Triamidetriamine bearing macrobicyclic and macrotricyclic ligands: Potential applications in the development of copper-64 radiopharmaceuticals

Article abstract of DOI:10.1021/ic4024508

A versatile and straightforward synthetic approach is described for the preparation of triamide bearing analogues of sarcophagine hexaazamacrobicyclic cage ligands without the need for a templating metal ion. Reaction of 1,1,1-tris(aminoethyl)ethane (tame) with 3 equiv of 2-chloroacetyl chloride, yields the tris(α-chloroamide) synthetic intermediate 6, which when treated with either 1,1,1-tris(aminoethyl)ethane or 1,4,7-triazacyclononane furnished two novel triamidetriamine cryptand ligands (7 and 8 respectively). The Co(III) and Cu(II) complexes of cryptand 7 were prepared; however, cryptand 8 could not be metalated. The cryptands and the Co(III) complex 9 have been characterized by elemental analysis, ¹H and ¹³C NMR spectroscopy, and X-ray crystallography. These studies confirm that the Co(III) complex 9 adopts an octahedral geometry with three facial deprotonated amido-donors and three facial amine donor groups. The Cu(II) complex 10 was characterized by elemental analysis, single crystal X-ray crystallography, cyclic voltammetry, and UV-visible absorption spectroscopy. In contrast to the Co(III) complex (9), the Cu(II) center adopts a square planar coordination geometry, with two amine and two deprotonated amido donor groups. Compound 10 exhibited a quasi-reversible, one-electron oxidation, which is assigned to the Cu^2+/3+ redox couple. These cryptands represent interesting ligands for radiopharmaceutical applications, and 7 has been labeled with ⁶⁴Cu to give ⁶⁴Cu-10. This complex showed good stability when subjected to L-cysteine challenge whereas low levels of decomplexation were evident in the presence of L-histidine.

Full text of DOI:10.1021/ic4024508

Article
pubs.acs.org/IC
Triamidetriamine Bearing Macrobicyclic and Macrotricyclic Ligands:
Potential Applications in the Development of Copper-64
Radiopharmaceuticals
†
‡
§
†
‡
Kel Vin Tan, Paul A. Pellegrini, Brian W. Skelton, Conor F. Hogan, Ivan Greguric,
,†
and Peter J. Barnard*
†
Department of Chemistry, La Trobe Institute for Molecular Science, La Trobe University, Victoria 3086, Australia
ANSTO LifeSciences, Australian Nuclear Science and Technology Organisation, Menai, New South Wales 2234, Australia
Centre for Microscopy, Characterisation and Analysis, The University of Western Australia, 35 Stirling Highway, Crawley, Western
‡
§
Australia 6009, Australia
*
S Supporting Information
ABSTRACT: A versatile and straightforward synthetic approach is
described for the preparation of triamide bearing analogues of
sarcophagine hexaazamacrobicyclic cage ligands without the need for
a templating metal ion. Reaction of 1,1,1-tris(aminoethyl)ethane
(
tame) with 3 equiv of 2-chloroacetyl chloride, yields the tris(α-
chloroamide) synthetic intermediate 6, which when treated with
either 1,1,1-tris(aminoethyl)ethane or 1,4,7-triazacyclononane fur-
nished two novel triamidetriamine cryptand ligands (7 and 8
respectively). The Co(III) and Cu(II) complexes of cryptand 7 were
prepared; however, cryptand 8 could not be metalated. The
cryptands and the Co(III) complex 9 have been characterized by
1
13
elemental analysis, H and C NMR spectroscopy, and X-ray
crystallography. These studies confirm that the Co(III) complex 9
adopts an octahedral geometry with three facial deprotonated amido-donors and three facial amine donor groups. The Cu(II)
complex 10 was characterized by elemental analysis, single crystal X-ray crystallography, cyclic voltammetry, and UV−visible
absorption spectroscopy. In contrast to the Co(III) complex (9), the Cu(II) center adopts a square planar coordination
geometry, with two amine and two deprotonated amido donor groups. Compound 10 exhibited a quasi-reversible, one-electron
2
+/3+
oxidation, which is assigned to the Cu
redox couple. These cryptands represent interesting ligands for radiopharmaceutical
6
4
64
applications, and 7 has been labeled with Cu to give Cu-10. This complex showed good stability when subjected to L-cysteine
challenge whereas low levels of decomplexation were evident in the presence of L-histidine.
INTRODUCTION
based imaging applications. For example, the Cu(II) complex of
NH ) Sar (1, Figure 1), displays very high thermodynamic
■
(
2
2
Copper-64 is a promising radionuclide for positron emission
tomography (PET) imaging. It is not only a positron-emitter
stability and, because of encapsulation of the metal ion in the
1
+
−
−
cage ligand, is inert to metal dissociation. In early studies, a
(
5
19%, E
= 656 keV) but also a β -emitter (39.6%, E
=
β max
β
max
(
NH ) Sar derivative - SarAr (2, Figure 1) was developed to
73 keV) with a half-life of 12.7 h and may therefore be used
2 2
1
for diagnostic and therapeutic applications. Numerous
chelating ligand systems have been evaluated for the develop-
ment of 64Cu-based PET imaging agents, and several excellent
allow the efficient bioconjugation of sarcophagine ligands, and
6
4
Cu-2 monoclonal antibody (mAb) immunoconjugates were
used for PET imaging of neuroblastoma and melanoma
1
−5
10−12
reviews exist. The chemistry of Cu in vivo is complicated by
the potential for bioreduction, where Cu(II) is reduced to
Cu(I) followed by transchelation, and this process is
fundamental to the mechanism by which Cu(II) thiosemicar-
bazone complexes (e.g., Cu-ATSM) accumulate in hypoxic
xenografts in mice.
More recently, Donnelly and co-
workers have introduced a pendent carboxylate functional
group to (NH ) Sar, yielding Sar-CO H (3, Figure 1), via a
2
2
2
2+
ring-opening reaction between [Cu(II)(NH ) Sar] and
glycolic anhydride. The Cu-3-rituximab immunobioconju-
2
2
13
64
6
,7
tissue. Additionally, bioreduction may be responsible for the
8
,9
gate was prepared, and this compound showed good in vivo
loss of radioactive copper from other ligand systems.
14
To overcome instability issues, ligands that form 64Cu(II)
complexes with high thermodynamic stability and kinetic
inertness are required. Hexaazamacrobicyclic cage ligands
plasma stability.
Received: September 26, 2013
Published: December 16, 2013
(
sarcophagines) are a promising ligand class for 64Cu(II)-
©
2013 American Chemical Society
468
dx.doi.org/10.1021/ic4024508 | Inorg. Chem. 2014, 53, 468−477

Inorganic Chemistry
Article
Figure 1. Structures of macrobicyclic chelators (1−3) and diamide/diamine tetraaza-macrocycles (4 and 5) used to form complexes with copper-64.
Deprotonated amide (amidate or amido) groups are strong
σ-donors and are well-known to stabilize metal ions in higher
oxidation states. Margerum and co-workers conducted ground
breaking early studies on this phenomenon and showed that
Cu(II) and Ni(II) complexes of peptides could be readily
ligand, and the complex showed no evidence of instability when
subjected to ligand challenge experiments with the metal
binding amino acid L-cysteine, however, incubation with L-
histidine resulted in low levels of decomplexation.
15−18
oxidized to their Cu(III) and Ni(III) counterparts.
More
RESULTS AND DISCUSSION
■
recently, Collins and co-workers have developed a range of
acyclic and macrocyclic amide containing ligands for the
stabilization of high oxidation state metal centers for use as
Synthesis. The triamidetriamine cryptand ligands (7 and 8)
were prepared by the reaction of the tripodal α-chloroamide
precursor (6) with either 1,1,1-tris(aminoethyl)ethane (tame)
or 1,4,7-triazacyclononane (tacn) respectively. The tris(α-
chloroamide) synthetic intermediate (6) was synthesized by
treating tame with 3 equiv of 2-chloroacetyl chloride (Scheme
19−23
green oxidation catalysts.
