; Anderson, C. J. Metal complexes as
diagnostic tools. Coord. Chem. Rev. 1999, 184, 3-66.
(8) Sibert, J. W.; Cory, A. H.; Cory, J. G. Lipophilic derivatives of
cyclam as new inhibitors of tumor cell growth. Chem. Commun.
(Cambridge, United Kingdom) 2002, 154-155.
(9) Liang, X.; Parkinson, J. A.; Weishaeupl, M.; Gould, R. O.; Paisey,
S. J.; et al. Structure and Dynamics of Metallomacrocycles:
Recognition of Zinc Xylyl-Bicyclam by an HIV Coreceptor. J. Am.
Chem. Soc. 2002, 124, 9105-9112.
(10) Bianchi, A.; Calabi, L.; Giorgi, C.; Losi, P.; Mariani, P.; et al.
Thermodynamic and structural aspects of manganese(II) com-
plexes with polyaminopolycarboxylic ligands based upon 1,4,7,10-
tetraazacyclododecane (cyclen). Crystal structure of dimeric
[MnL]2·2CH3OH containing the new ligand 1,4,7,10-tetraaza-
cyclododecane-1,4-diacetate. J. Chem. Soc., Dalton Trans. 2001,
917-922.
(11) Chappell, L. L.; Voss, D. A., Jr.; Horrocks, W. D., Jr.; Morrow,
J. R. Effect of Mixed Pendent Groups on the Solution and
Catalytic Properties of Europium(III) Macrocyclic Complexes:
Bifunctional and Monofunctional Amide and Alcohol Pendents
in Septadentate or Octadentate Ligands. Inorg. Chem. 1998, 37,
3989-3998.
(12) Aoki, S.; Kawatani, H.; Goto, T.; Kimura, E.; Shiro, M. A Double-
Functionalized Cyclen with Carbamoyl and Dansyl Groups
(Cyclen ) 1,4,7,10-Tetraazacyclododecane): A Selective Fluo-
rescent Probe for Y3+ and La3+. J. Am. Chem. Soc. 2001, 123,
1123-1132.
(13) Anderson, C. J.; Lewis, J. S. Radiopharmaceuticals for targeted
radiotherapy of cancer. Expert Opin. Ther. Patents 2000, 10,
1057-1069.
(14) Connett, J. M.; Anderson, C. J.; Guo, L.-W.; Schwarz, S. W.; Zinn,
K. R.; et al. Radioimmunotherapy with a 64Cu-labeled mono-
clonal antibody: A comparison with 67Cu. Proc. Natl. Acad. Sci.
U.S.A. 1996, 93, 6814-6818.
Determination of Partition Coefficients. The partition
coefficients were determined by adding 2-20 µL of 64Cu-
labeled complex (10-39 µCi) to a solution containing 1.5 mL
of octanol and 1.5 mL of buffer solution (0.01 M ammonium
acetate, pH 7.4), which are obtained from saturated octanol
buffer solutions. The resulting solutions were then vortexed
for 1 min and centrifuged for 5 min at 5000 rpm. A 1 mL
aliquot of the octanol layer was removed, back-extracted with
1 mL of buffer, vortexed, and centrifuged for 2 min at 5000
rpm. Aliquots (150 µL) of octanol and buffer were removed,
weighed, and counted (first partition coefficient). A 700 µL
aliquot of the previous octanol layer was removed and back-
extracted again with 700 µL of buffer, vortexed, and centri-
fuged as before. Aliquots (150 µL) of octanol and buffer were
removed, weighed, and counted (second partition coefficient).
A 400 µL aliquot of octanol layer of second back-extraction was
removed and back-extracted one more time with 400 µL of
buffer, vortexed, and centrifuged as before. Aliquots (150 µL)
of octanol and buffer were removed, weighed, and counted
(third partition coefficient). The partition coefficient was
calculated as a ratio of counts in the octanol fraction of unit
mass to counts in the buffer fraction of unit mass per back-
extraction. The average log P value of the three back-
extractions is reported.
Biodistribution Studies. The radioactive complexes were
diluted with saline for injection, except for 64Cu-1 and 64Cu-
3, where 0.1 M ammonium acetate buffer (pH 6.4) was used
instead. Mature female Sprague-Dawley rats (n ) 4 per time
point) weighing 170-240 g were anesthetized and injected
with 17-32 µCi of activity in a volume of 100-160 µL via the
tail vein. At selected time points postinjection (15 min, 4 h,
and 24 h), rats were sacrificed. Organs of interest (liver,
kidney, muscle, heart, and blood) were removed, weighted, and
counted. The percent of injected dose per gram (%ID/g) and
percent injected dose per organ (%ID/organ) were calculated
by comparison to weighed, counted standard solutions. All
animal experiments were conducted in compliance with the
Guidelines for the Care and Use of Research Animals estab-
lished by Washington University’s Animal Studies Committee.
(15) Boswell, C. A.; Sun, X.; Niu, W.; Weisman, G. R.; Wong, E. H.;
et al. Comparative in Vivo Stability of Copper-64-Labeled Cross-
Bridged and Conventional Tetraazamacrocyclic Complexes. J.
