H2azapa: A versatile acyclic multifunctional chelator for 67Ga, 64Cu, 111In, and 177Lu

Article abstract of DOI:10.1021/ic302225z

Preliminary experiments with the novel acyclic triazole-containing bifunctional chelator H₂azapa and the radiometals ⁶⁴Cu, ⁶⁷Ga, ¹¹¹In, and ¹⁷⁷Lu have established its significant versatile potential a

Full text of DOI:10.1021/ic302225z

Article
pubs.acs.org/IC
67
64
H2azapa: a Versatile Acyclic Multifunctional Chelator for Ga, Cu,
177
111In, and Lu
Gwendolyn A. Bailey,†,∥ Eric W. Price,†,‡,∥ Brian M. Zeglis,⊥ Cara L. Ferreira,§ Eszter Boros,†,‡
Michael J. Lacasse,† Brian O. Patrick,† Jason S. Lewis,⊥ Michael J. Adam,*,‡ and Chris Orvig*,†
†Medicinal Inorganic Chemistry Group, Department of Chemistry, University of British Columbia, 2036 Main Mall, Vancouver,
British Columbia V6T 1Z1, Canada
‡TRIUMF, 4004 Wesbrook Mall, Vancouver, British Columbia V6T 2A3, Canada
§Nordion, 4004 Wesbrook Mall, Vancouver, British Columbia V6T 2A3, Canada
⊥Memorial Sloan-Kettering Cancer Center (MSKCC), Memorial Hospital, 1275 York Avenue, New York, New York 10065, United
States
S
* Supporting Information
ABSTRACT: Preliminary experiments with the novel acyclic triazole-containing bifunctional chelator H2azapa and the
radiometals 64Cu, 67Ga, 111In, and 177Lu have established its significant versatile potential as an alternative to 1,4,7,10-
tetraazacyclododecane-1,4,7,10-tetraacetic acid (DOTA) for metal-based radiopharmaceuticals. Unlike DOTA, H2azapa
radiolabels quantitatively with 64Cu, 67Ga, 111In, and 177Lu in 10 min at room temperature. In vitro competition experiments
with human blood serum show that 64Cu remained predominantly chelate-bound, with only 2% transchelated to serum proteins
after 20 h. Biodistribution experiments with [64Cu(azapa)] in mice reveal uptake in various organs, particularly in the liver, lungs,
heart, intestines, and kidneys. When compared to [64Cu(DOTA)]2−, the lipophilic neutral [64Cu(azapa)] was cleared through the
gastrointestinal tract and accumulated in the liver, which is common for lipophilic compounds or free 64Cu. The chelator H2azapa
is a model complex for a click-based bifunctional chelating agent, and the lipophilic benzyl “place-holders” will be replaced by
hydrophilic peptides to modulate the pharmacokinetics and direct activity away from the liver and gut. The solid-state molecular
structure of [In(azapa)(H2O)][ClO4] reveals a very rare eight-coordinate distorted square antiprismatic geometry with one
triazole arm bound, and the structure of [64Cu(azapa)] shows a distorted octahedral geometry. The present study demonstrates
significant potential for bioconjugates of H2azapa as alternatives to DOTA in copper-based radiopharmaceuticals, with the highly
modular and “clickable” molecular scaffold of H2azapa easily modified into a variety of bioconjugates. H2azapa is a versatile
addition to the “pa” family, joining the previously published H2dedpa (67/68Ga and 64Cu), H4octapa (111In, 177Lu, and 90Y), and
H5decapa (225Ac) to cover a wide range of important nuclides.
the single photon emission computed tomography (SPECT)
isotopes 111In, 67Ga, and 99mTc have practical value for
widespread clinical use. In light of recent recurrent global
shortages of 99Mo, the parent isotope of the clinically important
diagnostic nuclide 99mTc, proper utilization of these “non-
standard” isotopes is of increasing importance.3 Radiometals
INTRODUCTION
■
There is currently strong interest in the development of new
and efficient chelators that form stable complexes with the
burgeoning variety of radiometal isotopes applicable for
diagnostic imaging and targeted radiotherapy, such as 67/68Ga,
111In, 64/67Cu, 86/90Y, 177Lu, 225Ac, 213Bi, 212Pb, and 89Zr.1,2
Radioisotopes that are readily available from commercial
generators and/or small cyclotrons, such as the positron
emission tomography (PET) isotopes 64Cu, 68Ga, and 89Zr and
Received: October 11, 2012
Published: October 29, 2012
© 2012 American Chemical Society
12575
dx.doi.org/10.1021/ic302225z | Inorg. Chem. 2012, 51, 12575−12589

Inorganic Chemistry
Article
Chart 1. Structures of Selected State-of-the-Art “Gold Standard” Chelators, and Chelators Studied or Discussed in This Work
that emit particle radiation (β−/α) have intermediate (12−240
h) half-lives suitable for targeted radiotherapy, and β+ emissions
for diagnostic imaging and dosimetry calculations (“theranos-
tic” isotopes) are of very high utility in cancer patients.4−7 One
isotope that fits into this class is 64Cu (t1/2 = 12.7 h; β+: 656
KeV, 19%; β−: 573 KeV, 40%), which is made especially
attractive because of the existence of imaging and/or
therapeutic surrogate isotopes with complementary half-lives
and decay properties that can be used in combination because
of their identical chemical properties and inherent interchange-
ability (e.g., 67Cu: t1/2 = 61.9 h, β−: 395−577 KeV, 100%).4−7
Contemporary research in the field of receptor-based
radiopharmaceutical targeting has focused almost exclusively
on bifunctional chelators (BFCs) based on the tetraazamacro-
cyclic polycarboxylates and derivatives of the acyclic dieth-
ylenetriamine pentaacetic acid (DTPA) scaffold (Chart 1).
1,4,7,10-Tetraazacyclododecane-1,4,7,10-tetraacetic acid
(DOTA) has historically been applied as an effective chelator
for a variety of radiometals, including isotopes of Cu(II),
Ga(III), Y(III), In(III), and Lu(III), while 1,4,7-triazacyclono-
nane-1,4,7-triacetic acid (NOTA) is the current benchmark
chelator for Ga(III) and is also very stable with Cu(II). The
downside to macrocycles like DOTA is the heating typically
required for quantitative radiolabeling; additionally, despite its
widespread clinical and preclinical use the instability of DOTA
with radioisotopes of copper has been well-documented.2,8−11
The second-generation DTPA derivative, CHX-A′′-DTPA
(Chart 1), chelates radioisotopes of Y(III), In(III), Lu(III),
and Bi(III) with improved stability compared to DTPA, and
although typically not as inert as DOTA in vivo, bifunctional
conjugates of CHX-A′′-DTPA have shown great promise for
antibody labeling with these four radiometals.12−16 The quest
for the ideal BFC for copper has been particularly challenging
because of the facile ligand exchange kinetics and biologically
accessible redox chemistry of Cu(II), but the availability of a
selection of copper radioisotopes (60/61/62/64/67Cu) with useful
properties for therapy and imaging diagnosis/dosimetry drives
basic research and development. The cross-bridged macrocycle
CB-TE2A (Chart 1) is a Cu(II) chelator with exceptional
stability in acid decomplexation experiments17 and unparalleled
in vivo inertness,9,10,18 but it requires unsuitably harsh
conditions (1−2 h at 95 °C) for radiolabeling. Alternatively,
the sarcophagine chelators (e.g., DiamSar), NOTA, and
recently developed methylphosphonate analogues of CB-
TE2A (CB-TE1A1P and CB-TE2P) can be efficiently labeled
at reduced temperatures.19−22 While investigation of these
chelators as bifunctional conjugates remains in its infancy,
preliminary biodistribution studies have shown the uncon-
jugated chelate-radiometal complexes to have slightly reduced
stabilities compared to CB-TE2A, as evinced by their increased
accumulation and slower clearance from the kidney, liver, and
bone marrow over 24 h.19−22 A new Tyr3-octreotide conjugate
of CB-TE1A1P has shown promise as it is radiolabeled with Cu
under mild conditions (40 °C over 1 h) and offers similar in
vivo performance to CB-TE2A conjugates.23
Our group has been investigating a new class of acyclic
chelators based on the pyridinecarboxylate scaffold (Chart 1) as
alternatives to the current chelator offerings that are used for
radiometals of strong practical significance in nuclear medicine.
Our interest in the scaffold originated in the identification of
H2dedpa,24 an acyclic chelator with optimal properties for 68Ga
and one of the first acyclic Ga(III) chelators known to combine
the advantages of efficient room temperature labeling and
excellent in vivo stability, as revealed by the rapid organ and
blood clearance and low uptake in the bone during
biodistribution studies in mice.24,25 Since then, we have been
making new additions to the “pa” family by derivatizing the
pyridinecarboxylate scaffold to make bifunctional conjugates of
H2dedpa, synthesizing chelators like H4octapa and H5decapa
with improved and tunable properties for radioisotopes of
In(III), Y(III), Lu(III), Re, Tc, and Cu(II), and evaluating
small, cationic, lipophilic tracers for perfusion-based cardiac
PET applications with 68Ga.24,26−30 Several bifunctional
analogues of pyridinecarboxylate ligands have been prepared
using the para-nitrobenzyl functionality as precursor to an
isothiocyanate (NCS) linker (Chart 1); however, the
cumbersome synthesis of backbone-derivatized H2dedpa
conjugates, together with the decreased efficiency of N,N′-
ethylenediamine derivatized bifunctional constructs with
67/68Ga, has motivated a need for new conjugation methods
and improved synthetic routes.24,26
Through our work to build on the pyridinecarboxylate-based
chelator H2dedpa we required new and efficient ways of
functionalizing the scaffold toward biovector (e.g., peptides and
12576
dx.doi.org/10.1021/ic302225z | Inorg. Chem. 2012, 51, 12575−12589

