Investigation into the Use of a Diaminodihydroxyaryl Derivative of Ethylenediaminetetraacetic Acid (DAHA-EDTA) for Cu-64 PET Imaging and Radioimmunotherapy

Article abstract of DOI:10.1071/CH16647

A diaminodihydroxyaryl derivative of ethylenediaminetetraacetic acid (DAHA-EDTA) was synthesised in two steps and evaluated for Cu-64 radiolabelling of the B72.3 antibody. The ligand complexes Cu-64 rapidly in a pH range 4 to 7. The Cu-64 complex of the parent species N,N′-bis(carboxymethyl)-N,N′-bis(2-hydroxyacetanilido)-1,2-diaminoethane (DHA-EDTA) shows good stability in serum at 37°C for up to 72h. Conjugation of the Cu-64-DAHA-EDTA to the B72.3 antibody was achieved using 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDC) as the activating agent. The reaction conditions were optimized for protein concentration and molar ratio of Cu-64-DAHA-EDTA and EDC to antibody. The specific activity of the final [Cu-64-DAHA-EDTA]-B72.3 product was 49MBqmg^-1 at the end of synthesis. The biodistribution of [Cu-64-DAHA-EDTA]-B72.3 in LS174t tumour-bearing nude mice was monitored over a 24h period. Maximum tumour uptake (25.8±7.5% IDg^-1) was achieved at 16h and maintained at 24h (21.6±1.8% IDg^-1). Rapid clearance of the [Cu-64-DAHA-EDTA]-B72.3 from the blood resulted in good tumour-to-blood ratios (≈ 3.3) within a shorter period (6h) than previously reported with B72.3 whole antibody and the LS174t tumour bearing nude mouse model.

Full text of DOI:10.1071/CH16647

RESEARCH FRONT
CSIRO PUBLISHING
Aust. J. Chem. 2017, 70, 614–622
Full Paper
http://dx.doi.org/10.1071/CH16647
Investigation into the Use of a Diaminodihydroxyaryl
Derivative of Ethylenediaminetetraacetic Acid (DAHA-
EDTA) for Cu-64 PET Imaging and Radioimmunotherapy
A B C
, ,
B D
,
B
Martalena Ramli,
Peter F. Schmidt,
Nadine Di Bartolo,
B E F
, ,
and Suzanne V. Smith
A
Centre for Radioisotope and Radiopharmaceutical Technology, National Nuclear
Energy Agency, B. 11, Kawasan PUSPIPTEK, Serpong–Tangerang Selatan 15314,
Banten, Indonesia.
B
Australian Nuclear Science and Technology Organization, Locked Bag 2001,
Kirrawee DC, Sydney, NSW 2232, Australia.
C
School of Chemistry, F11, University of Sydney, Sydney, NSW 2006, Australia.
D
Q Biotics, Suite 3A, Level 1, Taringa Central, 165 Moggill Road,
Taringa, Qld 4068, Australia.
E
Idaho Accelerator Centre, Idaho State University, 1500 Alvin Ricken Drive,
Pocatello, ID 83201, USA.
F
Corresponding author. Email: Suzanneoznq@gmail.com
A diaminodihydroxyaryl derivative of ethylenediaminetetraacetic acid (DAHA-EDTA) was synthesised in two steps and
evaluated for Cu-64 radiolabelling of the B72.3 antibody. The ligand complexes Cu-64 rapidly in a pH range 4 to 7. The
Cu-64 complex of the parent species N,N0-bis(carboxymethyl)-N,N0-bis(2-hydroxyacetanilido)-1,2-diaminoethane
(DHA-EDTA) shows good stability in serum at 378C for up to 72 h. Conjugation of the Cu-64-DAHA-EDTA to
the B72.3 antibody was achieved using 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDC) as the activating agent.
The reaction conditions were optimized for protein concentration and molar ratio of Cu-64-DAHA-EDTA and EDC to
antibody. The specific activity of the final [Cu-64-DAHA-EDTA]-B72.3 product was 49 MBq mgꢀ1 at the end of
synthesis. The biodistribution of [Cu-64-DAHA-EDTA]-B72.3 in LS174t tumour-bearing nude mice was monitored over
a 24 h period. Maximum tumour uptake (25.8 ꢁ 7.5 % ID gꢀ1) was achieved at 16 h and maintained at 24 h (21.6 ꢁ 1.8 %
ID gꢀ1). Rapid clearance of the [Cu-64-DAHA-EDTA]-B72.3 from the blood resulted in good tumour-to-blood ratios
(E 3.3) within a shorter period (6 h) than previously reported with B72.3 whole antibody and the LS174t tumour bearing
nude mouse model.