Amido donors have also been
successfully utilized in radiopharmaceutical development. For
example, 99mTc-Depreotide (Neotect), a clinical radiopharma-
ceutical used for imaging somatostatin receptor positive
2
4
tumors, displays N S chelation, consisting of amido (2),
3
25
Scheme 1. Synthesis of the tris(α-Chloroamide) Precursor 6
and the Cryptands 7 and 8 Utilizing 6 and the Triamines
thiolate, and amine donor groups. The related ligand system,
mercaptoacetyltriglycine (MAG3) in combination with 99mTc,
1
,1,1-tris(Aminoethyl)ethane (tame) or 1,4,7-
is an approved radiopharmaceutical used for the evaluation of
2
6
Triazacyclononane (tacn) (S3 and S6 Respectively,
Supporting Information)
renal transplant blood flow and function.
6
4
The synthesis and Cu radiolabeling studies of the
diamidediamine macrocycle 4 (Figure 1) has been re-
2
7,28
ported.
The Cu(II) complex of this ligand has a highly
negative reduction potential (−1.84 V vs SCE) and as such is
expected to be resistant to reduction in vivo, and
6
4
biodistribution studies for the Cu labeled complex showed
rapid clearance of radioactivity from the blood suggesting good
complex stability. A series of six aliphatic diamidediamine
macrocycles with different chelate ring sizes (e.g., 5, Figure 1)
6
4
29
have been radiolabeled with Cu. These studies showed
significant variation in the labeling efficiency for the different
6
4
macrocycles, and only ligand 5 readily formed a Cu complex
with high purity. Biodistribution studies for this complex
29
showed rapid blood, liver and kidney clearance.
We herein report the synthesis of two novel triamidetriamine
cryptand ligands, which differ in their molecular architecture,
with one being macrobicyclic and the other macrotricyclic. The
Co(III) and Cu(II) complexes of the macrobicyclic ligand have
been prepared and the Co(III) complex displays octahedral
coordination with three amine and three deprotonated amido
nitrogen atoms coordinated to the metal atom. In contrast, the
Cu(II) center is not encapsulated by the ligand but rather
adopts a square planar coordination geometry, with two amine
and two deprotonated amido donor groups. Electrochemical
and spectroelectrochemical studies were conducted to gain
insight into the oxidative stability of the Cu(II) complex.
Copper-64 radiolabeling was carried out for the macrobicyclic
1). As expected, the C symmetric molecule gave rise to simple
H and C NMR spectra with a broad triplet signal at 7.78
3
1
13
ppm observed for the amide N−H groups. Cryptands 7 and 8
were prepared from equimolar mixtures of 6 and the triamines:
tame or tacn respectively and the mild base K CO (Scheme 1).
2
3
The reactions were carried out under high dilution in refluxing
acetonitrile, in an effort to improve the yield of the cryptand
products. In both cases, relatively long reaction times were
4
69
dx.doi.org/10.1021/ic4024508 | Inorg. Chem. 2014, 53, 468−477

Inorganic Chemistry
Article
used, as over the course of several experiments it became
apparent that increased reaction times lead to improved
product yields. Markedly different yields (11% and 47%
respectively) were obtained for cryptands 7 and 8, and this
difference appears to be the result of the different levels of
conformational freedom associated with the triamines: tame
and tacn. Here, the acyclic tame molecule in conjunction with
the highly flexible α-chloroamide precursor (6) results in a poor
yield of the desired macrobicycle 7, while the lower level of
conformational freedom associated with the cyclic tacn
molecule leads to an improved yield of the macrotricycle 8.
Attempts were made to increase product yields by the
complex was separated from charged species by passage
through both cation and anion (CM- and DEAE-Sephadex)
exchange resins. Despite several attempts, the Co(III) complex
of cryptand 8 could not be synthesized. Because of our interest
in the development of these cryptand ligands for radio-
pharmaceutical applications, we also investigated the chemistry
of 7 and 8 with Cu(II). The Cu(II) complex 10, was prepared
by adding 3 equiv of NaOH to an equimolar aqueous mixture
of 7 and CuCl
(Scheme 2). As was the case with Co(III), the
2
Cu(II) complex of 8 could not be prepared.
1
13
The H NMR and C NMR spectra of 9 are consistent with
1
the Co(III) complex possessing C symmetry. The H NMR
3
−
−
introduction of a halide anion, for example, Cl or Br (as
spectrum is characterized by two singlet signals for the
inequivalent methyl groups, while the methylene groups of
the magnetically equivalent “straps” of the macrobicycle give
rise to three sets of AB doublet pairs. The AB patterns are
+
the Bu N salts), as a templating agent for the α-chloroamide
4
precursor (6) (via hydrogen bonding with amide N−H
groups). However, this variation in the synthetic procedure
resulted in much poorer yields of the desired products, with
unidentified byproducts being generated.
consistent with C symmetry as opposed to the C3v symmetry
3
of the ligand. In interpreting these spectra, it is useful to
consider the hexaaminemacrobicyclic cage family of ligands
The number of signals observed in the H and 13C NMR
spectra for 7 and 8 (Figures S2 and S3, Supporting
Information) are consistent with these cryptands exhibiting
C (point group C ) symmetry in solution. The amine and
1
1
developed by Sargeson and co-workers. As an example, the H
NMR spectrum of [Co(III)(sepulcrate)]3+ (11, Figure 2),
shows AB doublet pairs for the cap methylene groups and more
3
3v
31
amide N−H protons of 7 resonate as broad singlet and triplet
signals at 1.86 ppm and 7.21 ppm respectively, while for 8 the
amide N−H protons resonate as a broad triplet signal at 7.92
ppm. For the macrotricycle 8 the six methylene units of the
triazacyclononane molecular component are equivalent, giving
complex A B patterns for the ethylenediamine protons.
2 2
1
3
rise to a single C signal at 53.5 ppm. In contrast, the
methylene proton environments of this molecular fragment
(
H7a and H7b, Supporting Information) are inequivalent, with
one proton being “endo” and the other “exo” with respect to
the macrocycle, and these protons resonate as a relatively
complex AA′BB′ pattern.
Figure 2. Structure of [Co(III)(sepulcrate)]3+ (11).31
With the cryptands (7 and 8) on hand, we initially focused
on synthesizing the Co(III) complexes of these ligands.
Cobalt(III) was chosen as this cation has previously been
used to form neutral, octahedral complexes with two tripodal,
triamide ligands, where the amide N−H groups were
Here the right- or left-handed “twist” (Δ or Λ) induced in
the sepulcrate by the octahedral coordination geometry of the
metal, leads to inequivalent environments for the methylene
group protons and defines the configuration (R or S) for the six
30
deprotonated and coordinated to the metal center. The
Co(III) complex 9 was prepared by aerating an aqueous
31
coordinated amine groups. Similarly for 9, (as confirmed by
the X-ray crystallographic structural studies, vide infra), the
octahedral complex is formed as a racemic mixture of Δ and Λ
stereoisomers, and the “twist” in ligand coordination geometry
results in inequivalent proton environments for the methylene
group protons.
mixture of 7 and CoCl for three days (Scheme 2). The neutral
2
Scheme 2. Synthesis of Co(III) (9) and Cu(II) (10)
Complexes of Cryptand 7
Based on the structure of previously reported Cu(II)
complexes of hexaaminemacrobicyclic cage ligands, for
32
example, 2, it was anticipated that the cryptate ligand 7
may form an anionic, tetragonally distorted octahedral Cu(II)
complex, with all three amide N−H groups deprotonated and
coordinated to the metal center. No evidence of an anionic
complex could be detected in the negative ion ESI-MS
spectrum of 10, indicating that Cu(II) does not form a
complex involving deprotonation of all three amide groups of
the ligand. The base peak in the positive ion ESI-MS spectrum
of 10 (Supporting Information, Figure S4) is consistent with
+
the formula [C H CuN O ] . Here two amide groups are
16
29
6
3
deprotonated and coordinated to the Cu(II) center but the
third remains protonated and likely does not bind to the metal.
Consistent with this result, the most likely coordination
geometry is square-planar, where two of the secondary amine
groups also bind to the metal and the third amine is protonated
4
70
dx.doi.org/10.1021/ic4024508 | Inorg. Chem. 2014, 53, 468−477

Inorganic Chemistry
Article
yielding an ammonium group and the overall cationic charge on
the molecule.
complexes of diamidediamine macrocycles, for example, 4
2
8,33
and 5.