Inorganic Chemistry
Article
a
Scheme 1. Synthesis of H2azapa: Reagents and Conditions
a
(i) THF, RT, 3 h; (ii) Propargyl bromide, NaOtBu, TBAI, MeCN, 0 °C−RT, 48 h; (iii) 2:1 DCM/TFA, RT, 1 h; (iv) Methyl 6-bromomethyl
picolinate, Na2CO3, MeCN, RT, 48 h; (v) BnN3, sodium ascorbate (0.2 equiv), Cu(OAc)2 (2.1 equiv), 1:1 t-BuOH/H2O, RT, 16 h; (vi) Na2S, 1:1
THF/H2O, 24 h.
antibodies) conjugation. To this end, we were interested in a
“click” approach toward functionalization through incorpora-
tion of 1,2,3-triazole rings that could be synthesized via the
Huisgen 1,3-dipolar cycloaddition (“click” reaction) of alkynes
and azides. The model chelator H2azapa (Chart 1) was
designed as an elaboration of the “click-to-chelate” approach
pioneered by Mindt and co-workers for the construction and
conjugation of a BFC in a single, easy step31,32 since the triazole
rings would serve not only as convenient linkers between the
chelator and biovectors but also as chelating arms toward the
coordination sphere of large metal ions such as In(III), Y(III),
and Lu(III). Benzyl groups are employed as placeholders for
biological targeting vectors in the model chelator to facilitate
easy synthesis, characterization, and study of the metal
complexes. Herein we report the synthesis and characterization
of H2azapa, a novel and bifunctional triazole-containing acyclic
chelator with promising characteristics for copper radiophar-
maceuticals. The complexation behavior of H2azapa with
nonradioactive In(III), Ga(III), and Lu(III) was studied by
NMR spectroscopy, and crystal structures of [In(azapa)-
(H2O)][ClO4] and [Cu(azapa)] are presented. Radiolabeling
and serum stability experiments were performed with 67Ga,
111In, 64Cu, and 177Lu, and biodistribution experiments and
PET imaging were performed in healthy nude athymic mice
with [64Cu(azapa)].
noethane resulted in nonselective cleavage of both benzyl and
picolinate substituents from the en amines, although various
hydrogenation conditions (methanol or acetic acid as a solvent,
varying catalyst loading, varying concentration) were attemp-
ted.
Thinking ahead to potential challenges that could be
encountered during scale-up and commercialization of a BFC,
we pursued a protecting group strategy for the synthesis of
H2azapa that would be fast, easy, versatile, and high-yielding
with maximal purity, and also tunable for the synthesis of a
library of chelators bearing picolinate, acetate, triazole, and
other arms on a selection of alkylamino backbones. To this end,
a novel N,N′-tert-butoxycarbonyl (boc) protecting group
strategy was employed (Scheme 1), which relied on a
carbamate alkylation reaction of the boc-protected intermediate
1 by propargyl bromide, based on a similar but noncompatible
carbamate alkylation described in the literature (no reaction
occurred following literature protocols).33 This strategy led to
the successful synthesis of the alkyne precursor to H2azapa 4 in
six steps from commercially available ethylenediamine and 2,6-
pyridinedicarboxylate methyl ester, the precursor to methyl 6-
(bromomethyl)picolinate, in 10% cumulative yield, which is a
slight improvement over the synthetically analogous (but
noncompatible) previously applied N-benzyl protection strat-
egy toward the fully protected ligand precursor to H4octapa.28
For the carbamate alkylation, a nontrivial variant of a literature
protocol33 was performed, since it was found that working with
an insoluble, non-nucleophilic base (NaOtBu) in a polar
enolizable solvent (acetonitrile) afforded principally product,
whereas other conditions attempted ad hoc or from the
literature (e.g., combinations of bases such as Cs2CO3/
Na2CO3/K2CO3/KOtBu/NaOH and solvents such as DMF/
THF/tBuOH/MeOH) resulted in uncontrolled carbamate
alkylations and deprotections with no evidence of product
formation. In the carbamate deprotection step, attempts to
extract the free amine from a saturated NaHCO3 solution were
exceptionally inefficient (25 × 1 mL extractions), although they
eventually led to the isolation of product in 87% yield.
RESULTS AND DISCUSSION
■
Synthesis and Characterization. The bifunctional
chelator H2azapa was synthesized according to a reaction
scheme that was designed to improve on a previously reported
method to synthesize H4octapa (Chart 1, Scheme 1) and a
number of pyridinecarboxylate ligands in our current chelator
library.28 The previous strategy to synthesize H4octapa involved
benzyl protection of the primary amines on ethylenediamine
(en) using a simple reductive amination with benzaldehyde
followed by alkylation with two equivalents of tert-butyl
bromoacetate, and then benzyl deprotection to facilitate
subsequent alkylation with methyl 6-(bromomethyl)-
picolinate.28 The main disadvantage to this method is that
the hydrogenation reaction to remove the benzyl protecting
groups is not satisfactory in the presence of olefin, alkyne, or
picolinate substituents; the hydrogenation of N,N′-dibenzyl-
N,N′-bis[6-(methoxycarbonyl)pyridin-2-yl]methyl-1,2-diami-
During the synthesis of the H2azapa alkyne precursor 4
reported herein, it was found that the boc protecting group
strategy was a slight improvement over the benzyl protection
strategy used for the synthesis of H4octapa and related
chelators, both in terms of the simplicity of the synthetic
protocols and the overall yield of the synthesis. A drawback was
12577
dx.doi.org/10.1021/ic302225z | Inorg. Chem. 2012, 51, 12575−12589

Inorganic Chemistry
Article
Figure 1. 1H NMR spectra (DMSO-d6, 25 °C) of (top) H2azapa, 300 MHz; (middle) [Ga(azapa)]+, 600 MHz; (middle) [In(azapa)]+, 400 MHz;
and (bottom, 300 MHz, D2O) [Lu(azapa)]+, showing simple diastereotopic splittings for a single isomer of [Ga(azapa)]+ and major peak broadening
for solution fluxional interconversion between multiple isomers of [In(azapa)]+ and, to a lesser extent, [Lu(azapa)]+.
the time required for the synthesis (minimum 7 days), although
this could conceivably be reduced through optimization.
Unfortunately, attempts to react 1,2-di-tert-butylethane-1,2-
dicarbamate (1) with tert-butylbromoacetate or methyl 6-
(bromomethyl)picolinate did not result in the formation of
product; instead the intact starting material was reisolated in
quantitative yield, and it appears this reaction is only efficient
with propargyl bromide as the electrophile. It is likely that the
scope of the carbamate alkylation reaction is limited by the
reactivity of the alkyl halides, where propargyl halide > acetyl
halide > picolinyl alkyl halide. Ongoing investigation is
currently underway in our laboratories to develop an improved
protecting group strategy that would be applicable to all ligands
in our current chelator library and provide simple and high
yielding access to both “naked” chelators and their bifunctional
derivatives.
esters. The progress of the reaction could be followed by a
gradual but distinct color change from brown to blue-green
over the course of 4 h, after which a blue-green solid
corresponding to [Cu(azapa)] could be observed to precipitate
gradually out of solution. It was observed that super-
stoichiometric rather than catalytic Cu(OAc)2 was required to
effect quantitative triazole formation and methyl ester
deprotection under any reasonable time frame, possibly because
of the increased reduction potential of Cu(II) when tightly
bound to the six-coordinate picolinate core (N4O2) of
H2azapa/4, or the lack of free copper available to associate
with the alkyne/azide click complex during the cycloaddition
reaction. Stoichiometric Cu(II) may effect the picolinate methyl
ester deprotection of H2azapa by binding to the picolinate and
N,N′-en nitrogen atoms of the molecule and interacting with
the lone pairs on nearby carbonyl oxygen atoms to weaken the
double bonds, making them susceptible to nucleophilic attack
by an incoming water molecule.34 In addition, lowering the pKa
of aqua ligands bound to the N,N′-en-Cu(II) complex may
facilitate the production of hydroxide species in close proximity
to the methyl esters, resulting in ester cleavage.34 After initial
The reaction of 4 with two equivalents of benzyl azide in the
presence of Cu(I) generated in situ by the reduction of
Cu(OAc)2 with catalytic sodium ascorbate resulted in the
formation of two 1-benzyl-1,2,3-triazole rings and, simulta-
neously, the quantitative deprotection of the picolinate methyl
12578
dx.doi.org/10.1021/ic302225z | Inorg. Chem. 2012, 51, 12575−12589

Inorganic Chemistry
Article
1
Figure 2. Overlay of H NMR spectra of [In(azapa)]+ (400 MHz, DMSO-d6) at temperatures ranging from 25 to 135 °C, showing a gradual
averaging effect from increasingly rapid interconversion between multiple coordination isomers at higher temperatures.
isolation of the copper-bound chelator [Cu(azapa)] by
filtration, the free chelator was liberated by reaction of the
complex with an excess of Na2S, which sequestered any free or
bound Cu(II) as black/brown insoluble CuS that could be
removed by filtration.
complex of H2dedpa24 and its nitrogen para-nitrobenzyl
functionalized derivative, which showed highly symmetric
octahedral geometries at the Ga(III) center with the six-
coordinate picolinate core (N4O2) of the molecule bound in
each case. Peak broadening for the en backbone methylene
1
For receptor-targeted radiopharmaceuticals, a BFC should
exhibit minimal static or fluxional isomerization in solution,
since fluxional interconversion between different isomers is
thought to affect in vivo stability, and static isomerization may
lead to divergent biological properties, such as pharmacoki-
netics and biodistribution.35−38 The simple, readily interpret-
hydrogen atoms is also commonly observed in the H NMR
spectra of pyridinecarboxylate ligands coordinated to metals.24
The highly symmetric [Ga(azapa)]+ complex has C2 symmetry
in solution as confirmed by the 13C NMR spectrum, which
gives rise to only one signal per pair of equivalent carbon atoms
on either half of the coordinated ligand.
1
able H NMR spectrum of [Ga(azapa)]+ is consistent with a
1
In contrast to that of [Ga(azapa)]+, the H NMR spectrum
single isomer being present in solution (Figure 1). The
spectrum shows prominent diastereotopic splittings for the
methylene hydrogens associated with the picolinate and the
triazole arms at 4.75/4.28 ppm (2J = 18 Hz) and 4.00/3.92 ppm
(2J = 15 Hz), respectively. Methylene hydrogens on the en
bridge exhibit minimal diastereotopic splitting, and are manifest
of [In(azapa)]+ at room temperature shows a series of broad
signals consistent with solution fluxional behavior commonly
observed for complexes of larger trivalent metals such as
In(III), Y(III), and Lu(III) with chelators of high denticities
(Figure 1).35−37 Variable temperature H NMR spectroscopy
1
confirmed the solution fluxional behavior of [In(azapa)]+
(Figure 2), as increasing the temperature of the solution
stepwise from 25 to 135 °C resulted in the gradual sharpening
of peaks to a series of singlets quite distinct from those
observed for the free ligand (Figure 1) as the kinetics of
fluxional interconversion became more rapid with increasing
temperature, resulting in an overall averaging effect. Coales-
cence of a pair of broad signals located at 2.25 and 2.81 ppm at
25 °C can be observed at 65−85 °C, which could suggest an
asymmetric, seven-coordinate structure with one triazole arm
bound in solution, resulting in nonequivalence of the two pairs
1
in the H NMR spectra as a broad quartet, which is barely
resolvable at 600 MHz with a chemical shift of 3.10 ppm. The
benzyl methylene hydrogen atoms behave similarly, and are
resolved as a sharp quartet with a characteristic downfield shift
centered at 5.64 ppm. The less prominent diastereotopic
splittings exhibited by the benzyl and en methylene hydrogen
atoms can be explained by invoking partial flexibility in the en
backbone and partial free rotation of the triazole arms
extending away from the coordination center as unbound
(but potentially chelating) arms. This explanation is consistent
with previously reported crystal structures of the Ga(III)
12579
dx.doi.org/10.1021/ic302225z | Inorg. Chem. 2012, 51, 12575−12589

Inorganic Chemistry
Article
Figure 3. ORTEP drawing of the solid-state molecular structure of [In(azapa)(H2O)][ClO4] obtained by X-ray diffraction. Hydrogen atoms are
omitted for clarity.
of methylene hydrogens on either end of the en bridge at 25
°C.
Table 1. Selected Bond Angles and Lengths in the X-ray
Crystal Structure of [In(azapa)(H2O)][ClO4]
A solid-state molecular structure of [In(azapa)(H2O)][ClO4]
was obtained by X-ray crystallographic analysis of a colorless
rhombic plate crystal grown by slow evaporation of a saturated
solution of 1:1 MeOH/DMSO (Figure 3). The X-ray structure
reveals an eight-coordinate complex in the solid state featuring
a highly distorted square antiprismatic geometry with the
picolinate core, one triazole arm, and one aqua ligand bound to
In(III). While six- and seven-coordinate binding modes for
In(III) are common,2 very few empirical eight-coordinate
structures are known,37,39,40 and to our knowledge [In(azapa)-
(H2O)][ClO4] is the fourth such reported structure. Of the
reported structures, all display distorted square antiprismatic
geometries, and all are fully saturated by chelating arms.37,39,40
Given the propensity of In(III) for seven-coordinate binding
modes, an eight-coordinate structure featuring a heptadentate
chelator with an additional aqua solvent ligand bound is
exceptional.
The metal-to-ligand oxygen bond lengths in [In(azapa)-
(H2O)][ClO4] vary between 2.216/2.320 Å for the indium-
carboxylate bonds In−O(3)/O(1) and 2.219 Å for the indium-
aqua ligand bond In−O(5). The indium-en backbone nitrogen
bonds In−N(2)/N(7) are longer at 2.461/2.561 Å, while the
indium-triazole nitrogen bond In−N(8) is the shortest of the
indium−nitrogen bonds at 2.292 Å (Table 1). The In-X bond
lengths for equivalent donor atoms on either side of the ligand
C2 symmetry axis are similar but not identical, as is to be
expected for the asymmetric heptadentate binding mode with
indium. The triazole arm is bound to indium at the N3 position
of the triazole ring (see Chart 1), which is consistent with
previously reported crystal structures32,41−43 and calculated
structures44 in which the triazole chelating arm is attached to
the chelator at the C4 position of the triazole.42,45 The relatively
short In−N(8) bond in [In(azapa)(H2O)][ClO4] is consistent
with recent density functional theory (DFT) calculations31,46,47
which suggested that a greater concentration of electron density
bond
length [Å]
angle
degree [deg]
In−O(3)
In−O(5)
In−N(8)
In−O(1)
In−N(1)
In−N(6)
In−N(2)
In−N(7)
2.216(1)
2.219(1)
2.292(1)
2.320(1)
2.303(1)
2.345(1)
2.461(1)
2.561(1)
O(3)−In−O(6)
N(6)−In−N(7)
O(5)−In−N(6)
O(1)−In−O(5)
O(1)−In−N(8)
N(2)−In−O(3)
N(2)−In−N(8)
O(3)−In−O(5)
70.25(4)
68.10(4)
76.39(4)
74.03(4)
82.89(4)
91.69(4)
97.22(4)
102.51(4)
at the N3 position of the triazole in 1,4-substituted triazoles
should result in relatively strong electron donating capability to
a metal center.
Since characterization using NMR spectroscopy was not
feasible with the paramagnetic d9 Cu(II) complex of H2azapa,
the bulk composition of C, N, and H in a RP-HPLC-purified
batch of [Cu(azapa)] was confirmed by elemental analysis. A
small, blue-green, cubic crystal of [Cu(azapa)] suitable for
characterization using X-ray diffraction was grown by slow
evaporation of a saturated solution of purified complex in 50%
acetonitrile in water. The solid-state molecular structure of
[Cu(azapa)] features a distorted octahedral metal−ligand
environment typical of six-coordinate complexes of Cu(II)
(Figure 4).2 The central binding of H2azapa via the picolinate
(N4O2) core is typical for crystal structures reported by our
group thus far for Cu(II) and Ga(III) complexes of dedpa2−
and its bifunctional derivatives, corroborating the notion that
the picolinate core of these ligands provides a versatile
coordination pocket for a variety of metal ions.24,27,30 A high
degree of symmetry for these complexes, along with an
approximately equally distributed set of metal−ligand bond
lengths, is thought to correlate well with their high stability and
favorable biological properties.24,27 That [Cu(azapa)] is more
symmetric and evenly distributed than [Cu(dedpa)],27 and
nearly as symmetric as [Ga(dedpa)][ClO4]24 is belied only by
12580
dx.doi.org/10.1021/ic302225z | Inorg. Chem. 2012, 51, 12575−12589