Manuscript received: 15 November 2016.
Manuscript accepted: 23 February 2017.
Published online: 10 April 2017.
Introduction
isideal for PET imaging. Cu-67 has a bemissionthatissuitablefor
radiotherapy.
Personalised medicine involves the tailoring of therapy for
patients to ensure they get the best treatment with minimal side
effects.[1] Interest in the use of nuclear molecular imaging tech-
niques such as positron emission tomography (PET) for perso-
nalised medicine continues to grow. The high sensitivity and
specificity of PET agents allows one to track the radiolabelled
biomarker in vivo, and therefore determine the extent of disease
and identify dose limiting organs which can cause unwanted side
effects. Research into the production of new PET imaging agents
and novel methods for radiolabelling biomarkers is important to
the success of PET in personalised medicine.
When a PET emitting radioisotope is substituted with a b- or an
a-emitting radioisotope, the resultant surrogate agent (radiother-
apeuticagent)can thenbe used totreat the disease. Copper (Cu-64)
and copper-67 (Cu-67) are known as a theranostic pair of radio-
isotopes.[2,3] Cu-64 has a positron emission (bþ ¼ 0.65MeV) that
The positron energy for Cu-64 is similar to that of F-18
(bþ ¼ 0.64 MeV; t1/2 ¼ 1.8 h) so one can expect the resolution of
Cu-64 PET agents to be comparable to F-18 agents. The longer
half-life (t1/2 ¼ 12.7 h) of Cu-64 is more appropriate for labelling
larger molecules such as peptides and proteins and it also permits
the transport of the Cu-64 agents across continents.[2c,d,4] High
specific activity Cu-64 can be produced on a proton accelerator
via various nuclear reactions. For low energy protons, enriched
Ni-64 is the targetof choice, while different isotopically enriched
Zn targets are used when bombarding with moderate to high
energy protons.
The Cu-67 half-life is 61.9 h and its b emissions (bꢀ ¼ 121
avg
(57 %), 154 (22 %), 189 (20 %) keV) have been used in the
radiotherapy of systemic disease.[2,3,5,6] Its gamma emissions are
ideal for single photon emission tomography (SPECT). Cu-67 has
Journal compilation Ó CSIRO 2017
www.publish.csiro.au/journals/ajc

DAHA-EDTA for Cu-64 Radiolabelling for PET Imaging
615
been produced using high-energy accelerators (. 100 MeV
protons) or reactors intermittently over the years. Until now the
worldwide supply of Cu-67 has been hampered by low specific
activity, poor radiochemical and chemical product purity, and an
unreliable supply. However, new cross-section data and the
commissioning of several 70 MeV proton cyclotrons around the
world, as well as advances in engineering of linear accelerators in
recent times, has provided impetus to assess different production
routes.[4,5] Inparticularthephotonuclearreactionof68Zn(g,p)67Cu
using thick enriched Zn-68 targets has shown significant promise
for large scale production of high purity Cu-67.[5,6]
Both of these copper isotopes have been complexed by a
wide range of polyaza, polyazacarboxylate, and hexaazacage
ligands.[2d,7] The bi-functional chelator (BFC) forms of these
ligands are used to radiolabel biomarkers. BFCs possess two
functional parts: a chemically reactive group for covalent
attachment to the biomarker and a metal binding moiety for
complexing the radiometal ion.
The monoclonal antibody B72.3 has been widely investigated
for imaging and therapy of cancer. It recognises the tumour-
associated protein, (TAG)-72 antigen, that is expressed in several
epithelial derived cancers such as colorectal and ovarian.[8] B72.3
antibody wasthe first radiolabelled (In-111) antibody (OncoScint)
to be approved for imaging in ovarian cancer in humans. The
B72.3 antibody has been radiolabelled with a range of radio-
nuclides; directly with I-125 and I-131 and indirectly with Tc-
99m, In-111, Y-88, Pb-203, Cu-64, and Cu-67 using bi-functional
chelators.[2d,7–10] Open chain polyazacarboxylate derivatives of
ethylenediaminetetraacetic acid (EDTA) and diethylenetriamine-
pentaacetic acid (DTPA), polyaza and polyazacarboxylatemacro-
cycles, and hexaaza cage ligands have all been deployed to
radiolabel the B72.3 antibody with radiometals. Evaluation of
the targeting and stability of the resultant radiolabelled B72.3
immunoconjugate is commonly conducted in LS174t tumour-
bearing nude mice. The LS174t is a colorectal cancer cell line that
expresses Tag-72 antigen.