In particular, these square-planar complexes
The electronic absorption spectrum of 10 (Figure 3) was
recorded to further probe the coordination mode of this Cu(II)
consistently display a weak band in the visible region (∼500
nm, ε = ∼100 M− cm ). Density functional theory (DFT)
calculations for the Cu(II) complex of 4 indicate that this weak
absorption band results from the redistribution of electron
1
−1
28
density between metal centered d orbitals. Based on the mass
spectrometric and electronic absorption results, it appears likely
9
that the Jahn−Teller distortion associated with the d electron
configuration of Cu(II) causes the metal to adopt a square-
planar coordination geometry, with two amido and two amine
donors.
To investigate differences in the interaction of cryptands 7
and 8 with CuCl , equimolar aqueous solutions of the chosen
2
cryptand and CuCl were mixed and UV−visible spectra were
2
recorded at 5 and 60 min after mixing. The spectra (Figures S5
2
+
and S6 respectively, Supporting Information) show that Cu
complexation by cryptand 7 was rapid with the 248 and 516 nm
absorption bands evident at 5 min. Absorption bands arising at
∼
250 nm have been previously assigned to a charge-transfer
2+
transition associated with Cu coordinated to deprotonated
34
Figure 3. Ultraviolet−visible spectrum of an aqueous solution of 10
amide nitrogen atoms. In contrast the only change observed
for cryptand 8 after mixing was a red shift in the high energy
n→π* transition of the ligand amide group, which is likely to be
associated with a weak interaction between the metal and the
carbonyl oxygen. These results support the notion that for
−
3
(
1.0 × 10 M).
complex. The spectrum is characterized by a moderately
−
1
intense absorption band centered at 248 nm (ε = 4207 M
−1
−1
2+
cm ) and a broad, weak band centered at 516 nm (ε = 97 M
cryptand 8, Cu -amine coordination does not occur and as a
−1
cm ). The electronic absorption spectrum of 10 closely
result amide group deprotonation and coordination is
resembles those of a number of square-planar Cu(II)
unfavorable. In comparison to 7, ligand 8 is a relatively rigid
Table 1. Crystal Refinement Data
6
7
8
9
10
empirical formula
formula weight
temperature/K
crystal system
space group
a/Å
C H N O Cl
C H N O
C H N O
C17.5 H CoN O
C H CuN O
4
1
1
18
3
3
3
16 30
6
3
17 30
6
3
33
6
4.5
17 32
6
346.63
173
354.46
366.47
458.43
173
448.03
173
173
173
monoclinic
Pc
monoclinic
monoclinic
monoclinic
monoclinic
P2 /c
1
P2 /c
1
P2 /n
1
P2 /c
1
7.9147(2)
11.1818(2)
9.5726(3)
113.626(3)
776.17(3)
2
15.9861(2)
14.9331(2)
15.0549(2)
10.337(2)
15.943(3)
10.759(2)
94.84(3)
1766.9(6)
4
15.7221(7)
7.8240(3)
17.6366(7)
106.127(4)
2084.10(15)
4
11.2087(6)
8.4478(5)
b/Å
c/Å
22.7388(11)
114.940(4)
1952.33(18)
4
β/deg
volume/Å3
97.8150(10)
3560.56(8)
8
Z
3
ρ(calc) mg/mm
1.483
1.322
1.378
1.461
1.524
μ/mm−
1
5.45
0.766
0.097
0.863
1.156
Crystal size/mm3
0.15 × 0.12 × 0.1
Cu−Kα 1.54184
7.9 to 130.04°
−6 ≤ h ≤ 9
0.15 × 0.08 × 0.02
Cu−Kα 1.54184
5.58 to 147.84°
−19 ≤ h ≤ 19
−18 ≤ k ≤ 16
−17 ≤ l ≤ 18
14408
0.12 × 0.12 × 0.05
Mo−Kα 0.71073
5.838 to 54.204°
−13 ≤ h ≤ 13
−20 ≤ k ≤ 20
−13 ≤ l ≤ 13
30912
0.15 × 0.1 × 0.1
Mo−Kα 0.71073
4.8 to 58.8°
−18 ≤ h ≤ 20
−6 ≤ k ≤ 10
−22 ≤ l ≤ 24
12100
0.3 × 0.12 × 0.03
Mo−Kα 0.71073
6.04 to 52.74°
−9 ≤ h ≤ 14
−10 ≤ k ≤ 10
−28 ≤ l ≤ 28
12047
wavelength/Å
2θ range for data collection
index ranges
−
−
11 ≤ k ≤ 13
11 ≤ l ≤ 11
reflections collected
2449
independent reflections
data/restraints/parameters
goodness-of-fit on F2
1604 [R(int) = 0.0202] 7014 [R(int) = 0.0255] 3896 [R(int) = 0.0262] 4966 [R(int) = 0.039] 3993 [R(int) = 0.0341]
1604/2/182
1.1
7014/7/479
0.861
3896/0/236
1.046
4966/1/274
1.023
3993/2/261
1.042
final R indexes [I ≥ 2σ (I)] R1 = 0.0400
R1 = 0.0413
R1 = 0.0352
R1 = 0.0471
R1 = 0.0368
wR = 0.1014
2
wR = 0.1136
2
wR = 0.0873
2
wR = 0.1010
2
wR = 0.0886
2
final R indexes [all data]
R1 = 0.0401
R1 = 0.0518
R1 = 0.0403
R1 = 0.0783
R1 = 0.0467
wR = 0.1018
wR = 0.1250
wR = 0.0908
wR = 0.1161
wR = 0.0936
2
2
2
2
2
largest diff. peak/hole/e Å−
3
0.37/−0.42
0.32/−0.28
0.31/−0.23
0.60/−0.47
0.93/−0.52
Flack parameter
0.017(19)
4
71
dx.doi.org/10.1021/ic4024508 | Inorg. Chem. 2014, 53, 468−477

Inorganic Chemistry
Article
Table 2. Selected Bond Distances (Å) from the X-ray Crystal Structures of 6−10
6
7
8
9 (M = Co(III))
10 (M =Cu(II))
N2A−C3A
N2B−C3B
N2C−C3C
N2A−M
N2B−M
N2C−M
N5A−M
N5B−M
1.338(5)
1.318(5)
1.330(5)
1.347(2)
1.328(2)
1.355(2)
1.342(2)
1.335(2)
1.342(2)
1.308(3)
1.306(3)
1.315(4)
1.890(2)
1.900(2)
1.898(2)
1.965(2)
1.967(2)
1.954(2)
1.342(3)
1.309(3)
1.318(3)
1.928(2)
1.934(2)
2.403(2)
2.028(2)
2.003(2)
N5C−M
35
Figure 4. ORTEP structures of (a) 7 (Z′ = 2, the second conformational isomeric form of the cryptand has been omitted for clarity) and (b) 8
N−H hydrogen atoms shown). Thermal ellipsoids are shown at 50% probability levels.
(
35
Figure 5. ORTEP structures of (a) 9 (two cocrystallized methanol molecules omitted for clarity) and (b) 10 (one cocrystallized methanol
molecule omitted for clarity) (N−H hydrogen atoms shown). Thermal ellipsoids are shown at 50% probability levels.
molecule, and the X-ray structure (vide infra) shows that the
amine lone pairs are endocyclic (endohedral) and may not be
accessible to the metal center.
(N2B−O3B = 2.869(5) Å, symmetry operator x, 1−y, 1/2+z)
hydrogen bonds present. The cryptands 7 and 8 both crystallize
in the monoclinic space group P2 /c, and the structures of these
1
Structural Studies. The crystal structures of compounds
molecules are shown in Figure 4, panels a and b, respectively. In
the case of 7, two independent conformational isomeric forms
of the cryptand are present in the asymmetric unit. As expected,
all amide groups for 7 and 8 adopt the trans configuration, and
extensive amide and amine group hydrogen bonding
interactions are present.