Inorganic Chemistry
Article
Figure 4. ORTEP drawing of the solid-state molecular structure of [Cu(azapa)]. Hydrogen atoms are omitted for clarity.
a
Table 2. Selected Bond Angles and Lengths in the X-ray Crystal Structure of [Cu(azapa)]
bond
length [Å] Cu(azapa)
length [Å] Cu(dedpa)27
length [Å] Ga(dedpa)24
angle
degree [deg]
Cu−N(1)
Cu−N(1)i
Cu−O(1)
Cu−O(1)i
Cu−N(2)
Cu−N(2)i
1.935(1)
1.9386(13)
2.0008(12)
2.3014(11)
2.0430(10)
2.1364(13)
2.3171(13)
1.987(2)
1.990(2)
1.971(1)
1.983(1)
2.112(2)
2.113(1)
O(3)−In−O(6)
N(6)−In−N(7)
O(5)−In−N(6)
O(1)−In−O(5)
O(1)−In−N(8)
N(2)−In−O(3)
70.25(4)
68.10(4)
76.39(4)
74.03(4)
82.89(4)
91.69(4)
”
2.120(1)
”
2.326(1)
”
a
Relevant bond lengths are compared to the analogous bond lengths in the previously reported X-ray crystal structure of [Ga(dedpa)]+ and
[Cu(dedpa)].24,27
the relatively long Cu−N(2) bond distances (Table 2), which
suggest that, like in [In(azapa)][ClO4] and [Ga(dedpa)]-
[ClO4], the en backbone may be somewhat loosely bound to
the coordination center. The relatively long and weak Cu−
N(2) bonds could conceivably facilitate attack from a biological
nucleophile in vivo; on the other hand, the chelating ability of
the triazole arms, which are covalently fixed to the en backbone
and held in close proximity to the metal center, could also play
a role in fluxional solution behavior by transiently binding
Cu(II) to improve stability and inertness. Alternatively, it is
conceivable that the elongated Cu−N(2) distances could be a
simple result of energetic contributions from packing in the
solid-state.
Potentiometric titrations of H2azapa with copper were
attempted so that the binding affinity (log KML, pM) could
be determined, but the insolubility of [Cu(azapa)] obviated
these experiments and cyclic voltammetry experiments. Syn-
thesis of pegylated hydrophilic derivatives of H2azapa is
underway, since clickable hydrophilic polyethylene glycol
(PEG) moieties should be more amenable toward aqueous
experiments than the lipophilic benzyl groups of H2azapa. The
modular synthetic scheme for H2azapa will allow for the
“clickable” alkyne precursor 4 to be easily transformed into a
variety of derivatives.
°C in a microwave reactor for quantitative labeling with 111In,
and 60 min at 90 °C for 64Cu and 177Lu.28 The single, sharp
peak in the HPLC radio-traces obtained for [67Ga(azapa)]+ (tR
= 12.8 min) and [111In(azapa)]+ (tR = 11.4 min) is consistent
with the NMR solution structural data for these complexes,
which indicate a single, robust, highly symmetric coordination
isomer for [67Ga(azapa)]+ and rapid fluxional interconversion
between multiple isomers for [111In(azapa)]+ on the NMR/
HPLC time scale. Single peaks were also observed in the HPLC
radiotraces of [177Lu(azapa)]+ (tR = 19.3 min) and [64Cu-
(azapa)] (tR = 20.9 min) during subsequent radiolabeling
experiments. RP-HPLC retention times for these radiometals
indicated the increased lipophilicity of the Mn+(azapa)
complexes compared to the Mn+(DOTA) and Mn+(DTPA)
standards. For example, retention times for the 177Lu/64Cu
complexes of H2azapa were tR = 19.3/20.9 min, and for the
same complexes of DOTA they were tR = 6.8/7.1 min. The
absence of free radiometal (tR ∼ 0−4 min) in the radioactive
trace was confirmed in all cases by the RP-HPLC radiotraces.
Blood Serum Stability Studies. The serum stability data
for the complexes of H2azapa with 67Ga, 111In, 64Cu, and 177Lu
establish H2azapa as a viable and important target for further
investigation with 64Cu (Table 3). [64Cu(azapa)] displays
excellent stability over the 20 h time period studied (37 °C),
with less than 2% transchelated to blood serum proteins at that
time, compared to [64Cu(DOTA)]2−, which remained only
77% and 74% intact after 1 and 20 h, respectively. The data also
compare favorably to a recent report of a mouse serum
competition experiment of H2dedpa and a N,N′-para-nitro-
benzyl functionalized cRGDyk conjugate,27 H2RGD-2, with
Radiolabeling Experiments. Initial radiolabeling experi-
ments have confirmed the efficiency of H2azapa to rapidly form
radioactive complexes of 67Ga, 111In, 64Cu, and 177Lu
quantitatively at ambient temperature in aqueous buffer (pH
4−5, NaOAc, 10 min), as determined by integration of RP-
HPLC radio-traces. In contrast, DOTA required 20 min at 80
12581
dx.doi.org/10.1021/ic302225z | Inorg. Chem. 2012, 51, 12575−12589

Inorganic Chemistry
Article
temperature are highly sought to advance receptor-based
targeting using antibodies and other sensitive biological vectors
for PET with 64Cu and dosimetry/therapy using
64Cu/67Cu.48−52 Antibodies have longer biological half-lives,
sluggishly distribute to target tissues, and must be labeled at
reduced temperatures (20−37 °C) to avoid protein denatura-
tion and retarded immunoreactivity and binding affinity.49,50,53
PET using antibody targeting is typically optimized at ∼24−72
h post injection, which requires the chelate-radiometal complex
to be stable over that entire time period.54,55 Furthermore,
tumor targeting for therapy using isotopes such as 67Cu, 177Lu,
and 90Y require an even longer biological stability window,
since the chelate-radiometal complex should be delivered to the
target tumor and then remain intact until the majority of the
isotope has decayed (e.g., 67Cu: t1/2 = 61.9 h).50,55 Kinetic
inertness has been shown to be a more accurate predictor of the
biological stability of chelate-radiometal complexes than
thermodynamic stability,2,9,56,57 and to date the most relevant
assays for biological stability, besides in vivo experiments, are in
vitro blood serum/apo-transferrin/superoxide dismutase com-
petition challenges.
Table 3. Stability Data Collected in Mouse Blood Serum at
25 °C for 67Ga and 111In and in Human Blood Serum with
Agitation (550 rpm) at 37 °C for 64Cu and 177Lu, with
a
H2azapa and Selected Chelator Standards
(a) short-term (b) medium-term (c) long-term
b
b
complex
stability (%)
stability (%)
stability (%)
[67Ga(azapa)]+
[67Ga(dedpa)]+
69 ( 2)
>99
48 ( 2)
>99
52 ( 4)
>99
[67Ga(dp-N-
88
69
51
NO2)]+
[67Ga(dp-bb-
NO2)]+
98
97
97
[
[
[
111In(azapa)]+
111In(DTPA)]2−
111In(DOTA)]−
95.1 ( 0.6)
87 ( 2)
65 ( 4)
88 ( 2)
89 ( 2)
90 ( 2)
[64Cu(azapa)]
[64Cu(DOTA)]2−
99.9 ( 0.1)
77 ( 6)
98.3 ( 0.2)
74 ( 1)
[
[
[
177Lu(azapa)]+
177Lu(DOTA)]−
177Lu(DTPA)]2−
55 ( 4)
46 ( 1)
87 ( 2)
82 ( 2)
87.7 ( 0.7)
77 ( 1)
The blood serum stability data for chelators with 67Ga, 111In,
and 177Lu reveal the stability of H2azapa with each of these
radiometals. The H2azapa-bound 177Lu and 111In are 54% and
35% transchelated by serum proteins after 24 h, compared with
our leading candidate for these radiometals, H4octapa, which is
>85% stable at 24 h with both 177Lu and 111In (<15%
transchelated).28 H2azapa displays roughly comparable stability
as a complex with 67Ga to the similar nitrogen functionalized
para-nitrobenzyl H2dedpa derivative H2dp-N-NO2, which was
roughly 50% intact after 2 h in a similar apo-transferrin
challenge experiment.24 Between the quartet of H2azapa,
H2dedpa,24 H4octapa,28 and H5decapa,28 we have a large
number of clinically relevant and high impact metal ions/
isotopes well covered.
A much more demanding test of the biological stability of the
chelate-radiometal complex is a biodistribution experiment in
animals, where the ability of the complex to resist trans-
chelation by native biological chelating proteins and enzymes
like superoxide dismutase (SOD), serum albumin, and
transferrin is ultimately tested by its effective clearance through
the kidneys (ideally) and low uptake in the liver, lung, spleen,
a
The % stability shown is the percentage of chelator-bound radiometal
and the error is expressed as standard deviation (n = 3). (a) Short-
term stability data at 15 min for [67Ga(azapa)]+ and 10 min for all
other complexes; (b) medium-term stability data at 1 h, except for
177Lu complexes which were at 1.5 h; (c) long-term stability data at 2 h
for 67Ga complexes, 20 h for 64Cu complexes, and 24 h for 111In and
b
177Lu complexes. Relevant data for the 67Ga complexes of H2dedpa
and the nitrogen and backbone para-nitrobenzyl functionalized
H2dedpa derivatives H2dp-N-NO2 and H2dp-bb-NO2 are from
previously reported apo-transferrin challenge experiments.24
64Cu in which 23% of 64Cu from [64Cu(dedpa)] and 12% of
64Cu from [64Cu(RGD-2)] was transchelated to serum proteins
after 24 h at ambient temperature,27 suggesting that H2azapa is
a more stable/inert chelator for copper than are H2dedpa and
its N,N′-en-nitrogen functionalized derivatives. Chelators that
remain robust with isotopes of copper over a much longer time
period (e.g., 24−72 h) while efficiently labeling at room
Figure 5. Biodistribution of [64Cu(azapa)] and [64Cu(DOTA)]2− showing organ and tissue uptake as percent injected dose per gram (% ID/g)
obtained over 24 h in healthy athymic nude mice; Y-axis is normalized at 25% ID/g for clarity.
12582
dx.doi.org/10.1021/ic302225z | Inorg. Chem. 2012, 51, 12575−12589