Med. Chem. 2004, 47, 1465-1474.
Acknowledgment. The authors thank Lynne Jones
and Nichole Mercer for technical assistance and thank
Dr. Nigam P. Rath in University of Missouri at St. Louis
for crystal structure determination. This work was
supported by the United States Department of Energy
(DEFG02-087ER-60512). Mass spectrometry was pro-
vided by the Washington University Mass Spectrometry
Resource, an NIH Research Resource (Grant No.
P41RR00954).
(16) Lewis, J. S.; Laforest, R.; Buettner, T. L.; Song, S.-K.; Fujiba-
yashi, Y.; et al. Copper-64-diacetyl-bis(N4-methylthiosemicar-
bazone): An agent for radiotherapy. Proc. Natl. Acad. Sci. U.S.A.
2001, 98, 1206-1211.
(17) McCarthy, D. W.; Shefer, R. E.; Klinkowstein, R. E.; Bass, L.
A.; Margeneau, W. H.; et al. Efficient production of high specific
activity 64Cu using a biomedical cyclotron. Nucl. Med. Biol.
1997, 24, 35-43.
(18) Yoo, J.; Reichert, D. E.; Welch, M. J. Regioselective N-substitu-
tion of cyclen with two different alkyl groups: Synthesis of all
possible isomers. Chem. Commun. (Cambridge, United Kingdom)
2003, 766-767.
(19) Jones-Wilson, T. M.; Deal, K. A.; Anderson, C. J.; McCarthy, D.
W.; Kovacs, Z.; et al. The in vivo behavior of copper-64-labeled
azamacrocyclic complexes. Nucl. Med. Biol. 1998, 25, 523-530.
(20) Denat, F.; Brandes, S.; Guilard, R. Strategies for the regio-
selective N-functionalization of tetraazacycloalkanes. From cy-
clam and cyclen towards more sophisticated molecules. Synlett
2000, 561-574.
(21) Theodoridis, G. Nitrogen protecting groups: Recent develop-
ments and new applications. Tetrahedron 2000, 56, 2339-2358.
(22) Boldrini, V.; Giovenzana, G. B.; Pagliarin, R.; Palmisano, G.;
Sisti, M. Expeditious N-monoalkylation of 1,4,7,10-tetraaza-
cyclododecane (cyclen) via formamido protection. Tetrahedron
Lett. 2000, 41, 6527-6530.
(23) Liu, S.; Edwards, D. S. Bifunctional Chelators for Therapeutic
Lanthanide Radiopharmaceuticals. Bioconjugate Chem. 2001,
12, 7-34.
Supporting Information Available: X-ray crystallo-
graphic files for Cu-2c, including crystal data and structure
refinement, atomic coordinates, and equivalent isotropic dis-
placements parameters, bond lengths and angles, anisotropic
displacement paramagnets, hydrogen coordinates, and projec-
tion plot with 30% probability ellipsoids. This material is
available free of charge via the Internet at http://pubs.acs.org.
References
(1) Anderson, C. J.; Welch, M. J. Radiometal-Labeled Agents (Non-
Technetium) for Diagnostic Imaging. Chem. Rev. (Washington,
D. C.) 1999, 99, 2219-2234.
(2) Caravan, P.; Ellison, J. J.; McMurry, T. J.; Lauffer, R. B.
Gadolinium(III) Chelates as MRI Contrast Agents: Structure,
Dynamics, and Applications. Chem. Rev. (Washington, D. C.)
1999, 99, 2293-2352.
(24) Kovacs, Z.; Sherry, A. D. A general synthesis of 1,7-disubstituted
1,4,7,10-tetraazacyclododecanes. J. Chem. Soc., Chem. Commun.
1995, 185-186.

Cyclen-Based Copper-64 Complexes
Journal of Medicinal Chemistry, 2004, Vol. 47, No. 26 6637
(25) Bellouard, F.; Chuburu, F.; Kervarec, N.; Toupet, L.; Triki, S.;
et al. cis-Diprotected cyclams and cyclens: A new route to
symmetrically or asymmetrically 1,4-disubstituted tetraaza-
macrocycles and to asymmetrically tetrasubstituted derivatives.
J. Chem. Soc., Perkin Trans. 1 1999, 3499-3505.
(26) Edwards, W. B.; Reichert, D. E.; Welch, M. J.; d’Avignon, D. A.
b-Cyclodextrin dimers as potential tumor pre-targeting agents.
Chem. Commun. (Cambridge, United Kingdom) 2001, 1312-
1313.
(27) Kimura, E.; Aoki, S.; Koike, T.; Shiro, M. A Tris(zinc-1,4,7,10-
tetraazacyclododecane) Complex as a New Receptor for Phos-
phate Dianions in Aqueous Solution. J. Am. Chem. Soc. 1997,
119, 3068-3076.