Many of the BFCs used are produced by a lengthy synthetic
route. It was of interest to us to see if a simpler BFC ligand
system could used to radiolabel proteins. In this study a new
ligand, a diaminodihydroxyaryl derivative of EDTA (N,N0-bis
(carboxymethyl)-N,N0-bis(2-hydroxy-5-aminoacetanilido)-1,2-
diaminoethane, DAHA-EDTA) was synthesised and evaluated
for the radiolabelling of the B72.3 antibody with Cu-64. Con-
ditions for complexing Cu-64 and radiolabelling of the B72.3
antibody are presented. The targetting and clearance properties
of the resultant radioimmunoconjugate, [Cu-64-DAHA-
EDTA]-B72.3 were investigated in LS174t tumour-bearing
nude mice.
Radiolabelling DHA-EDTA and DAHA-EDTA with Cu-64
The complexation of Cu-64 with DAHA-EDTA and DHA-
EDTA in various buffers (0.1 M sodium acetate, buffered to pH
4, 5, and 6, and 0.1 M sodium phosphate buffer of pH 6 and 7)
was investigated. Both ligands complexed (.95 %) the Cu-64
within 10 min at 378C for all buffers.
As the Cu-64-DAHA-EDTA reaction mixture would be used
to conjugate directly to the antibody, pH 6 phosphate buffer was
chosen for optimising the specific activity of the Cu-64-DAHA-
EDTA complex. Reaction mixtures with ligand-to-Cu molar
ratios of 300, 100, 50, 25, 10, 5, 1.5, and 1.2 all gave .95 %
complexation within 10 min at 378C.
Serum Stability of the Cu-64-DHA-EDTA (Parent Species)
The Cu-64-DHA-EDTA complex was incubated in human
serum at 378C. Aliquots were removed from the reaction mix-
ture at 2, 4, 24, 48, and 72 h and separated on a Sephadex G-25
column. Radioactivity elution profiles showed no protein
binding of the Cu-64 and more than 95 % of the copper was
associated with DHA-EDTA after 72 h.
Radiolabelling of B72.3 Antibody with Cu-64-DAHA-EDTA
Several conditions to conjugate or radiolabel the B72.3 antibody
with Cu-64-DAHA-EDTA complex were attempted. These
included the use of different buffers, varying the concentrations
of protein and the molar ratios of 1-ethyl-3-(3-dimethylamino-
propyl)carbodiimide (EDC) and Cu-64-DAHA-EDTA to anti-
body. Varying the buffer conditions (i.e. pH 4 and 5 0.1 M
sodium acetate and pH 6, 7, and 8 0.1 M sodium phosphate)
showed that the best radiolabelling efficiency (15 %) was
achieved using 0.1 M phosphate buffer at pH 6. All other buffer
conditions yielded lower radiolabelling efficiencies (#8 %).
Increasing the antibody concentration from 1 to 5 mg mLꢀ1 did
result in increased labelling efficiency but unfortunately also
increased the amount of cross-linked protein. For 3 mg mLꢀ1 of
antibody, the labelling efficiency was 15 % and the cross-linked
protein was 4 %. For reactions with 5 mg mLꢀ1 of antibody, the
labelling efficiency remained at 15 %, however, the cross-linked
protein increased significantly to 22 %. (Note: the molar ratios
of EDC and Cu-64-DAHA-EDTA to antibody were set at 500
and 100, respectively). The activating agent, EDC, can cause the
antibody to cross-link but it is likely that the Cu-64-DAHA-
EDTA complex with its two functional groups (anilino nitrogen)
available for covalent attachment to the protein, contributed to
the increase in cross-linked antibody. Therefore the optimum
antibody concentration for the conjugation reaction was set at
3 mg mLꢀ1
.
Additional investigations to optimise the yield of the conju-
gation reaction involved varying the molar ratio of the Cu-64-
DAHA-EDTA complex (5, 10, 20, 50, and 100 molar ratio) and
the activating agent, EDC (100, 500, and 1000 molar ratio), to
antibody. The optimal molar ratio of EDC to antibody B72.3
was found to be 500. Increasing the EDC concentration to 1000
resulted in increased cross-linked antibody from 4 to 8 %.