Inorganic Chemistry
Article
Figure 6. PET images of two (A and B) healthy female nude athymic mice injected with [64Cu(azapa)] and imaged at 1, 4, and 24 h post injection,
showing rapid clearance through the gut, with ∼8%ID/g remaining in the liver after 24 h.
Table 4. Biodistribution of [64Cu(azapa)] and [64Cu(DOTA)]2− in Healthy Nude Athymic Mice (n = 4) Showing Organ Uptake
as % ID/g, with the Error Expressed as Standard Deviation (SD), at 15 min, 1 h, 4 h, and 24 h Time Points
organ
blood
15 min ( SD)
1 h ( SD)
4 h ( SD)
24 h ( SD)
[64Cu(azapa)]
1.62 ( 0.20)
2.91 ( 0.43)
6.39 ( 0.71)
15.97 ( 1.43)
2.27 ( 0.33)
5.93 ( 0.43)
2.11 ( 0.50)
14.28 ( 5.77)
6.84 ( 1.28)
0.45 ( 0.08)
1.30 ( 0.39)
1.77 ( 0.45)
[64Cu(DOTA]]2−
0.21 ( 0.02)
0.37 ( 0.04)
0.89 ( 0.20)
1.91 ( 0.07)
0.34 ( 0.11)
0.70 ( 0.17)
0.37 ( 0.10)
1.01 ( 0.47)
1.66 ( 0.07)
0.09 ( 0.02)
0.45 ( 0.13)
0.40 ( 0.04)
1.72 ( 0.12)
2.12 ( 0.12)
5.10 ( 0.28)
17.71 ( 3.48)
1.50 ( 0.22)
2.64 ( 0.75)
1.28 ( 0.21)
63.73 ( 26.06)
8.87 ( 0.93)
1.27 ( 1.31)
1.27 ( 0.06)
2.52 ( 0.59)
1.62 ( 0.20)
3.81 ( 0.55)
8.24 ( 2.03)
12.39 ( 2.64)
3.72 ( 0.97)
5.11 ( 0.62)
21.12 ( 3.44)
5.18 ( 0.85)
6.98 ( 0.77)
0.47 ( 0.06)
2.38 ( 0.78)
2.38 ( 0.54)
1.19 ( 0.08)
3.44 ( 0.44)
4.30 ( 0.60)
8.14 ( 0.94)
3.13 ( 0.96)
2.13 ( 0.58)
1.93 ( 0.44)
3.05 ( 0.18)
4.80 ( 0.60)
0.48 ( 0.07)
1.88 ( 0.50)
1.04 ( 0.36)
heart
lung
liver
spleen
stomach
large intestine
small intestine
kidney
muscle
bone
skin
blood
2.77 ( 0.95)
1.53 ( 0.43)
2.61 ( 0.66)
2.32 ( 0.61)
0.90 ( 0.23)
1.03 ( 0.21)
0.42 ( 0.03)
1.53 ( 0.43)
6.81 ( 1.22)
0.72 ( 0.29)
1.17 ( 0.57)
1.76 ( 0.47)
0.22 ( 0.03)
0.57 ( 0.10)
1.25 ( 0.60)
1.64 ( 0.50)
0.69 ( 0.18)
0.72 ( 0.20)
2.19 ( 0.53)
0.88 ( 0.36)
1.59 ( 0.20)
0.09 ( 0.03)
0.55 ( 0.40)
0.34 ( 0.11)
0.16 ( 0.01)
0.52 ( 0.07)
0.71 ( 0.12)
1.04 ( 0.10)
0.48 ( 0.16)
0.35 ( 0.10)
0.35 ( 0.08)
0.44 ( 0.08)
0.93 ( 0.05)
0.09 ( 0.03)
0.27 ( 0.07)
0.17 ( 0.03)
heart
lung
liver
spleen
stomach
large intestine
small intestine
kidney
muscle
bone
skin
kidneys, intestines, blood, and bone. The clearance and
biological distribution pattern of a metal complex is dependent
on the size, shape, charge, thermodynamic stability, kinetic
inertness, and overall polarity of the complex, such that these
same properties can be manipulated to modulate pharmacoki-
netics and promote effective uptake into, or away from, various
organs.30,38 It is sometimes difficult to discern whether a
complex shows uptake in an organ because of its instability/
transchelation to proteins within that organ (or blood), or
whether the intact complex is being taken up and retained
because of its physical properties. Metabolism studies, which
can help identify % authentic intact (% AI) complex within a
given organ vs “free” radiometal can be useful to this end.9,10
For bifunctional chelate-based radiopharmaceuticals, the
biodistribution profile is largely dominated by the properties
of the attached biovector (e.g., antibody, peptide, nanoparticle),
particularly in the case of large targeting vectors such as
antibodies (MW∼150 kDa) and antibody fragments.18,20,24,26
The biodistribution of [64Cu(azapa)] shows elevated uptake
in all organs compared to [64Cu(DOTA)]2−, particularly the
liver, lung, heart, spleen, stomach, intestines, and kidneys
(Figure 5). [64Cu(azapa)] can be observed to move through the
digestive tract over 24 h, with activity levels peaking at 15 min
post injection in the small intestines (63.7−14.3%ID/g from 15
min to 1 h), peaking in the large intestine at 4 h (21.1%ID/g),
and finally being excreted through the feces, as pictured via
PET (Figure 6), confirming elimination through the digestive
system (Table 4). High uptake in the stomach was also
observed, with activity levels peaking at 1 h (5.6%ID/g).
Although metabolism and clearance appeared to be mostly
through the hepatobiliary and digestive system, there was also
some kidney clearance observed, and the amount of activity
decreased over 24 h in both of these organs (liver: 17.7−8.1%
12583
dx.doi.org/10.1021/ic302225z | Inorg. Chem. 2012, 51, 12575−12589

Inorganic Chemistry
Article
ID/g; kidney: 8.9−4.8%ID/g, 15 min-24 h). While an increase
in 64Cu uptake in the liver and kidneys is often cited as being
evidence for instability of the 64Cu chelate-radiometal complex
in vivo, the complex [64Cu(azapa)] is accumulated in many
other organs as well, strongly suggesting distribution of the
intact lipophilic complex. DOTA in particular is known to
display instability and transchelation of 64Cu in mice with
bioconjugate systems,8−11 and the differences in the biodis-
tribution of H2azapa and DOTA, particularly the elevated
uptake of [64Cu(azapa)] in the stomach, intestines, heart, and
lung, suggest that the mechanism of uptake in these organs is
related to the physical characteristics of the intact complex, as
opposed to the instability of [64Cu(azapa)] and resultant
transchelation to proteins in these organs.
Persistent uptake in the liver, lung, heart, and gut are
expected for small lipophilic molecules, and elevated uptake in
the blood and kidneys is also often observed.58−60 The pressing
issue with 64Cu is that “free” or transchelated Cu(II) is known
to accumulate in the liver, and highly lipophilic molecules (such
as [Cu(azapa)]) are also known to accumulate in the liver and
kidneys, which makes the standard assessment of stability by
monitoring the amount of 64Cu accumulated in the liver
problematic. Furthermore, with the high abundance of
superoxide dismutase (SOD) in the liver, accumulation and
retention of the intact and lipophilic [64Cu(azapa)] complex in
the liver could result in very harsh metabolic conditions and a
very strict Cu(II) transchelation challenge with SOD.56 This is
in sharp contrast to the hydrophilic [Cu(DOTA)]2− complex,
which cleared very quickly through the kidneys/bladder/urine,
and since a majority of the injected activity was cleared within
15 min, the complex was not subject to the same harsh and
degrading environment (liver) as [64Cu(azapa)].
To observe a fair competition between these chelators,
bioconjugates (e.g., antibody) must be synthesized and
compared in vivo. The rapid clearance of high amounts of
64Cu through the digestive system, as well as its high uptake in
the lungs and heart, suggest that [64Cu(azapa)] remains intact.
Recent investigation into a series of small, cationic, lipophilic
derivatives of H2dedpa that bore lipophilic methoxyphenyl
substituents showed that with 67/68Ga the complexes that were
the most stable in vitro with 67Ga, but that were concurrently
the most lipophilic by RP-HPLC, displayed greatly elevated
uptake in the liver, intestine, heart, and lung, although they
remained stable as confirmed by low bone uptake of “free”
67Ga.30 Uptake and clearance from the kidneys was also
prominent in the biodistribution for each of these compounds,
particularly those that were unstable in vitro. For comparison,
Ando et al. injected 64CuCl2 into Donryu rats with implanted
sarcomas, and found that activity accumulation was highest for
“free” metal in the liver and kidney,61 while for 64Cu complexes
of DOTA, TETA, and cross-bridged cyclams known to be
unstable in vivo,10,11,56 abnormal accumulation in the blood and
bone was also commonly observed.10,11,56
that future elaborations of the H2azapa scaffold as hydrophilic
peptide bioconjugates will remain intact in vivo and work well
as 64Cu PET imaging agents. Efforts are currently underway to
synthesize hydrophilic PEG-3 and peptide conjugates of
H2azapa to modulate the pharmacokinetics away from the
gut, liver, and lungs, when compared to H2azapa, toward
selective tumor uptake for imaging and therapy. It is anticipated
that the propensity of [64Cu(azapa)] toward persistent organ
uptake is solely a result of its lipophilic character and not
chelate instability, and the modular and “clickable” H2azapa
scaffold should be an excellent candidate for receptor-based
targeting using larger, more sensitive biological targeting
vectors like peptides and antibody fragments that require low
temperature radiolabeling.
The ability of H2azapa to radiolabel 177Lu, 111In, 67Ga, and
64Cu quantitatively in 10 min at ambient temperature shows
impressive flexibility and versatility. It is also of interest to note
that high heart uptake was observed over 24 h (uptake of 3.8%
ID/g at 4 h and 3.4%ID/g at 24 h), with a heart/blood ratio of
1.9 at 1 h and 2.9 at 24 h, which is comparable to many
H2dedpa cardiac derivatives recently published,30 and fortu-
itously suggests that a modified version of this neutrally charged
Cu(II) chelator scaffold could be used for heart imaging.
Further, the neutral and lipophilic physical properties of
H2azapa could lead to applications in tumor hypoxia imaging
through decoration of the molecular scaffold with nitro-
imidazole functionality. The modularity of the “clickable”
precursor 4 (alkyne functionalized chelator) adds a large degree
of versatility to this scaffold, as many different bifunctional
derivatives can be synthesized from the same precursor (e.g.,
peptide, antibody, antibody fragment, nanoparticle, PEG chain,
nitro-imidazole), which will allow for facile and simple synthesis
of these derivatives with little or no changes in synthetic
methodology. H2azapa provides a highly versatile and flexible
molecular framework for synthesizing facile click-based
bioconjugates; an important addition to the “pa” family,
which currently includes the quartet of chelators H2azapa,
H2dedpa,24 H4octapa,28 and H5decapa,28 and covers a wide
base of radiometal ions for imaging and therapy applications.
CONCLUSIONS
■
We have presented the synthesis, characterization, complex-
ation, and radiolabeling of H2azapa, the first known example of
a bifunctional triazole-containing acyclic chelator based on a
“clickable” version of the pyridinecarboxylate scaffold. Besides
investigating its complexation and stability properties with a
selection of radiometals of current interest in nuclear medicine,
we were interested to investigate the feasibility of an alternative
“click” approach toward biovector conjugation and derivatiza-
tion of the scaffold toward potentially higher denticity systems
with larger metals such as In(III), Y(III), and Lu(III). Our
hypothesis was that the formation of robust, coordinatively
saturated complexes in the solution phase would result in
kinetically inert chelate-radiometal complexes in vivo.
The reason for elevated uptake of [64Cu(azapa)] in various
organs is difficult; however, the model ligand H2azapa was
never intended as a biologically relevant chelator, with its
lipophilic benzyl “placeholders” being excellent for studies using
standard analytical techniques but not intended for use in final
radiopharmaceutical formulations. The elevated uptake of
[64Cu(azapa)] in the small intestine, large intestines, stomach,
spleen, heart, and lungs suggests that the physical properties of
the neutral, lipophilic complex facilitate uptake of the intact
complex into these organs. This promising observation suggests
From our preliminary investigation, H2azapa appears to be a
viable candidate as a “clickable” bifunctional chelator for copper
radiopharmaceuticals, and a valuable addition to the “pa” family
of chelators (H2dedpa, H4octapa, and H5decapa). This is
demonstrated by the ability of H2azapa to quantitatively
radiolabel 64Cu in 10 min at ambient temperature in aqueous
buffer (pH 4−5), and by its robust inertness in human blood
serum, in which it shows only 2% transchelation by serum
proteins after a 20 h incubation period. This is in stark contrast
12584
dx.doi.org/10.1021/ic302225z | Inorg. Chem. 2012, 51, 12575−12589