6
−10 were determined, and the crystallographic refinement
data are given in Table 1 and selected bond distances in Table
. The tripodal precursor 6 crystallized in the chiral space group
Pc and the structure of this compound is shown in Figure S7
Supporting Information). Although compound 6 is achiral, the
2
(
conformational isomeric form of 6 found in the solid state is
chiral. The structure of 6 is strongly influenced by amide group
N−H···O hydrogen bonding, with both intra- (N2A−O3B =
The Co(III) complex, 9 crystallizes in the monoclinic space
group P2 /n, and the structure is shown in Figure 5a. The
1
Co(III) center sits in a relatively undistorted octahedral
coordination geometry with three facial sites occupied by the
2.942(4) Å and N2C−O3A = 3.000(6) Å) and intermolecular
4
72
dx.doi.org/10.1021/ic4024508 | Inorg. Chem. 2014, 53, 468−477

Inorganic Chemistry
Article
deprotonated amido nitrogen atoms and the other three by the
amine nitrogen atoms. The complex possesses four chiral
centers, these being the Co(III) center (Δ and Λ) and the
three coordinated amine groups (R and S). The chirality of the
amine groups is defined by the Co(III) center, where if Co(III)
=
(
Δ the coordinated amine groups adopt an S configuration
alternatively, Co(III) = Λ, amine = R). The three five-
membered chelate rings are nonplanar and adopt an ob
obverse) conformation with respect to the C axis through the
(
3
metal center, yielding an overall ob molecular configuration.
3
The ob configuration is the same as that reported previously
3
for the Co(III) complex of an acyclic tripodal, triamidetriamine
30
ligand [Co(2,2′,2″-trioxosen-3H)]. The average of the Co−
N
amido
distances (1.896(2) Å) is significantly shorter than that
of the Co−N
bond lengths (1.962(2) Å). This difference is
expected, as anionic amido groups are strong σ-donors and
exert a marked trans-influence, leading to a lengthening of the
amine
Figure 6. (a) Cyclic voltammetric responses at varying scan rate for a
1
.0 mM aqueous solution of 10 containing 0.1 M LiClO as supporting
Co−N
bonds. Interestingly the Co−N
bond lengths for
4
amine
amine
−1/2
electrolyte. The current axis has been normalized by a factor of ν
9
are shorter than those reported previously for [Co(2,2′,2″-
(where ν is scan rate) to highlight kinetic effects. (b) Changes in the
trioxosen-3H)], where the average of the three Co−N
amine
30
UV−visible absorbance spectrum during electrolysis of 1.0 mM
solution of 10 in water containing 0.1 M LiClO4 as supporting
electrolyte. The potential of the platinum gauze working electrode was
held at a constant value of 1.0 V vs Ag/AgCl during the course of the
oxidation.
bond lengths was 1.996(4) Å. This shortening of the Co−
Namine distances most likely arises as a result of the constraints
imposed on these bonds by the relatively inflexible macro-
bicyclic ligand. Such a notion is supported by comparison to
structurally related Co(III) complexes of hexaaminemacrobicy-
clic cage complexes, for example, the average of the Co−N
amine
36
redox couple. The redox couple is electrochemically and
chemically quasi-reversible at scan rates of approximately 0.2 V
5
+
bond lengths for [Co{(NMe ) -Sar}] was 1.961(6) Å. To
3
2
−1
the best of our knowledge there is only one other example of an
amide bearing sarcophagine cryptand and for the Co(III)
s
and above, with the ratio of oxidative and reductive peak
currents (ip,ox/ip,red) being close to unity and the peak-to-peak
−
1
complex of this monoamide molecule, [Co(III)(Me,CO H-
splitting (ΔE ) being 90 mV at 0.2 V s . However the
2
p
2
+
37
oxosar−H] , the Co−N
distance was 1.917(4) Å.
response becomes progressively less chemically reversible as the
scan rate is decreased, until the return wave disappears entirely
at about 0.01 V s . These observations are consistent with an
EC type mechanism where the complex undergoes a chemical
transformation on oxidation. That this homogeneous reaction
can be out-run at moderate voltammetric time scales suggests
that the kinetics for this process are relatively slow.
amido
The Cu(II) complex, 10 crystallizes in the monoclinic space
−1
group P2 /c, and the structure is shown in Figure 5b. The X-ray
1
structure confirms the results of the mass spectrometric and
UV−visible studies and shows that the Cu(II) center sits in a
distorted square planar geometry, with coordination from two
deprotonated amido nitrogen atoms and two amine nitrogen
donors. As was the case for the Co(III) complex 9, the average
To probe the nature of the oxidized species, spectroelec-
trochemical measurements were performed, and the changes in
the visible absorption spectrum while electrochemically
oxidizing a 1.0 mM solution of 10 are shown in Figure 6b.
The potential was maintained at 1.0 V versus the Ag/AgCl
reference electrode, and a spectrum acquired every 10 s until no
further spectroscopic changes were observed. A general increase
in absorption is observed with an oxidizing potential applied,
with new absorbance bands appearing at ∼320 and 415 nm
accompanied by a visual color change from pale purple to
green. Such changes are consistent with those observed
previously for Cu(III) complexes of ligands bearing deproto-
of the Cu−N
distances (1.931(2) Å) is significantly less
amido
than that of the Cu−N
bond lengths (2.016(2) Å). Similar
amine
or slightly shorter distances have been reported previously for
two related square planar Cu(II) complexes of aliphatic
diamidediamine macrocyclic ligands, these being 1.948(2) Å
3
8
39
and 2.019(2) Å and 1.887(3) Å and 1.991(2) Å for the
bond lengths, respectively. Exami-
Cu−N
and Cu−N
amido
amine
nation of the structure shows a weak axial interaction between
the Cu(II) atom and the third amine group of the
macrobicyclic molecule, with this Cu−N
distance being
amine
2
.403(2) Å. The square-planar coordination geometry is
40
41
distorted with the Cu(II) center being displaced above the
least-squares plane defined by the four nitrogen atoms (N2B,
N2C, N5B, and N5C) by 0.261(1) Å toward the third axial
amine group. The two amine nitrogen atoms coordinated to
the Cu(II) center (N5B and N5C) are chiral with one of these
groups adopting an R and the other an S configuration, and as
such 10 can be described as a meso compound.
Electrochemistry and Spectroelectrochemistry. The
electrochemical properties of 10 were studied using cyclic
voltammetry, and Figure 6a shows the cyclic voltammetric
response for an aqueous solution of 10 (1 mM) containing 0.1
nated amide donors, for example, biuret, tetraglycine, and
28
the diamide macrocycle 4. The observed spectral changes due
to the oxidative process are only partially reversible, and the
spectrum for the original Cu(II) complex was not fully
regenerated through bulk electrolysis when a potential more
negative than the oxidation peak was applied (results not
shown). This is consistent with the voltammetric response
presented in Figure 6a, which is also irreversible at long
experimental time scales (slow scan rates). The absence of an
isosbestic point also supports the notion of an EC type
mechanism accompanying the oxidation process.
M LiClO as supporting electrolyte. When scanned anodically,
Copper-64 Labeling Studies. Cryptand 7 was radio-
labeled with Cu(II) by heating a solution of 7 (0.1 mg) and
4
64
the complex exhibits a one-electron oxidation process centered
at 0.75 V vs Ag/AgCl, which can be assigned to the Cu2+/3+
64
CuCl (23.1 MBq) at 86 °C for 1 h in ammonium acetate
2
4
73
dx.doi.org/10.1021/ic4024508 | Inorg. Chem. 2014, 53, 468−477

Inorganic Chemistry
Article
buffer (0.4 M, pH 5.5) to give 64Cu-10 in a decay corrected
yield of 75.7% after HPLC isolation and mobile phase
evaporation, which took 93 min including the 1 h reaction
time. The yield based on integration of the activity channel
chromatogram was somewhat higher indicating that 84% of the
64Cu,10−14,42 and the simple yet versatile synthetic approach
described here represents a novel strategy for the synthesis of
cage ligands. Previous attempts to prepare triamidetriamine
cryptands by capping the acyclic [Co(2,2′,2″-trioxosen-3H)]
precursor with methanal and nitromethane were unsuccessful,
presumably because of the electronic influence of the
6
4
radioactivity was associated with the Cu-10 peak; this
discrepancy is understandable on account of the losses
associated with collection of the radiolabeled compound and
geometric and attenuation considerations in the dose calibrator.