Table 1 summarises the effect of the molar ratio of Cu-64-
DAHA-EDTA to antibody on the radiobelling efficiency of
B72.3. The data in Table 1 shows how the radiolabelling
efficiency decreases as the molar ratio of Cu-64-DAHA-EDTA
to antibody increases, but the number of ligands attached
increases. Unfortunately the percentage of cross-linked anti-
body increases as the molar ratio of Cu-64-DAHA-EDTA to
antibody was increased (.20 %).
Results
The DAHA-EDTA, its precursor N,N0-bis(carboxymethyl)-N,
N0-bis(2-hydroxy-5-nitroacetanilido)-1,2-diaminoethane (DNHA-
EDTA), and parent N,N0-bis(carboxymethyl)-N,N0-bis(2-hydro-
xyacetanilido)-1,2-diaminoethane (DHA-EDTA) derivative
were synthesised according to literature methods.[11,12] Ethyle-
nediaminetetraacetic dianhydride was refluxed with a slight
excess of the appropriate aminophenol derivative in dry aceto-
nitrile under nitrogen. DAHA-EDTA was synthesised by reduc-
ing the nitro groups of DNHA-EDTA at room temperature under
nitrogen using an excess of NaBH4 on Pd/C. The purity of each
ligand isolated was confirmed by 1H and 13C NMR spectroscopy,
mass spectrometry, and FT-IR spectroscopy.

616
M. Ramli et al.
Table 1. Effect of Cu-64-DAHA-EDTA-to-antibody molar ratio on
radiolabelling efficiency
shows that the tumour-to-blood ratio is 2.4, 3.3, 3.2, and 5.6
at 5, 6, 16, and 24 h p.i., respectively. The tumour-to-kidney
ratios improve over time (1.5, 1.7, 2.3, and 2.1, respectively).
However the tumour-to-liver ratios stay at ,1.0 improving
moderately to 1.3 at 24 h p.i.
Molar ratio of EDC to antibody: 500 : 1; concentration of antibody:
3 mg mLꢀ1. Reaction mixtures were incubated at 378C for 1 h
Molar ratio of
Radiolabelling
efficiency [%]
Cu-DAHA-EDTA
conjugated to
Cu-64-DAHA-EDTA
to antibody
Analysis of Metabolites
B72.3 antibody
Blood from the mice at each time point was withdrawn into a
heparinised syringe, then transferred to microcentrifuge tubes
and counted in a gamma counter. Each blood sample was then
centrifuged at 2415 g for 6 min and the fractions separated,
weighed, and then counted in a gamma counter. The recovered
plasma from each time point was pooled and a 500 mL aliquot of
the combined sample was loaded on to a pre-blocked Sephadex-
G25 column (eluent: 0.01 M phosphate buffered saline (PBS) at
pH 7.2, flow rate: 1.85 mL minꢀ1). Only two radioactive species
were present; they were identified as [Cu-64-DAHA-EDTA]-
B72.3 (retention time: 24 min) and Cu-64-DAHA-EDTA
(retention time: 48 min). The Cu-64-DAHA-EDTA fractions
represented ,2 % and 4 % of the total activity in plasma at 6 and
16 h, respectively. At 24 h the percentage of unconjugated
Cu-64-DAHA-EDTAincreasedto16 % oftotalactivityinplasma.
100
50
20
10
5
15
23
28
34
40
15
12
6
3
2
The highest yield and purest product for the conjugation
reaction was achieved with an antibody concentration of
3 mg mLꢀ1 and a molar ratio of antibody to EDC to Cu-64-
DAHA-EDTA of 1 : 500 : 5 in phosphate buffer (pH 6) at 378C
for 1 h.
As the conjugation reaction results in a mixture of products,
the crude [Cu-64-DAHA-EDTA]-B72.3 had to be purified
further. All reaction mixtures were loaded onto a size-exclusion
column (BioSep 3000)-HPLC for further purification. The
isolated [Cu-64-DAHA-EDTA]-B72.3 product was free from
both Cu-64-DAHA-EDTA and macro-aggregates formed from
cross-linked antibody. The radiochemical purity of the [Cu-64-
DAHA-EDTA]-B72.3 isolated was 96.7 ꢁ 1.6 % and its specific
activity was 49 ꢁ 0.9 MBq mgꢀ1 at the end of synthesis and
quality control. The relative immunoreactivity of [Cu-64-
DAHA-EDTA]-B72.3 was 87 ꢁ 2.6 % of unmodified B72.3.