Inorganic Chemistry
Article
connected to a Waters 600 controller, a Waters 2487 dual wavelength
absorbance detector, and a Waters delta 600 pump. Radiochemical
experiments with 67Ga and 111In were performed at TRIUMF/
Nordion in Vancouver, BC. 67Ga and 111In were cyclotron produced
(Advanced Cyclotron Systems, Model TR30) by proton bombard-
ment through the reactions 68Zn(p,2n)67Ga and 111Cd(p,n)111In and
provided by Nordion as 67GaCl3 and 111InCl3 solutions in 0.05 M HCl.
Analysis of radiolabeled compounds was performed using an analytical
HPLC system consisting of a Waters xbridge BEH130 C18 reverse
phase (150 × 6 mm) analytical column and a Waters Alliance HT
2795 separation module equipped with a Raytest Gabi Star NaI (Tl)
detector (Raytest GmbH, Straubenhardt, Germany) and a Waters 996
photodiode array (PDA) detector. Blood serum competition experi-
ments with 67Ga/111In were performed using previously frozen mouse
blood serum (Sigma, M5905−5 mL). Blood serum competition
solutions were analyzed using GE Healthcare Life Sciences PD-10
desalting columns (GE Healthcare, United Kingdom, MW < 5000 Da
filter) conditioned by elution of 20 mL of phosphate-buffered saline
(PBS) before use. Radioactivity counting for 67Ga and 111In was
performed using a Capintec CRC 15R dose calibrator (Capintec,
Ramsey, NJ).
to DOTA, which required elevated temperatures for radio-
labeling and was only 74% intact after 20 h incubation period
with human blood serum, further substantiating its nonoptimal
properties for both antibody and peptide labeling with 64Cu.
The solid-state molecular structure of [64Cu(azapa)] confirmed
a coordinatively saturated six-coordinate structure with the
picolinate (N4O2) core of the molecule bound and the triazole
arms extending away from the metal ion center. Although the
In(III) complex of H2azapa was unstable in serum and was not
further investigated in vivo, it formed a highly unusual and rare
eight-coordinate complex with In(III) showing coordination to
the picolinate core, one triazole arm, and one aqua ligand, one
of only a handful of 8-coordinate In(III) complexes.
Despite the high stability of [64Cu(azapa)] in the presence of
biological serum proteins, uptake in a variety of organs, notably
the liver, stomach, intestines, and kidneys was observed in nude
athymic mice, most likely because of its neutrally charged and
highly lipophilic physical properties. [64Cu(azapa)] cleared
from the liver and passed through the intestines and gut
relatively rapidly in mice, with ∼8% ID/g remaining in the liver
after 24 h; however, persistent uptake in the liver, lung, heart,
and gut are expected for small, lipophilic molecules, and
H2azapa is merely a model ligand designed for preliminary
study. [64Cu(azapa)] demonstrated exceptional stability
compared to [64Cu(DOTA)]2− in serum, and the uptake of
[64Cu(azapa)] in organs and tissues can be explained by the
high lipophilicity of the intact, neutral complex, which suggests
great potential for future elaboration with hydrophilic peptide
biovectors. The modular design of this “clickable” system will
facilitate the fast and simple synthesis of a number of new
derivatives. Further investigation into the biological stability of
bifunctional bioconjugates of H2azapa exploiting hydrophilic
peptides and PEG groups will likely secure the potential of
H2azapa as a useful bifunctional chelator for copper.
Radiochemical experiments with 177Lu and 64Cu were performed at
Memorial Sloan-Kettering Cancer Center (MSKCC) in New York and
utilized the following materials and methods. 177Lu was procured from
Perkin-Elmer (Perkin-Elmer Life and Analytical Sciences, Wellesley,
MA, effective specific activity of 29.27 Ci/mg) as 177LuCl3 in 0.05 M
HCl. 64Cu was purchased from Washington University, St. Louis
(Washington University School of Medicine Cyclotron, model CS-15,
Cyclotron Corp., 64Ni(p,n)64Cu) and purified as previously described
to yield [64Cu]CuCl2 with an effective specific activity of 200−400
mCi/μg (7.4−14.8 GBq/μg).62 All radiolabeling chemistry was
performed with ultrapure water (>18.2 MΩ cm−1 at 25 °C, Milli-Q,
Millipore, Billerica, MA) which had been passed through a 10 cm
column of Chelex resin (BioRad Laboratories, Hercules, CA). The
HPLC used was a Shimadzu SPD-20A prominence UV/vis, LC-20AB
prominence LC, a Bioscan flow-count radiation detector, and a C18
reverse phase column (Phenomenex Luna Analytical 250 × 4.6 mm).
Blood serum competition experiments with 64Cu/177Lu were
performed using human blood serum (Sigma, Sera, human, aseptically
filled, S7023−100 mL). Human blood serum competition solutions
were agitated at 550 rpm and held at 37 °C using an Eppendorf
Thermomixer and then analyzed using GE Healthcare Life Sciences
PD-10 desalting columns as described above for 67Ga and 111In serum
competition solutions. Radioactivity in samples was measured using a
Capintec CRC-15R dose calibrator (Capintec, Ramsey, NJ), and for
biodistribution studies a Perkin-Elmer (Waltham, MA) Automated
Wizard Gamma Counter was used for counting organ activities and
creating calibration curves. PET imaging was performed using a micro-
PET R4 rodent scanner (Concord Microsystems).
EXPERIMENTAL SECTION
■
Materials and Methods. All solvents and reagents were
purchased from commercial suppliers (Sigma Aldrich, St. Louis,
MO; TCI America, Portland, OR; Fisher Scientific, Waltham, MA)
and were used as received unless otherwise indicated. Methyl 6-
(bromomethyl)picolinate was synthesized according to a literature
protocol.28 Water used was ultrapure (18.2 MΩ cm−1 at 25 °C, Milli-
Q, Millipore, Billerica, MA). The analytical thin-layer chromatography
(TLC) plates were aluminum-backed ultrapure silica gel (Siliaplate, 60
Å pore size, 250 μM plate thickness, Silicycle, Quebec, QC). Flash
column silica gel was provided by Silicycle (Siliaflash Irregular Silica
Gels F60, 60 Å pore size, 40−63 mm particle size, Silicycle, Quebec,
QC). Automated column chromatography was performed using a
Teledyne Isco (Lincoln, NE) CombiFlash Rf automated system with
solid load cartridges packed with flash column silica gel and RediSep Rf
Gold reusable normal-phase silica columns (Teledyne Isco, Lincoln,
Di-tert-butyl Ethane-1,2-dicarbamate (1). To a solution of
ethylenediamine (770 μL, 11.5 mmol) in tetrahydrofuran (THF, 60
mL) was added dropwise a solution of di-tert-butyl dicarbonate (5.449
g, 25.0 mmol, 2.2 equiv) in THF (10 mL). A white solid precipitated
immediately. The resulting white suspension was stirred under argon
at ambient temperature and monitored by TLC and EI-MS. TLC
showed the gradual disappearance of tert-butyl dicarbonate (Rf: 0.65,
TLC in 25% EtOAc in hexanes/KMnO4 staining) in favor of product
(Rf: 0.25). After 3 h the reaction mixture was concentrated in vacuo to
dryness, and the resulting white solid was redissolved in dichloro-
methane (100 mL). The organic layer was washed with water (2 × 20
mL) and brine (1 × 20 mL). The combined aqueous layers were dried
(MgSO4), filtered, and concentrated in vacuo to dryness. The resulting
colorless solid was dissolved in a minimum volume (∼5 mL) of
dichloromethane and purified by column chromatography (Combi-
Flash Rf automated column system; 40 g HP silica; eluted with a
gradient of 0−100% EtOAc in hexanes) to afford the product 1 as a
white solid (1.49 g, 50%, Rf: 0.25, TLC in 25% EtOAc in hexanes/
1
NE). H and 13C NMR spectra were recorded on Bruker AV300,
AV400, or AV600 instruments; all spectra were internally referenced
to residual solvent peaks except for 13C NMR spectra in D2O, which
were externally referenced to a sample of CH3OH/D2O. Low-
resolution mass spectrometry was performed using a Waters liquid
chromatography−mass spectrometer (LC-MS) consisting of a Waters
ZQ quadrupole spectrometer equipped with an ESCI electrospray/
chemical ionization ion source and a Waters 2695 HPLC system
(Waters, Milford, MA). High-resolution electrospray-ionization mass
spectrometry (EI-MS) was performed on a Waters Micromass LCT
time-of-flight instrument. Microanalysis for C, H, and N was
performed on a Carlo Erba EA 1108 elemental analyzer. The HPLC
system used for purification of nonradioactive compounds consisted of
a semipreparative reverse phase C18 Phenomenex synergi hydro-RP
(80 Å pore size, 250 × 21.2 mm, Phenomenex, Torrance, CA) column
1
KMnO4 staining). H NMR (300 MHz, CDCl3, 25 °C) δ: 5.08 (s,
2H), 3.18 (s, 4H), 1.40 (s, 18H). 13C NMR (75 MHz, CDCl3, 25 °C)
12585
dx.doi.org/10.1021/ic302225z | Inorg. Chem. 2012, 51, 12575−12589