(28) Azuma, N.; Kohno, Y.; Nemoto, F.; Kajikawa, Y.; Ishizu, K.; et
al. Spectroscopic studies on copper(II) complexes of chiral
cyclens: [CuN4Cl] chromophores varying from square pyramidal
to trigonal bipyramidal stereochemistry. Inorg. Chim. Acta 1994,
215, 109-121.
(29) Styka, M. C.; Smierciak, R. C.; Blinn, E. L.; DeSimone, R. E.;
Passariello, J. V. Copper(II) complexes containing a 12-mem-
bered macrocyclic ligand. Inorg. Chem. 1978, 17, 82-86.
(30) Webb, R. L.; Mino, M. L.; Blinn, E. L.; Pinkerton, A. A.
Comparison of copper(II) complexes containing various saturated
12-membered macrocycles. Inorg. Chem. 1993, 32, 1396-1402.
(31) Soibinet, M.; Dechamps-Olivier, I.; Guillon, E.; Barbier, J.-P.;
Aplincourt, M.; et al. XAS, ESR and potentiometric studies
of three dinuclear N,N′-para-xylylenebis(tetraazamacrocycle)-
copper(II) complexessX-ray crystal structure of [N,N′-p-xylyl-
enebis(cyclen)]copper(II). Eur. J. Inorg. Chem. 2003, 1984-1994.
(32) DeSimone, R. E.; Blinn, E. L.; Mucker, K. F. Structural aspects
of a macrocyclic metal complex containing a copper(II) ion with
unusual spectral and kinetic properties. Inorg. Nucl. Chem. Lett.
1980, 16, 23-26.
(36) Kim, S.-Y.; Jung, I.-S.; Lee, E.; Kim, J.; Sakamoto, S.; et al.
Macrocycles within macrocycles: Cyclen, cyclam, and their
transition metal complexes encapsulated in cucurbit[8]uril.
Angew. Chem., Int. Ed. 2001, 40, 2119-2121.
(37) Bhattacharya, P. K. Ring size effect on copper(II) complexes of
cyclic tetraamines. Indian J. Chem., Sect. A 1980, 19A, 1086-
1088.
(38) Tsang, B. W.; Mathias, C. J.; Fanwick, P. E.; Green, M. A.
Structure-distribution relationships for metal-labeled myo-
cardial imaging agents: Comparison of a series of cationic
gallium(III) complexes with hexadentate bis(salicylaldimine)
ligands. J. Med. Chem. 1994, 37, 4400-4406.
(39) Di Vaira, M.; Mani, F.; Stoppioni, P. Synthesis and coordination
chemistry of 1,7-bis(carboxymethyl)-4,10-bis(1-methylimidazol-
2-ylmethyl)-1,4,7,10-tetraazacyclododecane. J. Chem. Soc., Dal-
ton Trans. 1998, 1879-1884.
(40) Meyer, M.; Dahaoui-Gindrey, V.; Lecomte, C.; Guilard, R.
Conformations and coordination schemes of carboxylate and
carbamoyl derivatives of the tetraazamacrocycles cyclen and
cyclam, and the relation to their protonation states. Coord.
Chem. Rev. 1998, 178-180, 1313-1405.
(41) Sun, X.; Wuest, M.; Weisman, G. R.; Wong, E. H.; Reed, D. P.;
et al. Radiolabeling and In Vivo Behavior of Copper-64-Labeled
Cross-Bridged Cyclam Ligands. J. Med. Chem. 2002, 45, 469-
477.
(42) Sun, Y.; Anderson, C. J.; Pajeau, T. S.; Reichert, D. E.; Hancock,
R. D.; et al. Indium(III) and Gallium(III) Complexes of Bis-
(aminoethanethiol) Ligands with Different Denticities: Stabili-
ties, Molecular Modeling, and in Vivo Behavior. J. Med. Chem.
1996, 39, 458-470.
(43) Cutler, C. S.; Giron, M. C.; Reichert, D. E.; Snyder, A. Z.; Herrero,
P.; et al. Evaluation of gallium-68 tris(2-mercaptobenzyl)-
amine: A complex with brain and myocardial uptake. Nucl. Med.
Biol. 1999, 26, 305-316.
(33) Yoo, J.; Reichert, D. E.; Welch, M. J. Unpublished results.
(34) Miyoshi, K.; Tanaka, H.; Kimura, E.; Tsuboyama, S.; Murata,
S.; et al. Electrochemical and spectroscopic studies on copper(II)
complexes of macrocyclic ligands as models for square-pyramidal
metal active sites of copper(II) complexes of bleomycin and
glutathione. Inorg. Chim. Acta 1983, 78, 23-30.
(44) Cutler, C. S.; Wuest, M.; Anderson, C. J.; Reichert, D. E.; Sun,
Y.; et al. Labeling and in vivo evaluation of novel copper(II)
dioxotetraazamacrocyclic complexes. Nucl. Med. Biol. 2000, 27,
375-380.
(35) Hay, R. W.; Pujari, M. P. The dissociation of [Cu(cyclen)]2+ in
acid solution. A kinetic study. Inorg. Chim. Acta 1985, 100, L1-
L3.
JM0496990