A typical high-performance liquid chromatography (HPLC)
profile of crude and purified product is given in the Supplemen-
tary Material (see Fig S1, Supplementary Material).
Dosimetry of Cu-64-DAHA-EDTA-B72.3
MIRDOSE 3 software was used to estimate the dosimetry of the
[Cu-64-DAHA-EDTA]-B72.3 in a similar manner to that
reported elsewhere.[3,13] Typically, the Restime software pro-
gram (developed by Eric Hetherington, ANSTO) was used to
calculate the residence time of radioactivity in the various
organs given in Table 2. The residence times were then used to
estimate human organ doses with the MIRDOSE 3 software for a
70 kg reference adult. The analysis assumes identical behaviour
and similar pharmacokinetics of [Cu-64-DAHA-EDTA]-B72.3
in mice and humans. Data are presented in Table S3 (Supple-
mentary Material). The total body dose of 0.011 mGy MBqꢀ1 is
lower than that reported for other radiolabelled immunoconju-
gates of B72.3.[3,9,10]
Stability of [Cu-64-DAHA-EDTA]-B72.3
The [Cu-64-DAHA-EDTA]-B72.3 was incubated in pH 6
phosphate buffer at 48C and 378C for up to 48 h. Aliquots were
removed from the solutions at various time intervals and ana-
lysed by size-exclusion HPLC. The data (i.e. UV-vis and
radioactivity profiles) showed the purity of the [Cu-64-DAHA-
EDTA]-B72.3 was maintained (. 95 %) up to 24 h at both
temperatures, however significant breakdown (up to 10 %) was
evident after 48 h.
Discussion
The study demonstrated that DAHA-EDTA could be prepared in
high yield using a simple two-step synthesis. DAHA-EDTA
complexation of copper was rapid over the pH range of 4 to 7. A
molar ratio of DAHA-EDTA to Cu of 1.2 in phosphate buffer at
pH 6.0 gave .95 % labelling efficiency within 10 min at 378C.
The serum stability of its parent analogue, Cu-64-DHA-
EDTA, was found to be comparable to Cu-67-nitrobenzyl-TETA,
its macrocyclic counterpart.[14] Surprisingly the Cu-64-DHA-
EDTA was found to be more stable (.95% intact) in serum
than Cu-67-DTPA-NH-Bu (,23%) and Cu-67-1-p-nitrobenzyl
BEDTA (,14%) counterparts after 3 days incubation.[14]
Conjugation of Cu-64-DAHA-EDTA to B72.3 antibody is
performed in two chemical steps. First the Cu-64-DAHA-EDTA
complex is formed in phosphate buffer at pH 6 and the resultant
Biodistribution of [Cu-64-DAHA-EDTA]-B72.3 in
LS174t Tumour-Bearing Nude Mice
The [Cu-64-DAHA-EDTA]-B72.3 was injected into LS174t
tumour-bearing nude mice, which were then sacrificed at 0.5, 1, 2,
3, 4, 5, 6, 16, and 24 h post-injection (p.i.) and their organs har-
vested. Five animals per time point were used and selected data are
summarised in Table 2 and Table 3. A complete set of biodis-
tribution data are provided in Tables S1 and S2 (Supplementary
Material).
reaction mixture added to the antibody solution (3 mg mLꢀ1
)
Table 2 shows the percentage injected dose per gram of organ
(%ID gꢀ1) at each time point. Table 3 gives the calculated ratio
for tumour to blood, kidney, and liver at each time point. The
clearance of [Cu-64-DAHA-EDTA]-B72.3 is bi-phasic with
two distinct half lives: a fast a phase of 2.5 ꢁ 0.75 h and a longer
b phase of 60 ꢁ 6.5 h.
containing an excess of EDC in phosphate buffer at pH 6. The
conjugation reaction mixture is incubated at 378C for 1 h. The
final purification step involves size-exclusion chromatography
to isolate the [Cu-64-DAHA-EDTA]-B72.3, free from macro-
aggregates, unreacted Cu-64-DAHA-EDTA, and by-products.
The influence of antibody, EDC, and Cu-64-DAHA-EDTA
concentration on the conjugation reaction was investigated.