The identity of the peak was confirmed by coinjection of a
30
deprotonated amido groups. A diamide containing sarcoph-
agine has been proposed as a reaction intermediate formed
from 3,6,10,13,16,19-hexaazabicyclo[6.6.6]eicosane and dime-
43
thyloxalate; however, this compound was not isolated.
6
4
Cu-10 and Cu-10 mixture (Figure 7), where the peak maxima
Cryptand 7 was metalated with Co(III) or Cu(II). In contrast,
and despite numerous attempts, cryptand 8 could not be
metalated with these metal ions. The low metal-binding affinity
of 8 was unexpected as tacn and tacn based ligands generally
bind as facial tridentates, and visual examination of the crystal
structure of 8 suggests a significant degree of preorganization of
the tacn molecular component for metalation as a facial
tridentate. In addition, the structurally related complex
[
Co(III)(nosartacn)]3+ (12, Figure 8) of a hexaaminemacro-
Figure 7. Radioactivity (a) and absorbance (254 nm) (b) HPLC
chromatograms obtained from the coinjection of a mixture of 64Cu-10
and Cu-10. The peak maxima for the radioactivity channel and the 254
nm absorbance channel are coincident (allowing for the arrangement
of the radioactivity and UV detection systems).
Figure 8. Structure of [Co(III)(nosartacn)]3+ (12).44
44
tricyclic ligand has been reported previously. In contrast to
the present work however, 12 was prepared by capping the
3
+
trigonal face of the precursor complex [Co(taetacn)] , as
opposed to direct metalation of the free cryptand ligand.
44
of the radioactivity channel and the 254 nm absorbance
6
4
coincided. Earlier attempts to label 7 with Cu were
unsuccessful on account of trace impurities in the ligand,
prior to HPLC purification. To evaluate the stability of 64Cu-
A possible explanation for our inability to metalate 8 may be
associated with the lower flexibility of this ligand because of the
introduction of the third macrocyclic ring. This change is likely
to lower the conformational freedom of 8 when compared to 7
and may decrease the capacity of the ligand to accommodate an
octahedral metal ion and instead impose the less favorable
trigonal prismatic coordination geometry. Additionally, the X-
ray structure of 8 shows that the amine nitrogen atom lone
pairs are endocyclic (endohedral) and may not be accessible to
the metal center.
1
0, ligand challenge experiments were conducted using the
well-known metal binding amino acids: L-histidine and L-
cysteine. For these experiments, samples of Cu-10 (∼1.3
MBq) reconstituted in phosphate buffered saline (PBS) pH =
6
4
7
.4, were incubated at 37 °C for 24 h with either no addition
−3
(
control) or with L-cysteine or L-histidine (1 × 10 M). Radio-
HPLC analysis Figures S8−S10 (Supporting Information)
showed evidence for trans-chelation in the presence of L-
histidine over the course of the experiment with the
radiochemical purity being 93% and 70.2% after 6 and 24 h
respectively. The L-cysteine challenge on the other hand
resulted in 64Cu-10 of a radiochemical purity of greater than
The X-ray crystal structure for the Co(III) complex 9
1
13
support the results of the H and C NMR studies, which
show that the complex possesses C symmetry. The structure of
3
this complex is closely related to those of Sargeson’s
hexaaminemacrobicyclic cage family of ligands, with the three
five-membered chelate rings adopting an ob conformation with
9
8% after the 24 h period, which was very similar to the PBS
6
4
control. These results show that Cu-10 displays good stability
respect to the C axis. Similarly the X-ray crystal structure of the
3
in the presence of very high concentrations of competing
Cu(II) complex 10 confirms the results of the mass
spectrometric and UV−visible studies and shows that the
deprotonated amido nitrogen donors favor a tetragonally
distorted square-planar coordination geometry for the Cu(II)
center.
6
4
ligands and that trans-chelation of Cu is more facile in the
presence of N-donor amino acid, L-histidine.
CONCLUSION
The first examples of triamidetriamine cryptand ligands (7 and
) were successfully prepared from the tris(α-chloroamide)
■
8
Cyclic voltammetric studies of 10 show an oxidative response
with E° = 0.75 V (versus Ag/AgCl). This process is reversible
at most scan rates but becomes irreversible at scan rates of
synthetic intermediate 6 and the triamines: tame or tacn,
respectively. Encapsulating ligands are of significant current
interest for radiopharmaceutical applications utilizing
−1
about 0.01 V s , suggesting a kinetically slow chemical reaction
following oxidation. Spectroelectrochemical experiments con-
4
74
dx.doi.org/10.1021/ic4024508 | Inorg. Chem. 2014, 53, 468−477

Inorganic Chemistry
Article
firm the formation of a new product when electrolysis is carried
out for relatively long periods of time. Although the nature of
the decomposition product formed on oxidation is unknown,
the results are consistent with a scenario where the metal center
withdraws electron density from one of the metal ligand bonds
on oxidation, resulting in cleavage of that bond and formation
of an aquo complex. Because of our interest in the development
of these cryptand ligands for radiopharmaceutical applications,
Electrochemical and Spectroelectrochemical Studies. All
electrochemical experiments were conducted at room temperature in
aqueous 0.1 M LiClO4 which was purged with argon prior to
experiments. Cyclic voltammetry was performed using a CHI660E
potentiostat (CH Instruments, Austin, TX, U.S.A.). The electro-
chemical cell was a conventional three-electrode configuration,
2
consisting of a glassy carbon 3 mm (0.07068 cm ) working electrode
shrouded in Teflon (CH Instruments), a 1 cm2 platinum gauze
auxiliary electrode, and an aqueous Ag/AgCl (3 M KCl) reference
electrode (CH Instruments). Spectro-electrochemical measurements
were performed using a Varian Cary UV−visible spectrometer and a
CH660B potentiostat (CH Instruments) with a 1 mm path length thin
layer quartz spectroelectrochemical cell (ALS, Tokyo, Japan). A fine
platinum gauze working electrode, platinum wire counter electrode,
6
4
64
ligand 7 was radiolabeled with Cu. Although, the Cu(II) ion
was not encapsulated by 7 in a manner analogous to hexaamine
3
2,45 64
sarcophagine ligands,
Cu-10 showed no evidence of
decomposition when incubated with the amino acid L-cysteine
6
4
and moderate levels of Cu transchelation in the presence of
the N-donating L-histidine. We are currently investigating the
and Ag/AgCl (3 M KCl) reference electrode were employed.
Radiochemistry. All radiochemistry reactions involving 64Cu were
6
8
chemistry of cryptand 7 with Ga(III) as Ga has excellent
carried out in suitably lead shielded fume hoods in plastic-ware that
had been soaked in 4 M HCl overnight followed by washing with
+
+
properties for PET imaging (β = 89%, E
= 1.92 MeV, t1/2
β max
=
68 min), and the results of these studies will be reported in
deionized H O (18.2 MΩ) and drying. Radioactivity measurements
2
due course.
were performed on a CAPINTEC CRC-15R Dose Calibrator with
6
4
calibration factor set to #035. Purification of the Cu-10 radio-
2
complex was performed on a Waters Empower controlled HPLC with
EXPERIMENTAL SECTION
■
a Waters 600E pump, a Rheodyne manual injection port, Waters 2998
General Procedures. Unless otherwise stated, all manipulations
were performed under a nitrogen atmosphere. All chemicals and
reagents were purchased from commercial sources (Sigma Aldrich or
Alfa Aesar) and were of analytical grade or higher, and were used as
received. Anhydrous tetrahydrofuran was distilled from potassium
Photodiode Array, and a Carroll & Ramsay Associates 105S γ-detector
6
4
with a sodium iodide crystal. Stability analysis of the Cu-10 radio-
2
complex was performed on a Waters Empower controlled HPLC with
a Waters 600E pump, a Rheodyne manual injection port, a Waters
2489 Dual channel UV/visible detector, and a LabLogic IN/US
Systems γ-RAM radio-HPLC detector.
under nitrogen prior to use. Anhydrous acetonitrile was distilled from
́
64
CaH under nitrogen, and stored over molecular sieves (4 Å, activated
HPLC Purification of 7 for Cu Radiolabeling. Purification of
2
2
at 250 °C) prior to use. Potassium carbonate was dried in a furnace at
the macrobicyclic ligand 7 was performed on a Waters Empower
2
00 °C for 12 h prior to use. NMR spectra were recorded on a Bruker
controlled Delta Prep HPLC including a Prep LC Controller, 717
Autosampler, Prep Degasser, and a 486 Tunable Absorbance Detector.