Inorganic Chemistry
Article
δ: 156.28, 79.19, 40.68, 28.28. HR-ESI-MS calcd. for
[C12H24N2O4+Na]+: 261.1814; found: 261.1814.
and a plastic-tipped autopipet to prevent a friction-induced explosion.
Sodium ascorbate (27 mg, 0.14 mmol, 0.2 equiv) was added, followed
by copper(II) acetate monohydrate (276 mg, 1.38 mmol, 2 equiv).
The suspension turned dark brown on addition of copper(II) acetate,
but over 4 h the solid gradually dissolved and the solution lightened to
blue-green. The reaction mixture was stirred at ambient temperature
overnight (16 h) after which time a light blue-green precipitate was
observed. The suspension was concentrated in vacuo to dryness. The
resulting light blue-green solid was dissolved in 1:1 THF/water (50
mL) and sodium sulfide (829 mg, 3.5 mmol, 5 equiv) was added. A
brown solid precipitated immediately, and the suspension was left to
stir at ambient temperature for 24 h. The brown solid was filtered out
using a fine fritted filter and the colorless filtrate was concentrated in
vacuo to dryness. The resulting colorless solid was dissolved in a
minimum volume of water and prepurified using a Waters C18 Sep-
Pak 6 cc Vac cartridge (1 g sorbent, 55−100 μm particle size, gradient:
100% water to 50% MeOH/water to 100% MeOH) in 6−8 small
portions. The cartridge was conditioned with methanol (2 × 2 mL)
and water (2 × 2 mL) before each use. UV activity was confirmed in
the 50% MeOH/water fractions by TLC before they were combined
and concentrated in vacuo to dryness. The partially purified white solid
was purified by semipreparative HPLC (A: 0.1% TFA, B: 100%
MeCN, 0−100% B over 25 min, tR (broad) 15.7−18.0 min, 50 mg
injections) to afford the product 5·TFA as a white crystalline solid
N,N′-Propargyl-N,N′-tert-butoxycarbonyl-1,2-diamino-
ethane (2). 1 (500 mg, 1.92 mmol) was suspended in dry acetonitrile
(20 mL, distilled over CaH2 in house) and sodium tert-butoxide (462
mg, 4.80 mmol, 2.5 equiv) was added. The white suspension colored
immediately to yellow. The suspension was cooled (0 °C) and
propargyl bromide (1242 μL, 0.5 mmol, 6 equiv) in dry acetonitrile (5
mL) was added dropwise over 5−10 min. The reaction mixture
darkened to light orange during the addition. Tetrabutyl ammonium
iodide (TBAI) (1.4 g, 2 equiv) was added and the reaction mixture was
left to stir, warming to room temperature as the ice water bath melted.
TLC revealed the 100% consumption of starting material (Rf: 0.45,
TLC in 40% EtOAc in hexanes) in favor of product (Rf: 0.65) after 48
h. Sodium tert-butoxide was filtered out and rinsed well with
acetonitrile, and the combined filtrate was concentrated in vacuo to
dryness. The resulting orange oil was dissolved in dichloromethane (5
mL) and the mixture was filtered over a short silica plug, rinsing well
with dichloromethane (50 mL) to elute the toxic propargyl bromide.
The filtrate was set aside and the silica was washed with ethyl acetate
(50 mL). The ethyl acetate eluent was concentrated in vacuo to
dryness and the resulting yellow oil was purified by column
chromatography (CombiFlash Rf automated column system; 24 g
HP silica; eluted with a gradient of 0−50% EtOAc in hexanes) to
afford the product 2 as a yellow solid (485 mg, 75%, Rf: 0.65, TLC in
40% EtOAc in hexanes/KMnO4 staining). 1H NMR (300 MHz,
CDCl3, 25 °C) δ: 4.02 (d, 4H), 3.44 (s, 4H), 2.18 (s, 2H), 1.42 (s,
18H). 13C NMR (75 MHz, CDCl3, 25 °C) δ: 154.49, 80.19, 79.22,
71.67, 43.98, 36.40, 28.08. HR-ESI-MS calcd. for [C18H28N2O4+H]+:
337.2127; found: 337.2127, [M+H]+.
1
(272 mg, 44% based on 2 TFA per molecule). H NMR (300 MHz,
MeOD, 25 °C) δ: 8.18 (s, 2H), 7.87 (m, 4H), 7.49 (d, 2H), 7.24 (s,
10H), 5.49 (s, 4H), 4.39 (s, 8H), 3.56 (s, 4H). 13C NMR (75 MHz,
MeOD, 25 °C) δ: 167.30, 155.86, 148.58, 140.68, 139.99, 136.54,
130.17, 129.80, 129.39, 128.17, 127.66, 125.80, 58.24, 55.14, 50.98,
50.20. The synthesis was continued to form the HCl salt of H2azapa.
5·TFA (272 mg, 0.30 mmol) was dissolved in acetonitrile (3 mL). 0.1
M HCl (10 mL) was added, and the resulting solution was
concentrated in vacuo to dryness. This procedure was repeated 4−5
times to afford the white solid 5·2HCl·1.5H2O which was dried under
high vacuum for several days before use (removal of TFA confirmed
via 19F NMR). NMR structural assignments were confirmed using
two-dimensional NMR experiments (Supporting Information, Figures
N,N′-Propargyl-1,2-diaminoethane (3). To a solution of 2 (134
mg, 0.40 mmol) in dichloromethane (2 mL) was added trifluoroacetic
acid (TFA, 1 mL), and the yellow solution darkened to orange. The
reaction mixture was stirred for 1 h at ambient temperature and then
concentrated in vacuo to dryness. The resulting brown oil was
dissolved in saturated NaHCO3 (2 mL), and the resulting aqueous
layer was extracted using dichloromethane (25 × 1 mL) until the
extractions stained only faintly by TLC with KMnO4 staining. The
combined organic layers were dried (Na2SO4), filtered, and
concentrated in vacuo to dryness. The resulting orange oil 3 was
carried onto the next step without further purification (47 mg, 87%).
1H NMR (300 MHz, CDCl3, 25 °C) δ: 3.45 (s, 4H), 2.85 (s, 4H), 2.22
1
S13−S14). H NMR (600 MHz, D2O, 25 °C) δ: 7.57 (d, 2H; H3),
7.53 (s, 2H; H5), 7.41 (t, 2H, H4), 6.90 (m, 12H; H11, PhH), 5.13 (s,
4H: H12), 3.37 (s, 4H: H9), 3.34 (s, 4H, H7), 2.26 (s, 4H, H8). 13C
NMR (150 MHz, D2O, 25 °C) δ: 172.67 (C1), 157.18 (C10), 153.12
(C4), 143.63 (C6), 138.09 (C2), 134.90 (C13), 128.98 (C15), 128.58
(C16), 128.00 (C14), 125.17 (C11), 124.85 (C5), 122.49 (C3), 59.73
(C9), 53.73 (C12), 50.37 (C8), 48.42 (C7). HR-ESI-MS calcd. for
[C36H36N10O4+H]+: 673.2999; found: 673.3000, [M+H]+. Anal.
Calcd. (found) for C36H36N10O4·2HCl·1.5H2O: C, 55.96 (56.06); H,
5.35 (5.23); N, 18.13 (18.11).
(s, 2H), 1.77 (s, 2H). 13C NMR (75 MHz, CDCl3, 25 °C) δ: 82.07,
71.20, 47.49, 37.82. HR-ESI-MS calcd. for [C8H12N2+H]+: 137.1079;
found: 137.1076, [M+H]+.
N,N′-[6-(Methoxycarbonyl)pyridin-2-yl]methyl-N,N′-prop-
argyl-1,2-diaminoethane (4). To a solution of 3 (196 mg, 1.43
mmol) in dry acetonitrile (9 mL, distilled over CaH2) was added
methyl 6-(bromomethyl)picolinate28 (695 mg, 3.02 mmol, 2.1 equiv)
and sodium carbonate (320 mg, 3.02 mmol, 2.1 equiv). The reaction
was stirred at ambient temperature. Reaction monitoring by TLC
revealed the formation and consumption of a singly alkylated
intermediate (Rf: 0.50, TLC in 10% MeOH in DCM/UV activity)
and formation of product (Rf: 0.55). After 48 h, sodium carbonate was
filtered out, and the filtrate was concentrated in vacuo to dryness. The
resulting brown oil was purified by column chromatography
(CombiFlash Rf automated column system; 24 g HP silica; eluted
with a gradient of 0−10% MeOH in DCM) to afford the product 4 as
a yellow solid (387 mg, 62%, Rf: 0.55, TLC in 10% MeOH in DCM/
Metal Complexation Experiments: General Procedure.
H2azapa·2HCl·1.5H2O (∼0.015 mmol) was suspended in 0.1 M
HCl (1.5 mL), and the appropriate metal chloride or perchlorate (1.2
equiv) was added. The pH was adjusted to 4−5 using 0.1 M NaOH,
and then the solution was stirred at 60 °C. Any remaining solid
dissolved upon heating. The reaction was stirred overnight (24 h),
maintaining the same temperature.
[Cu(azapa)]. H2azapa·2HCl·1.5H2O (13.8 mg, 0.018 mmol) was
reacted with copper(II) chloride dihydrate (3.7 mg, 0.0215 mmol, 1.2
equiv). After 24 h, a light blue precipitate was filtered out and purified
by semipreparative HPLC (gradient: A: water, B: MeCN, 0−100% B
over 25 min, tR (broad) 16.2−17.5 min) to afford the product
1
[Cu(azapa)] as
a green solid. HR-ESI-MS calcd. for
UV activity). H NMR (300 MHz, CDCl3, 25 °C) δ: 7.96 (d, 2H),
[C36H3463CuN10O4+Na]+: 756.1958; found: 756.1951, [M+Na]+.
Anal. Calcd. (found) for C36H34CuN10O4·1.5H2O: C, 56.80 (56.88);
H, 4.90 (4.81); N, 18.40 (18.38).
7.71 (m, 4H), 3.94 (s, 6H), 3.90 (s, 4H), 3.42 (s, 4H), 2.71 (s, 4H),
2.18 (s, 2H). 13C NMR (75 MHz, CDCl3, 25 °C) δ: 165.82, 160.12,
147.37, 137.34, 125.99, 123.60, 78.51, 73.34, 60.17, 52.86, 51.03,
42.84. HR-ESI-MS calcd. for [C24H26N4O4+H]+: 435.2032; found:
435.2030, [M+H]+.
N,N′-[1-Benzyl-1,2,3-triazole-4-yl]methyl-N,N′-[6-(carboxy)-
pyridin-2-yl]-1,2-diaminoethane (H2azapa·2HCl·1.5H2O)
(5·2HCl·1.5H2O). 4 (300 mg, 0.69 mmol, 1 equiv) was suspended
in 1:1 tert-butanol/water, and benzyl azide (172 μL, 1.38 mmol, 2
equiv) was carefully added using precautionary blast shield protection
[Ga(azapa)][ClO4]. H2azapa·2HCl·1.5H2O (9.7 mg, 0.013 mmol)
was reacted with gallium(III) perchlorate hexahydrate (7.2 mg, 0.015
mmol, 1.2 equiv). After 24 h, a white crystalline precipitate was filtered
out to afford [Ga(azapa)][ClO4]. NMR structural assignments were
confirmed using two-dimensional NMR experiments (Supporting
1
Information, Figures S18 and S19). H NMR (600 MHz, DMSO-d6,
25 °C) δ: 8.65 (t, 2H; H4), 8.34 (m, 4H; H3, H11), 8.20 (d, 2H; H5),
12586
dx.doi.org/10.1021/ic302225z | Inorg. Chem. 2012, 51, 12575−12589

Inorganic Chemistry
Article
7.32 (m, 10H; PhH), 5.65 (q, 4H; H12, H12′), 4.74 and 4.28 (2d, 4H;
H2azapa required heating at 37 °C and frequent vortexing for 1−2 h to
dissolve. Aliquots of each chelator stock solution were transferred to
Corning 2.0 mL self-standing microcentrifuge tubes containing ∼1.2
mCi of 177Lu to a concentration of ∼182 mM, and made up to 1 mL
with NaOAc buffer (10 mM, pH 5.0). H2azapa and DTPA were
allowed to radiolabel at ambient temperature for 10 min, and DOTA
was radiolabeled for 1 h at 90 °C. Radiometal complexes were then
evaluated using radio-HPLC (linear gradient A: 0.1% TFA in H2O, B:
MeCN, 0−80% B over 30 min): [177Lu(azapa)]+ tR = 19.3 min, >99%;
2
2
H7, H7′, J = 18 Hz), 4.00 and 3.92 (2d, 4H; H9, H9′, J = 15 Hz),
3.10 (m, 4H; H8, H8′). 13C NMR (150 MHz, DMSO-d6, 25 °C) δ:
162.35 (C1), 150.57 (C6), 146.52 (C4), 143.58 (C2), 137.69 (C10),
135.74 (C13), 128.84 (PhC), 128.43 (C5), 128.24 (PhC), 127.94
(PhC), 127.10 (C11), 123.47 (C3), 56.31 (C7), 52.99 (C12), 47.80
(C8), 46.01 (C9). HR-ESI-MS calcd. for [C36H3469GaN10O4]+:
739.2020; found: 739.2014, [M]+.
[In(azapa)(H2O)][ClO4]. H2azapa·2HCl·1.5H2O (9.8 mg, 0.013
mmol) was reacted with indium(III) perchlorate hexahydrate (8.5 mg,
0.015 mmol, 1.2 equiv). After 24 h, a white crystalline precipitate was
[
177Lu(DOTA)]− tR = 6.8 min, >99%, [177Lu(DTPA)]2− tR = 5.2 min,
>99%. 64Cu labeling was performed in a similar manner with H2azapa
and DOTA at the same chelator concentrations previously used for
labeling with 177Lu (182 mM) and with ∼800−1000 μCi of 64Cu.
DOTA was radiolabeled at 90 °C for 60 min, and H2azapa in 10 min at
ambient temperature, and both were evaluated by RP-HPLC
[64Cu(azapa)] tR = 20.9 min, >99%; [64Cu(DOTA)]2− tR = 7.1 min,
>99%.
1
filtered out to afford the product [In(azapa)(H2O)][ClO4]. H NMR
(400 MHz, DMSO-d6, 25 °C) δ: 8.54−6.87 (br m, 18H), 5.49 (br s,
1
4H) 4.80−4.18 (m, 4H), 2.84 (s, 2H), 2.24 (s, 2H). H NMR (400
MHz, DMSO-d6, 135 °C) δ: 8.19 (m, 4H), 7.98 (s, 2H), 7.57 (s, 2H),
7.29 (m, 6H), 7.12 (m, 4H), 5.42 (s, 4H), 3.93 (s, 4H), 3.66 (s, 4H),
2.59 (s, 4H). HR-ESI-MS calcd. for [C36H34115InN10O4]+: 785.1803;
found: 785.1806, [M]+.
67Ga/111In Mouse Serum Competition Experiments. Frozen
mouse serum was allowed to thaw for 30 min, and 750 μL of
[67Ga(azapa)]+ solution was added in triplicate to 20 mL sterile vials
containing 1250 μL of thawed mouse serum and 500 μL of PBS. The
solutions sat at ambient temperature, and 800 μL aliquots were
removed from each vial at 15 min, 1 h, and 2 h time points and diluted
to a final volume of 2.5 mL with PBS in sterile vials. A 500 μL aliquot
of [111In(azapa)]+ was added in triplicate to 20 mL sterile vials
containing 750 μL of thawed mouse serum and 250 μL of PBS. The
solutions were allowed to sit at ambient temperature, and 750 μL
aliquots were removed at 1 and 24 h time points, and diluted to a final
volume of 2.5 mL with PBS in sterile vials. The activities were counted,
and the diluted mouse serum competitions were immediately loaded
onto conditioned PD-10 size-exclusion columns. The 2.5 mL loading
volumes were eluted and discarded into dedicated 111In or 67Ga
radioactive waste containers, and the columns were then eluted with
an additional 3.5 mL of PBS, which was collected into fresh 20 mL
sterile vials. The activities of the eluents, which contained trans-
chelated metal ions bound/associated with serum proteins, were
counted and compared with the activity values obtained exactly prior
to column loading to determine the percent radiometal ions that had
been retained on the column as chelate-bound activity.
[Lu(azapa)][ClO4]. H2azapa·2HCl·1.5H2O (9.8 mg, 0.013 mmol)
was reacted with lutetium(III) chloride hexahydrate (6.1 mg, 0.015
mmol, 1.2 equiv). After 4 h the product was confirmed via mass
spectrometry, and the solvent was removed in vacuo to yield
1
[Lu(azapa)][ClO4]. H NMR (300 MHz, D2O, 25 °C) δ: 8.09 (br
m, 2H), 7.98 (br m, 2H), 7.89 (s, 2H), 7.49 (br m, 2H), 7.31 (br m,
6H), 7.09 (br m, 4H), 5.32 (s, 4H) 4.14−3.91 (m, 6H), 3.65 (br m,
2H), 2.80−2.72 (m, 4H). 13C NMR (400 MHz, D2O, 25 °C) δ:
172.52, 157.86, 150.34, 142.96, 134.38, 129.25, 129.03, 128.4, 125.63,
125.51, 124.06, 59.70, 54.73, 54.41, 52.99, 49.08. HR-ESI-MS calcd.
for [C36H34175LuN10O4]+: 845.2172; found: 845.2166 [M]+.
X-ray Crystallography. A small blue prism crystal of [Cu(azapa)]
having approximate dimensions of 0.10 × 0.41 × 0.45 mm was grown
by slow evaporation of a saturated solution in 1:1 MeCN/H2O. A
brown plate crystal of [In(azapa)(H2O)][ClO4]·2.5H2O having
approximate dimensions of 0.10 × 0.22 × 0.27 mm was grown by
slow evaporation of a saturated solution in 1:1 MeOH/DMSO. The
crystals were mounted onto respective glass fibers, and measurements
were made on a Bruker Apex DUO diffractometer at −183 °C with
graphite-monochromated Mo−Kα radiation. All data were collected
and integrated using the Bruker SAINT63 software package. Data were
corrected for adsorption effects, and Lorentz and polarization effects
using the SADABS64 program. The structures were solved by direct
methods using the SIR9765 software and refined using SHELXL-9766
via the WinGX67 interface. [In(azapa)][ClO4] crystallizes with 3
molecules of water in the crystal lattice. A summary of crystallographic
data is presented in Table 1 for [In(azapa)(H2O)][ClO4] and in Table
2 for [Cu(azapa)]. CIF files for the two complexes as well as relevant
bond lengths and angles are found in the Supporting Information.
67Ga/111In Radiolabeling. For 67Ga radiolabeling experiments, a 1
mg/mL stock solution of H2azapa was made by dissolving 1.8 mg in
1.8 mL NaOAc buffer (10 mM, pH 4) with mild heat and vortex-
mixing. An aliquot of chelator stock (750 μL) was transferred to a 20
mL sterile vial, and 67GaCl3 (10 μL, ∼5 mCi) solution was added
followed by NaOAc buffer (1.785 mL, 10 mM, pH 4) up to a total
chelator concentration of 0.3 mg/mL. 111In labeling was performed
similarly, with 111InCl3 (15 μL, ∼6 mCi) and NaOAc buffer (1.240
mL, 10 mM, pH 5) being added to an aliquot of chelator stock
solution (495 μL). The mixtures were agitated briefly and then
allowed to react for 10 min at ambient temperature. The radiometal
complex solutions were analyzed by analytical RP-HPLC (linear
gradient A: 0.05% TFA, B: MeCN, 0−100% B over 20 min) to
determine radiolabeling yields through peak integration: [67Ga-
(azapa)]+ tR = 12.8 min, >99%; [111In(azapa)]+ tR = 11.4 min,
>99%. The retention times for the radioactive metal complexes
determined by the Raytest NaI (Tl) detector were confirmed by
analyzing the fully characterized nonradioactive metal complexes by
RP-HPLC using the PDA detector: [Ga(azapa)]+ tR = 12.7 min;
[In(azapa)]+ tR = 11.4 min.
177Lu/64Cu Blood Serum Competition Experiments. Frozen
human blood serum thawed for 30 min, and 750 μL aliquots were
transferred to 2.0 mL Corning centrifuge vials. 300 μL of each
177Lu(chelator) was transferred to the blood serum, along with 450 μL
of PBS to a total volume of 1.5 mL (in triplicate for each chelator).
The final 177Lu(chelate) concentration present in serum was ∼36 mM.
Serum competition samples were then incubated at 37 0.1 °C with
constant agitation (550 rpm) and analyzed via PD-10 size-exclusion
column elution (filters MW < 5000 Da) at 1.5 and 24 h time points
and counted using a Capintec CRC-15R dose calibrator. 750 μL of
each serum/177Lu(chelator) competition solution (in triplicate) was
removed from the competition vial, diluted to 2.5 mL with PBS, and
counted. The diluted aliquot of serum competition mixture was loaded
onto a conditioned PD-10 column. The loading volume (2.5 mL) was
eluted into radioactive waste, and then an additional 3.5 mL of PBS
was loaded, collected, and counted in the dose calibrator as the serum-
bound 177Lu (nonchelate bound). Percent stability was reported as a
percentage of 177Lu still chelate-bound and not associated with serum
proteins (MW < 5000 Da). Human serum competitions with 64Cu
were performed in an identical fashion as described above for 177Lu,
but with time points of 1 and 20 h.
64Cu Biodistribution Studies. Biodistribution studies evaluated
the basic stability and clearance of the 64Cu complexes of the
commonly used chelator DOTA, and the novel entrant H2azapa.
Although the chelator NOTA is also widely used with 64Cu, and is
known to be superior to DOTA in terms of in vivo stability and
radiolabeling kinetics, DOTA was used as an initial first benchmark
because of its wide availability and common use (currently used in 2 of
177Lu/64Cu Radiolabeling. For 177Lu experiments, the chelators
H2azapa, DTPA, and DOTA were all dissolved up to 1 mg/mL in
NaOAc buffer (10 mM, pH 5.0) as stock solutions. DTPA and
3
64Cu-based clinical trials), and its superior properties with the
isotopes 111In and 177Lu used in this study.9,19,68 DOTA and H2azapa
were radiolabeled with 64Cu under the same conditions as described
12587
dx.doi.org/10.1021/ic302225z | Inorg. Chem. 2012, 51, 12575−12589

Inorganic Chemistry
Article
above, and prepared for biodistribution studies in healthy nude
athymic mice (female, 6−8 weeks old). The 64Cu(chelate) complexes
were purified via RP-HPLC, concentrated in vacuo to dryness, and
then suspended in 2:1 sterile saline 0.9% NaCl:PBS (pH 7) to a
concentration and dose of ∼30 μCi in 200 μL per mouse (specific
activity ∼10 μCi/μg, ∼ 3 μg 64Cu(chelate) per mouse). Each mouse
was intravenously injected through the tail vein and sacrificed by CO2
inhalation at time points of 15 min, 1 h, 4 h, and 24 h (n = 4). Tissues
collected after sacrifice included blood, heart, lungs, liver, spleen,
kidneys, large intestine, small intestine, muscle, bone (femur), and
skin. All organs were rinsed in water after removal and air-dried for 5
min. Tissues were weighed and counted (calibrated for 64Cu), the
counts were background- and decay-corrected from the time of
injection and then converted to the percentage of injected dose (%ID)
per gram of organ tissue (%ID/g). The 64Cu counts measured in each
organ were converted to activity using a calibration curve created from
known standards of 64Cu.
ACKNOWLEDGMENTS
■
We acknowledge Nordion (Canada) and the Natural Sciences
and Engineering Research Council (NSERC) of Canada for
grant support, NSERC for CGS-M/CGS-D fellowships
(E.W.P.) and a summer USRA scholarship (G.A.B), the
University of British Columbia for 4YF fellowships (E.B. and
E.W.P.), and Kuntalkumar Sevak from MSKCC (Lewis Lab)
for skillful tail-vein injections of [64Cu(azapa)]. C.O. acknowl-
edges the Canada Council for the Arts for a Killam Research
Fellowship (2011−2013).
REFERENCES
■
(1) Zeglis, B. M.; Lewis, J. S. Dalton Trans. 2011, 40, 6168−6195.
(2) Wadas, T. J.; Wong, E. H.; Weisman, G. R.; Anderson, C. J.
Chem. Rev. 2010, 110, 2858−2902.
Small-Animal 64Cu-PET Imaging. PET imaging was performed
using a micro-PET R4 rodent scanner (Concord Microsystems) to
visually confirm biodistribution results and monitor clearance of 64Cu.
Two healthy, nude athymic mice (female, 6−8 weeks old) were
administered doses of ∼300 μCi in 200 μL of 2:1 sterile saline (0.9%
NaCl):PBS (pH 7) via intravenous tail vein injection (specific activity
∼10 μCi/μg, ∼30 μg of [64Cu(azapa)] per mouse). Approximately 5
min prior to PET image acquisition, mice were anesthetized via
inhalation of 2% isoflurane (Baxter Healthcare, Deerfield, IL)/oxygen
gas mixture, placed on the scanner bed, and anesthesia was maintained
during imaging using a reduced 1% isoflurane/oxygen mixture. Images
were acquired at time points 1 h, 4 h, and 24 h, with PET data being
recorded via static scans with a minimum of 10 million coincident
events (30−60 min). An energy window of 350−700 keV and a
coincidence timing window of 6 ns were used. Data were sorted into
2D histograms by Fourier rebinning, and transverse images were
reconstructed by filtered back-projection (FBP) into a 128 × 128 × 63
(0.72 × 0.72 × 1.3 mm3) matrix. The image data were normalized to
correct for nonuniformity of response of the PET, dead-time count
losses, positron branching ratio, and physical decay to the time of
injection, but no attenuation, scatter, or partial-volume averaging
correction was applied. The counting rates in the reconstructed images
were converted to activity concentrations (percentage injected dose
[%ID] per gram of tissue) by use of a system calibration factor derived
from the imaging of a mouse-sized water-equivalent phantom
containing 64Cu. Images were analyzed using ASIPro VM software
(Concorde Microsystems).
(3) Jones-Wilson, T. M.; Motekaitis, R. J.; Sun, Y.; Anderson, C. J.;
Martell, A. E.; Welch, M. J. Nucl. Med. Biol. 1995, 22, 859−868.
(4) Laforest, R.; Dehdashti, F.; Lewis, J. S.; Schwarz, S. W. Eur. J.
Nucl. Med. Mol. Imaging 2005, 32, 764−770.
(5) Blower, P. J.; Lewis, J. S.; Zweit, J. Nucl. Med. Biol. 1996, 23,
957−980.
(6) Anderson, C. J.; Ferdani, R. Cancer Biother. Radiopharm. 2009,
24, 379−393.
(7) Novak-Hofer, I.; Schubiger, A. Eur. J. Nucl. Med. Mol. Imaging
2002, 29, 821−830.
(8) McQuade, P.; Miao, Y.; Yoo, Y.; Quinn, T. P.; Welch, M. J.;
Lewis, J. S. J. Med. Chem. 2005, 48, 2985−2992.
(9) Boswell, C. A.; Sun, X.; Niu, W.; Weisman, G. R.; Wong, E. H.;
Rheingold, A. L.; Anderson, C. J. J. Med. Chem. 2004, 47, 1465−1474.
(10) Sun, X.; Wuest, M.; Weisman, G. R.; Wong, E. H.; Reed, D. P.;
Boswell, C. A.; Motekaitis, R.; Martell, A. E.; Welch, M. J.; Anderson,
C. J. J. Med. Chem. 2002, 45, 469−477.
(11) Sprague, J. E.; Peng, Y.; Sun, X.; Weisman, G. R.; Wong, E. H.;
Achilefu, S.; Anderson, C. J. Clin. Cancer. Res. 2004, 10, 8674−8682.
(12) Clifford, T.; Boswell, C. A.; Biddlecombe, G. B.; Lewis, J. S.;
Brechbiel, M. W. J. Med. Chem. 2006, 49, 4297−4304.
(13) Lee, F. T.; Mountain, A. J.; Kelly, M. P.; Hall, C.; Rigopoulos,
A.; Johns, T. G.; Smyth, F. E.; Brechbiel, M. W.; Nice, E. C.; Burgess,
A. W.; Scott, A. M. Clin. Cancer Res. 2005, 11, 7080s−7086s.
́
(14) Vandenbulcke, K.; Vos, F.; Offner, F.; Philippe, J.; Apostolidis,
C.; Molinet, R.; Nikula, T.; Bacher, K.; Gelder, V.; Vral, A.; Lahorte,
C.; Thierens, H.; Dierckx, R.; Slegers, G. Eur. J. Nucl. Med. Mol.
Imaging 2003, 30, 1357−1364.
ASSOCIATED CONTENT
(15) Camera, L.; Kinuya, S.; Garmestani, K.; Wu, C.; Brechbiel, M.
W.; Pai, L. H.; McMurry, T. J.; Gansow, O. A.; Pastan, I.; Paik, C. H.;
Carrasquillo, J. A. J. Nucl. Med. 1994, 35, 882−889.
■
S
* Supporting Information
Crystallographic information files (CIF) for the X-ray crystal
structures along with relevant distance and bond angle data,
1H/13C NMR spectra of all synthesized compounds, two-
dimensional HSQC and HMBC NMR spectra of H2azapa and
[Ga(azapa)]+, and HPLC radio-traces confirming quantitative
radiolabeling of H2azapa with all four radiometals. This material
is available free of charge via the Internet at http://pubs.acs.org.
(16) Nayak, T.; Regino, C.; Wong, K.; Milenic, D.; Garmestani, K.;
Baidoo, K.; Szajek, L.; Brechbiel, M. Eur. J. Nucl. Med. Mol. Imaging
2010, 37, 1368−1376.
(17) 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, 4829−4833.
(18) Wei, L.; Ye, Y.; Wadas, T. J.; Lewis, J. S.; Welch, M. J.; Achilefu,
S.; Anderson, C. J. Nucl. Med. Biol. 2009, 36, 277−285.
(19) 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,
23 (5), 1029−1039.
AUTHOR INFORMATION
■
Corresponding Author
(20) Ferdani, R.; Stigers, D. J.; Fiamengo, A. L.; Wei, L.; Li, B. T. Y.;
Golen, J. A.; Rheingold, A. L.; Weisman, G. R.; Wong, E. H.;
Anderson, C. J. Dalton Trans. 2012, 41, 1938−1950.
(21) Stigers, D. J.; Ferdani, R.; Weisman, G. R.; Wong, E. H.;
Anderson, C. J.; Golen, J. A.; Moore, C.; Rheingold, A. L. Dalton
Trans. 2009, 39, 1699−1701.
(22) Kotek, J.; Lubal, P.; Hermann, P.; Cisarova, I.; Lukes, I.; Godula,
T.; Svobodova, I.; Taborsky, P.; Havel, J. Chem.Eur. J. 2003, 9, 233−
248.
*Phone: (604) 222-7527 (M.J.A.), (604) 822-4449 (C.O.).
Fax: (604) 222-1074 (M.J.A.), (604) 822-2847 (C.O.). E-mail:
adam@triumf.ca (M.J.A.), orvig@chem.ubc.ca (C.O.).
Author Contributions
∥These authors contributed equally to this work.
Notes
The authors declare no competing financial interest.
12588
dx.doi.org/10.1021/ic302225z | Inorg. Chem. 2012, 51, 12575−12589