Compound 7 (5 mg) was dissolved in the minimum volume of 1:1
acetonitrile: water and aliquots were injected into the HPLC system
using an Atlantis PrepT3 10 μm OBD C-18 column (19 × 150 mm)
running a gradient with the following conditions; 0−7 min, 20 mL/
min, water (0.1% formic acid); 7−7.1 min 20−30 mL/min water
(0.1% formic acid); 7.1−8 min 0−2% acetonitrile in water (0.1%
formic acid) with monitoring absorbance at 210 nm. The peak at 7−8
min was collected in acid-washed glassware and evaporated to dryness
(3.6 mg) under high vacuum and reconstituted in water (0.36 mL) for
radiolabeling experiments.
1
13
Avance ARX-300 (300.14 MHz for H, 75.48 MHz for C) and
1
13
Bruker Avance ARX-500 (500.13 MHz for H, 125.77 MHz for C)
spectrometer and were internally referenced to solvent resonances.
Mass spectra were obtained using a Bruker Esquire6000 mass
spectrometer fitted with an Agilent electrospray (ESI) ion source.
UV−visible spectra were recorded using an Agilent Technologies Cary
3
00 UV−visible spectrophotometer using quartz cuvettes (1 cm). The
46−48 49,50
triamines tame
and tacn
were prepared using modified
literature procedures (Supporting Information). The compound
1
13
numbering scheme used for H and C NMR assignments are given
in the Supporting Information.
64Cu-10 Radiolabeling. To a 0.5 mL microcentrifuge tube was
added an aqueous solution of macrobicyclic ligand 7 (10 μL of a 10
X-ray Crystallography. Single crystals suitable for X-ray
crystallography were grown as follows: the diffusion of vapors between
hexane and a chloroform solution of the title compound (6); the
diffusion of vapors between diethyl ether and chloroform solutions of
the title compounds (7 and 8) or methanol solutions of the title
compounds (9 and 10). Crystallographic data for all structures
determined are given in Tables 1 and 2. For all samples, crystals were
removed from the crystallization vial and immediately coated with
Paratone oil on a glass slide. A suitable crystal was mounted in
Paratone oil on a glass fiber and cooled rapidly to 173 K in a stream of
cold N2 using an Oxford Instruments low temperature device.
Diffraction data were measured using an Oxford Gemini diffractometer
mounted with Mo−Kα (λ = 0.71073 Å) or Cu−Kα (λ = 1.54184 Å)
radiation. Data were reduced and corrected for absorption using the
mg/mL solution; 100 μg), aqueous NH OAc (0.4 M, pH 5.5, 50 μL)
4
and 64CuCl in 0.02 M HCl, (20 μL, 23.1 MBq). The vial was then
2
capped and the mixture heated at 85 °C for 1 h. The reaction mixture
was injected into a preparative radio-HPLC system using an Atlantis
Prep T3 5 μm C-18 column (10 × 250 mm) running a gradient with
the following conditions; 0−10 min, 2.5 mL/min 5% acetonitrile in
water (0.1% formic acid); 10−11 min, 5−50% acetonitrile in water
(0.1% formic acid) and the radio-peak at 7.46 min was collected, and
evaporated to dryness under high vacuum, yielding 16.07 MBq, 75.7%
decay corrected yield from 93 min operation.
64
Cu-10 Stability Studies. In 0.5 mL microcentrifuge tubes 64Cu-
10 (1.3 MBq) was incubated in phosphate buffered saline (PBS) (pH
51
−3
CrysAlis Pro program. The structures were solved using Direct
= 7.4, 222 μL) or in PBS with either L-cysteine or L-histidine (1 × 10
Methods and refined by the Full-Matrix Least-Squares refinement
M, pH = 7.4, final volume = 222 μL) for 24 h at 37 °C. Stability was
determined at 6 and 24 h time-points by injections on a radio-HPLC
system and comparing the radiochemical purity.
2
52
techniques on F using SHELXL97. The non-hydrogen atoms were
refined anisotropically and hydrogen atoms were placed geometrically
and refined using the riding model. Coordinates and anisotropic
thermal parameters of all non-hydrogen atoms were refined. All
Synthesis. 6. To a solution of tame (S3) (2.00 g, 0.017 mol) in
chloroform (50 mL) under an atmosphere of N , was added an
2
2
53
calculations were carried out using the program Olex . Images were
aqueous solution of K CO (1.0 M, 50 mL). The mixture was cooled
2
3
35
generated by using ORTEP-3. Further X-ray diffraction (XRD)
to 0 °C and, with vigorous stirring, a solution of 2-chloroacetyl
chloride (4.06 mL, 0.051 mol) in chloroform (50 mL) was added
dropwise. After 2 h, the mixture was warmed to room temperature
(RT), and the organic phase was separated and washed with water (2
details are provided in the Supporting Information. CCDC 961215−
961219 contains the supplementary crystallographic data for this
paper. These data can be obtained free of charge from The Cambridge
Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/
cif
× 20 mL) and then dried with MgSO . The solvent was removed on
4
the rotary evaporator, yielding the crude product as a pale yellow solid.
4
75
dx.doi.org/10.1021/ic4024508 | Inorg. Chem. 2014, 53, 468−477

Inorganic Chemistry
Article
The crude product was purified on silica using an eluent composed of
C H N O Cu = 416.15. Found: C, 45.91; H, 7.00; N, 20.40%.
C H N O Cu requires C, 46.20; H, 6.78; N, 20.20%.
16 28 6 3
16
29
6
3
4
1
:1 chloroform:methanol +25% aqueous ammonia solution (20 drops/
00 mL), affording 6 as an off-white solid. Yield: 4.44 g, (75.1%). H
1
NMR (500 MHz, CDCl ): δ 0.86 (s, 3H, CH , [H1]), 3.049 (d, 6H,
3
3
ASSOCIATED CONTENT
■
* Supporting Information
3
3
JHH = 6.5 Hz, CH , [H3]), 4.08 (s, 6H, CH , [H6]), 7.78 (t, 3H, J
2
2
HH
13
S
=
6.5 Hz, CONH, [H4]) ppm. C NMR (CDCl ): δ 19.1 (C1), 41.6
3
Synthetic details for the preparation of the triamines: tame and
(
(
C2), 42.8 (C6), 43.2 (C3), 168.2 (C5) ppm. ESI-MS: m/z = 346.1
1
13
+
H ) calcd. For C H N O Cl = 346.04. Found: C, 38.07; H, 5.34;
tacn. H and C NMR spectra for 7 and 8 and compound
11
19
3
3
3
1
13
N, 12.25%. C H Cl N O requires C, 38.11; H, 5.23; N, 12.12%.
numbering scheme used for H and C NMR assignment of
11
18
3
3
3
7
. To a solution of 6 (1.07 g, 3.07 mmol) and tame (S3) (0.36 g,
.07 mmol) in anhydrous acetonitrile (500 mL) under an atmosphere
of N was added K CO (2.54 g, 18.4 mmol). The resulting solution
6−9. Positive-ion ESI-mass spectrum for the Cu(II) complex
3
1
0 and UV−vis spectra for the reaction of cryptands 7 and 8
2
2
3
with CuCl . Additional X-ray crystallographic details for 6−10
and the ORTEP structure of compound 6. Radio-HPLC
2
was heated at 110 °C for 11 days. The solution was filtered while hot,
and the solvent was removed on the rotary evaporator to afford a
colorless oil. The crude product was purified on silica using an eluent
composed of 4:1 chloroform: methanol +25% aqueous ammonia
64
chromatograms obtained at 6 and 24 h for a sample of Cu-10
incubated for 24 h in phosphate buffered saline (PBS, pH 7.4)
and PBS with L-histidine and L-cysteine. This material is
available free of charge via the Internet at http://pubs.acs.org.
solution (20 drops/100 mL), affording 7 as a colorless oil that slowly
1
crystallized. Yield: 0.122 g, (11.2%). H NMR (500 MHz, CDCl ): δ
3
0
.86 (s, 3H, CH , [H1]), 0.88 (s, 3H, CH , [H10]), 1.86 (br, 3H, NH,
3
3
[
H7]), 2.62 (s, 6H, CH , [H8]), 3.25 (s, 6H, CH , [H6]), 3.43 (d, 6H,
AUTHOR INFORMATION
2
2
■
*
3
3
JHH = 7.0 Hz, CH , [H3]), 7.21 (t, 3H, J = 7.0 Hz, CONH, [H4])
2
HH
Corresponding Author
E-mail: p.barnard@latrobe.edu.au. Fax: (+)61 3 9479 1266.
ppm. 13C NMR (CDCl ): δ 21.7 (C10), 26.1 (C1), 38.5 (C2), 39.7
3
(
C9), 49.7 (C3), 55.8 (C6), 59.3 (C8), 173.8 (C5) ppm. ESI-MS: m/z
+
=
355.2 (H ) calcd. For C H N O = 355.24. Found: C, 52.60; H,
Notes
16
31
6
3
8
2
.84; N, 22.67%. C H N O ·0.6H O requires C, 52.61; H, 8.61; N,
The authors declare no competing financial interest.
16
30
6
3
2
3.01%.
. This compound was prepared from 6 (1.09 g, 3.14 mmol), tacn
S6) (0.39 g, 3.14 mmol), and K CO (2.61 g, 18.85 mmol) using the
8
ACKNOWLEDGMENTS
■
(
2
3
We thank the Australian Research Council for ARC LIEF
funding (LE100100109). The radiochemical work was
financially supported by the Australian Institute of Nuclear
Science and Engineering (AINSE), Lucas Heights Science &
Technology Centre, New Illawarra Rd, Menai, N.S.W., 2234,
method described for 7. Compound 8 was obtained as an off-white
solid. Yield: 0.541 g, (47.0%). H NMR (500 MHz, CDCl ): δ 0.98 (s,
3
CH , [H7b]), 3.36 (s, 6H, CH , [H6]), 3.379 (d, 6H, J = 6.5 Hz,
CH , [H3]), 7.92 (t, 3H, J = 6.5 Hz, CONH, [H4]) ppm.
NMR (CDCl ): δ 23.9 (C1), 39.8 (C2), 47.6 (C3), 53.5 (C7), 62.0
1
3
H, CH , [H1]), 2.59−2.65 (m, 6H, CH , [H7a]), 2.80−2.86 (m, 6H,
3
2
3
2
2
HH
3
13
C
2
HH
64
3
Australia. Cu radioisotope was kindly provided by Charmaine
+
(
C6), 171.4 (C5) ppm. ESI-MS: m/z = 367.2 (H ) calcd. For
Jeffery of the RAPID Laboratories, Medical Technology and
Physics, Sir Charles Gairdner Hospital, Nedlands, W.A., 6009,
Australia.
C H N O = 367.24. Found: C, 49.42; H, 7.53; N, 19.42%.
C H N O ·0.5CHCl requires C, 49.32; H, 7.21; N, 19.72%.
17
31
6
3
17
30
6
3
3
9
. A stirred solution of 7 (50.0 mg, 0.141 mmol) and CoCl .6H O
2 2
(
33.6 mg, 0.141 mmol) in water (10 mL) was aerated for 3 days,
REFERENCES
during which time the color changed from red to orange. The solution
was filtered through a pad of Celite, and anionic and cationic species
were removed by passing the reaction mixture through separate CM-
and DEAE-Sephadex columns. The solvent was removed on the rotary
evaporator to give an orange solid. The solid was triturated with
chloroform (2 × 10 mL) followed by acetonitrile (3 × 10 mL), before
being redissolved in ethanol (10 mL). The latter solution was filtered,
and the solvent removed on the rotary evaporator to afford 9 as an
orange solid. Yield: 6.0 mg, (10.4%). H NMR (500 MHz, CD OD): δ
0
=
■
(
(
1) Smith, S. V. J. Inorg. Biochem. 2004, 98 (11), 1874−1901.
2) Shokeen, M.; Anderson, C. J. Acc. Chem. Res. 2009, 42 (7), 832−
8
41.
(
3) Anderson, C. J.; Ferdani, R. Cancer Biother. Radiopharm. 2009,
24, 379−393.
(4) Wadas, T. J.; Wong, E. H.; Weisman, G. R.; Anderson, C. J.
Chem. Rev. 2010, 110 (5), 2858−2902.
1
3
(5) Donnelly, P. S. Dalton Trans. 2011, 40 (5), 999−1010.
(6) Holland, J. P.; Barnard, P. J.; Collison, D.; Dilworth, J. R.; Edge,
R.; Green, J. C.; McInnes, E. J. L. Chem.Eur. J. 2008, 14 (19), 5890−
5907.
2
.86 (s, 3H, CH , [H1]), 0.93 (s, 3H, CH , [H10]), 2.33 (d, 3H, J
3
3
HH
2
14.0 Hz, CH , [H8b]), 2.37 (d, 3H, J = 12.5 Hz, CH , [H3a]),
.96 (d, 3H, J = 14.0 Hz, CH , [H8a]), 3.15 (d, 3H, J = 12.5
2 HH 2
2
2
2
HH 2 HH
2
Hz, CH , [H3b]), 3.35 (d, 3H, J = 15.5 Hz, CH , [H6b]), 3.50 (d,
3
(
(
C H N O Co = 411.15. Found: C, 46.39; H, 6.25; N, 20.59%.
C H N O Co.0.1H O requires C, 46.63; H, 6.65; N, 20.39%.
(7) Holland, J. P.; Giansiracusa, J. H.; Bell, S. G.; Wong, L.-L.;
Dilworth, J. R. Phys. Med. Biol. 2009, 54 (7), 2103.
(8) Woodin, K. S.; Heroux, K. J.; Boswell, C. A.; Wong, E. H.;
Weisman, G. R.; Niu, W.; Tomellini, S. A.; Anderson, C. J.; Zakharov,
L. N.; Rheingold, A. L. Eur. J. Inorg. Chem. 2005, 2005 (23), 4829−
4833.
(9) Heroux, K. J.; Woodin, K. S.; Tranchemontagne, D. J.; Widger, P.
C. B.; Southwick, E.; Wong, E. H.; Weisman, G. R.; Tomellini, S. A.;
Wadas, T. J.; Anderson, C. J.; Kassel, S.; Golen, J. A.; Rheingold, A. L.
Dalton Trans. 2007, 0 (21), 2150−2162.
(10) Di Bartolo, N. M.; Sargeson, A. M.; Donlevy, T. M.; Smith, S. V.
J. Chem. Soc., Dalton Trans. 2001, 0 (15), 2303−2309.
(11) Di Bartolo, N.; Sargeson, A. M.; Smith, S. V. Org. Biomol. Chem.
2006, 4 (17), 3350−3357.
(12) Voss, S. D.; Smith, S. V.; DiBartolo, N.; McIntosh, L. J.; Cyr, E.
M.; Bonab, A. A.; Dearling, J. L. J.; Carter, E. A.; Fischman, A. J.;
Treves, S. T.; Gillies, S. D.; Sargeson, A. M.; Huston, J. S.; Packard, A.
B. Proc. Natl. Acad. Sci. 2007, 104 (44), 17489−17493.
2
HH
2
2
13
H, J = 15.5 Hz, CH , [H6a]) ppm. C NMR (CH OD): δ 21.2
HH 2 3
C1), 21.9 (C10), 42.4 (C9), 43.7 (C2), 51.9 (C3), 56.4 (C8), 62.2
+
C6), 176.6 (C5) ppm. ESI-MS: m/z = 411.1 (H ) calcd. For
16
28
6
3
16
27
6
3
2
10. The pH of a solution of 7 (40.0 mg, 0.113 mmol) and
CuCl .2H O (19.2 mg, 0.113 mmol) in water (20 mL) heated to 50
2
2
°C was adjusted to pH ∼ 10 with an aqueous solution of NaOH (68
mM, 5 mL). After 1 h the solution was filtered and evaporated to
dryness to afford a purple residue. The residual solid was extracted into
ethanol (10 mL), and the resulting solution was filtered. The solvent
was then removed on a rotary evaporator to give a purple solid. To
remove residual NaCl, the ethanol extraction process was repeated 3
more times. The crude product was then recrystallized via the diffusion
of vapors between a solution of the crude product in methanol and
diethyl ether, affording 10 as a deep blue/purple crystalline solid.