Inorganic Chemistry
Article
(23) Nguyen, K.; Parry, J. J.; Rogers, B. E.; Anderson, C. J. Nucl. Med.
Biol. 2012, 39, 187−197.
(24) Boros, E.; Ferreira, C. L.; Cawthray, J. F.; Price, E. W.; Patrick,
B. O.; Wester, D. W.; Adam, M. J.; Orvig, C. J. Am. Chem. Soc. 2010,
132, 15726−15733.
(25) Ferreiros-Martinez, R.; Esteban-Gomez, D.; Platas-Iglesias, C.;
Blas, A. d.; Rodriguez-Blas, T. Dalton Trans. 2008, 42, 5754−5765.
(26) Boros, E.; Ferreira, C. L.; Yapp, D. T. T.; Gill, R. K.; Price, E.
W.; Adam, M. J.; Orvig, C. Nucl. Med. Biol. 2012, 39, 785−794.
(27) Boros, E.; Cawthray, J. F.; Ferreira, C. L.; Patrick, B. O.; Adam,
M. J.; Orvig, C. Inorg. Chem. 2012, 51, 6279−6284.
(54) Anderson, C. J.; Connett, J. M.; Schwarz, S. W.; Rocque, P. A.;
Guo, L. W.; Philpott, G. W.; Zinn, K. R.; Meares, C. F.; Welch, M. J. J.
Nucl. Med. 1992, 33, 1685−1691.
(55) Zimmermann, K.; Grunberg, J.; Honer, M.; Amctamcy, S.;
Schubiger, P. A.; Novak-Hofer, I. Nucl. Med. Biol. 2003, 30, 417−427.
(56) Bass, L. A.; Wang, M.; Welch, M. J.; Anderson, C. J. Bioconjugate
Chem. 2000, 11, 527−532.
(57) Cole, W. C.; DeNardo, S. J.; Meares, C. F.; McCall, M. J.;
DeNardo, G. L.; Epstein, A. L.; O’Brien, H. A.; Moi, M. K. Int. J.
Radiat. Appl. Instrum., Part B 1986, 13, 363−368.
(58) Packard, A. B.; Kronauge, J. F.; Barbarics, E.; Kiani, S.; Treves, S.
T. Nucl. Med. Biol. 2002, 29, 289−294.
(59) John, E. K.; Green, M. A. J. Med. Chem. 1990, 33, 1764−1770.
(60) Kelly, J. D.; Forster, A. M.; Higley, B.; Archer, C. M.; Booker, F.
S.; Canning, L. R.; Chiu, K. W.; Edwards, B.; Gill, H. K.; McPartlin,
M.; Nagle, K.; Latham, I.; Pickett, R.; Storey, A.; Webbon, P. J. Nucl.
Med. 1993, 34, 222−227.
(61) Ando, A.; Ando, I.; Hiraki, T.; Hisada, K. Int. J. Radiat. Appl.
Instrum., Part B 1989, 16, 57−80.
(28) Price, E. W.; Cawthray, J. F.; Bailey, G. A.; Ferreira, C. L.; Boros,
E.; Adam, M. J.; Orvig, C. J. Am. Chem. Soc. 2012, 134, 8670−8683.
(29) Boros, E.; Lin, Y.-H. S.; Ferreira, C. L.; Patrick, B. O.; Hafeli, U.
O.; Adam, M. J.; Orvig, C. Dalton Trans. 2011, 40, 6253−6259.
(30) Boros, E.; Ferreira, C. L.; Patrick, B. O.; Adam, M. J.; Orvig, C.
Nucl. Med. Biol. 2011, 38, 1165−1174.
(31) Mindt, T. L.; Struthers, H.; Brans, L.; Anguelov, T.;
Schweinsberg, C.; Maes, V.; Tourwe,
Soc. 2006, 128, 15096−15097.
́
D.; Schibli, R. J. Am. Chem.
(62) McCarthy, D. W.; Shefer, R. E.; Klinkowstein, R. E.; Bass, L. A.;
Margeneau, W. H.; Cutler, C. S.; Anderson, C. J.; Welch, M. J. Nucl.
Med. Biol. 1997, 24, 35−43.
(63) SAINT, Version 7.68A; Bruker AXS Inc.: Madison, WI, 1997−
2010.
(32) Struthers, H.; Spingler, B.; Mindt, T. L.; Schibli, R. Chem.Eur.
J. 2008, 14, 6173−6183.
(33) Salvatore, R. N.; Shin, S. I.; Flanders, V. L.; Jung, K. W.
Tetrahedron Lett. 2001, 42, 1799−1801.
(64) Riss, P. J.; Kroll, C.; Nagel, V.; Rosch, F. Bioorg. Med. Chem. Lett.
̈
(34) Nakon, R.; Angelici, R. J. J. Am. Chem. Soc. 1973, 95, 3170−
3174.
2008, 18, 5364−5367.
(65) Altomare, A.; Burla, M. C.; Camalli, M.; Cascarano, G. L.;
Giacovazzo, C.; Guagliardi, A.; Moliterni, A. G. G.; Polidori, G.;
Spagna, R. SIR97; J. Appl. Crystallogr., Published Online 1999.
(66) SHELXTL, Version 2008/4; Bruker AXS Inc.: Madison, WI,
2008;
(35) Liu, S.; Pietryka, J.; Ellars, C. E.; Edwards, D. S. Bioconjugate
Chem. 2002, 13, 902−913.
(36) Hsieh, W.-Y.; Liu, S. Inorg. Chem. 2004, 43, 6006−6014.
(37) Liu, S.; He, Z.; Hsieh, W.-Y.; Fanwick, P. E. Inorg. Chem. 2003,
42, 8831−8837.
(67) Farrugia, L. J. WinGX; J. Appl. Crystallogr., Published Online
1999.
(68) Schneider, D. W.; Heitner, T.; Alicke, B.; Light, D. R.; McLean,
K.; Satozawa, N.; Parry, G.; Yoo, J.; Lewis, J. S.; Parry, R. J. Nucl. Med.
2009, 50, 435−443.
(38) Bartholoma, M. D.; Louie, A. S.; Valliant, J. F.; Zubieta, J. Chem.
Rev. 2010, 110, 2903−2920.
(39) Malyarick, M. A.; Ilyuhin, A. B.; Petrosyants, S. P. Main Group
Met. Chem. 1994, 17, 707−718.
(40) Maecke, H. R.; Riesen, A.; Ritter, W. J. Nucl. Med. 1989, 30,
1235−1239.
(41) Mindt, T. L.; Schweinsberg, C.; Brans, L.; Hagenbach, A.;
Abram, U.; Tourwe,
2009, 4, 529−539.
(42) Struthers, H.; Mindt, T. L.; Schibli, R. Dalton Trans. 2010, 39,
675−696.
́
D.; Garcia-Garayoa, E.; Schibli, R. ChemMedChem
(43) Tamanini, E.; Rigby, S. E. J.; Motevalli, M.; Todd, M. H.;
Watkinson, M. Chem.Eur. J. 2009, 15, 3720−3728.
(44) Lebedev, A. Y.; Holland, J. P.; Lewis, J. S. Chem. Commun. 2010,
46, 1706−1708.
(45) Bratsos, I.; Urankar, D.; Zangrando, E.; Genova-Kalou, P.;
Kosmrlj, J.; Alessio, E.; Turel, I. Dalton Trans. 2011, 40, 5188−5199.
(46) Maisonial, A.; Serafin, P.; Traikia, M.; Debiton, E.; Thery, V.;
Aitken, D. J.; Lemoine, P.; Viossat, B.; Gautier, A. Eur. J. Inorg. Chem.
2008, 2008, 298−305.
(47) Bastero, A.; Font, D.; Pericas, M. A. J. Org. Chem. 2007, 72,
2460−2468.
(48) Mirick, G. R.; O’Donnell, R. T.; DeNardo, S. J.; Shen, S.;
Meares, C. F.; DeNardo, G. L. Nucl. Med. Biol. 1999, 26, 841−845.
(49) Connett, J. M.; Anderson, C. J.; Guo, L.-W.; Schwarz, S. W.;
Zinn, K. R.; Rogers, B. E.; Siegel, B. A.; Philpott, G. W.; Welch, M. J.
Proc. Natl. Acad. Sci. U.S.A. 1996, 93, 6814−6818.
(50) DeNardo, G. L.; Kukis, D. L.; Shen, S.; DeNardo, D. A.; Meares,
C. F.; DeNardo, S. J. Clin. Cancer Res. 1999, 5, 533−541.
(51) Ferreira, C. L.; Yapp, D. T. T.; Crisp, S.; Sutherland, B. W.; Ng,
S. S. W.; Gleave, M.; Bensimon, C.; Jurek, P.; Kiefer, G. E. Eur. J. Nucl.
Med. Mol. Imaging 2010, 37, 2117−2126.
(52) Wadas, T. J.; Wong, E. H.; Weisman, G. R.; Anderson, C. J.
Curr. Pharm. Des. 2007, 13, 3−16.
(53) Rogers, B. E.; Anderson, C. J.; Connett, J. M.; Guo, L. W.;
Edwards, B.; Sherman, E. L. C.; Zinn, K. R.; Welch, M. J. Bioconjugate
Chem. 1996, 7, 511−522.
12589
dx.doi.org/10.1021/ic302225z | Inorg. Chem. 2012, 51, 12575−12589