+
Yield: 23.5 mg, (50%). ESI-MS: m/z = 416.2 (H ) calcd. For
4
76
dx.doi.org/10.1021/ic4024508 | Inorg. Chem. 2014, 53, 468−477

Inorganic Chemistry
Article
(
13) Ma, M. T.; Karas, J. A.; White, J. M.; Scanlon, D.; Donnelly, P. S.
Chem. Commun. 2009, 0 (22), 3237−3239.
14) Cooper, M. S.; Ma, M. T.; Sunassee, K.; Shaw, K. P.; Williams, J.
D.; Paul, R. L.; Donnelly, P. S.; Blower, P. J. Bioconjugate Chem. 2012,
3 (5), 1029−1039.
15) Margerum, D. W.; Chellappa, K. L.; Bossu, F. P.; Burce, G. L. J.
Am. Chem. Soc. 1975, 97 (23), 6894−6896.
16) Bossu, F. P.; Margerum, D. W. J. Am. Chem. Soc. 1976, 98 (13),
003−4004.
17) Kirksey, S. T.; Neubecker, T. A.; Margerum, D. W. J. Am. Chem.
Soc. 1979, 101 (6), 1631−1633.
18) Diaddario, L. L.; Robinson, W. R.; Margerum, D. W. Inorg.
Chem. 1983, 22 (7), 1021−1025.
19) Collins, T. J.; Richmond, T. G.; Santarsiero, B. D.; Treco, B. G.
R. T. J. Am. Chem. Soc. 1986, 108 (8), 2088−2090.
20) Anson, F. C.; Collins, T. J.; Richmond, T. G.; Santarsiero, B. D.;
Toth, J. E.; Treco, B. G. R. T. J. Am. Chem. Soc. 1987, 109 (10), 2974−
979.
21) Collins, T. J.; Kostka, K. L.; Munck, E.; Uffelman, E. S. J. Am.
Chem. Soc. 1990, 112 (14), 5637−5639.
22) Collins, T. J.; Kostka, K. L.; Uffelman, E. S.; Weinberger, T. L.
Inorg. Chem. 1991, 30 (22), 4204−4210.
23) Bartos, M. J.; Gordon-Wylie, S. W.; Fox, B. G.; James Wright, L.;
Weintraub, S. T.; Kauffmann, K. E.; Munck, E.; Kostka, K. L.;
(45) Ma, M. T.; Cooper, M. S.; Paul, R. L.; Shaw, K. P.; Karas, J. A.;
Scanlon, D.; White, J. M.; Blower, P. J.; Donnelly, P. S. Inorg. Chem.
2011, 50 (14), 6701−6710.
(
(46) Fleischer, E. B.; Gebala, A. E.; Levey, A.; Tasker, P. A. J. Org.
Chem. 1971, 36 (20), 3042−3044.
2
(
(47) Beaufort, L.; Delaude, L.; Noels, A. F. Tetrahedron 2007, 63
(30), 7003−7008.
(
4
(
(48) Hoveyda, H. R.; Karunaratne, V.; Nichols, C. J.; Rettig, S. J.;
Stephens, A. K. W.; Orvig, C. Can. J. Chem. 1998, 76 (4), 414−425.
(49) Madison, S. A.; Batal, D. J. Improved synthesis of 1,4,7-
triazacyclononane, U.S. Patent US5284944 A, 1992.
(50) Barta, C. A.; Bayly, S. R.; Read, P. W.; Patrick, B. O.; Thompson,
R. C.; Orvig, C. Inorg. Chem. 2008, 47 (7), 2280−2293.
(51) CrysAlisPro, 171.35.11; Oxford Diffraction Ltd.: Abingdon, U.K.,
2011.
(
(
(
(52) Sheldrick, G. M. Acta Crystallogr., Sect. A 2008, 64 (1), 112−
122.
(53) Dolomanov, O. V.; Bourhis, L. J.; Gildea, R. J.; Howard, J. A. K.;
2
(
Puschmann, H. J. Appl. Crystallogr. 2009, 42 (2), 339−341.
(
(
̈
Uffelman, E. S.; Rickard, C. E. F.; Noon, K. R.; Collins, T. J. Coord.
Chem. Rev. 1998, 174 (1), 361−390.
(
24) Liu, S. Adv. Drug Delivery Rev. 2008, 60 (12), 1347−1370.
25) Cantorias, M. V.; Howell, R. C.; Todaro, L.; Cyr, J. E.;
(
Berndorff, D.; Rogers, R. D.; Francesconi, L. C. Inorg. Chem. 2007, 46
18), 7326−7340.
26) O’malley, J. P.; Ziessman, H. A.; Chantarapitak, N. Clin. Nucl.
Med. 1993, 18 (1), 22−29.
27) Antunes, P.; Delgado, R.; Drew, M. G. B.; Fel
Inorg. Chem. 2007, 46 (8), 3144−3153.
28) Barnard, P. J.; Holland, J. P.; Bayly, S. R.; Wadas, T. J.;
Anderson, C. J.; Dilworth, J. R. Inorg. Chem. 2009, 48 (15), 7117−
126.
29) Cutler, C. S.; Wuest, M.; Anderson, C. J.; Reichert, D. E.; Sun,
Y.; Martell, A. E.; Welch, M. J. Nucl. Med. Biol. 2000, 27 (4), 375−380.
30) Angus, P. M.; Elliott, A. J.; Sargeson, A. M.; Willis, A. C. J.
(
(
(
́
ix, V.; Maecke, H.
(
7
(
(
Dalton Trans. 1999, 7, 1131−1136.
(
(
31) Sargeson, A. Chem. Br. 1979, 15 (1), 23−27.
32) Bernhardt, P. V.; Bramley, R.; Engelhardt, L. M.; Harrowfield, J.
M.; Hockless, D. C. R.; Korybut-Daszkiewicz, B. R.; Krausz, E. R.;
Morgan, T.; Sargeson, A. M. Inorg. Chem. 1995, 34 (14), 3589−3599.
(
33) Motekaitis, R. J.; Sun, Y.; Martell, A. E.; Welch, M. J. Can. J.
Chem. 1999, 77, 614−623.
(34) Sigel, H.; Martin, R. B. Chem. Rev. 1982, 82 (4), 385−426.
(35) Farrugia, L. J. Appl. Crystallogr. 1997, 30 (5 Part 1), 565.
(36) Bernhardt, P. V.; Bygott, A. M. T.; Geue, R. J.; Hendry, A. J.;
Korybut-Daszkiewicz, B. R.; Lay, P. A.; Pladziewicz, J. R.; Sargeson, A.
M.; Willis, A. C. Inorg. Chem. 1994, 33 (20), 4553−4561.
(
37) Geue, R.; Petri, W.; Sargeson, A.; Snow, M. Aust. J. Chem. 1992,
4
5 (10), 1681−1703.
38) Tang, J.-K.; Ou-Yang, Y.; Zhou, H.-B.; Li, Y.-Z.; Liao, D.-Z.;
Jiang, Z.-H.; Yan, S.-P.; Cheng, P. Cryst. Growth Des. 2005, 5 (2), 813−
19.
39) Achmatowicz, M.; Szumna, A.; Zielin
Tetrahedron 2005, 61 (38), 9031−9041.
40) Steggerda, J. J.; Bour, J. J.; Birker, P. J. M. W. L. Inorg. Chem.
971, 10 (6), 1202−1205.
41) Wong, L. F.; Cooper, J. C.; Margerum, D. W. J. Am. Chem. Soc.
976, 98 (23), 7268−7274.
42) Cai, H.; Fissekis, J.; Conti, P. S. Dalton Trans. 2009, 27, 5395−
400.
43) Korybut-Daszkiewicz, B.; Klimkiewicz, J. Chem. Commun. 1996,
0, 1175−1176.
44) Taylor, S.; Snow, M.; Hambley, T. Aust. J. Chem. 1983, 36 (12),
359−2368.
(
8
(
́
ski, T.; Jurczak, J.
(
1
(
1
(
5
(
1
(
2
4
77
dx.doi.org/10.1021/ic4024508 | Inorg. Chem. 2014, 53, 468−477