The optimum conjugation reactions conditions were set at
The [Cu-64-DAHA-EDTA]-B72.3 appears to be cleared
preferentially from the blood via the hepatic system. Table 3

DAHA-EDTA for Cu-64 Radiolabelling for PET Imaging
617
Table 2. Biodistribution of [Cu-64-DAHA-EDTA]-B72.3 in LS174t tumour bearing nude mice
Biodistribution after various time periods [% ID gꢀ1 (standard deviation)]A
Organ
1 h
2 h
4 h
6 h
16 h
24 h
Liver
25.2 (3.6)
6.1 (0.5)
14.8 (1.3)
2.1 (1.0)
4.3 (0.6)
3.5 (0.6)
14.5 (1.1)
8.6 (1.0)
17.7 (1.9)
5.2 (5.4)
8.8 (3.1)
9.2 (1.8)
16.7 (7.9)
24.4 (1.5)
7.3 (1.1)
14.0 (1.5)
2.1 (0.5)
5.6 (0.8)
3.6 (0.6)
14.7 (1.3)
9.1 (0.1)
13.8 (1.3)
22.6 (17.5)
9.2 (1.2)
12.1 (1.1)
27.6 (9.6)
25.6 (2.6)
7.8 (0.7)
12.3 (1.1)
1.6 (0.5)
5.1 (1.1)
3.3 (0.4)
14.5 (1.9)
7.9 (0.3)
9.8 (1.4)
5.6 (3.1)
8.7 (1.6)
14.1 (1.5)
18.9 (2.0)
24.4 (3.2)
7.2 (1.7)
11.2 (1.1)
1.0 (0.2)
4.7 (0.9)
2.7 (0.7)
12.7 (1.8)
6.8 (0.9)
7.7 (0.8)
4.9 (3.5)
7.7 (2.2)
13.5 (2.5)
25.8 (7.5)
17.1 (1.2)
6.4 (1.9)
9.3 (0.6)
1.3 (0.4)
3.0 (0.4)
2.8 (1.0)
10.0 (0.6)
5.2 (0.8)
6.3 (0.8)
9.6 (12.4)
4.2 (2.3)
8.5 (1.2)
20.0 (0.5)
16.0 (1.3)
6.7 (0.7)
9.8 (0.4)
1.2 (0.8)
3.3 (0.8)
2.5 (0.2)
11.0 (0.8)
6.0 (0.5)
3.9 (0.7)
10.8 (15.6)
3.5 (1.4)
7.3 (1.2)
21.6 (1.8)
Spleen
Kidney
Muscle
Skin
Bone
Lungs
Heart
Blood
Bladder
Stomach
GITB
Tumour
AThese values are the average of 5 mice per time point.
BGastrointestinal tract (GIT).
a molar ratio of EDC to Cu-64-DAHA-EDTA to antibody of
500 : 5 : 1 and a 3 mg mLꢀ1 concentration of antibody. The
purified [Cu-64-DAHA-EDTA]-B72.3 had a specific activity
of 49 MBq mgꢀ1 at the end of synthesis and quality control. The
radiochemical purity was .96 %, and the immunoreactivity of
.87 % was considered acceptable for evaluation in animals.
The biodistribution of [Cu-64-DAHA-EDTA]-B72.3 in
LS174t tumour-bearing mice showed that radiolabelled immu-
noconjugate cleared from the blood predominantly via the
hepatic system (see Tables 2 and 3). The clearance from the
blood was bi-phasic, a fast a phase (2.5 ꢁ 0.75 h) and a slower b
phase (60 ꢁ 6.5 h). The biological half-life of the [Cu-64-
DAHA-EDTA]-B72.3 was significantly faster compared with
other radiolabelled whole B72.3 antibody conjugates. For
example the Cu-64-SarAr-B72.3 has an a phase of 7.3 ꢁ 0.4 h
and a b phase of 100 ꢁ 30 h.[10] The % ID gꢀ1 of [Cu-64-
DAHA-EDTA]-B72.3 in the blood was 17.7 ꢁ 1.9 % ID gꢀ1 at
1 h and decreased dramatically to 8.7 ꢁ 1.6 % ID gꢀ1 at 6 h, and
to 3.9 ꢁ 0.7 % ID gꢀ1 at 24 h p.i. The tumour uptake was also
rapid. The % ID gꢀ1 tumour uptake increased considerably from
16.7 ꢁ 7.9 % ID gꢀ1 at 1.0 h to 18.9 ꢁ 2.0 % ID gꢀ1 at 4 h, and to
25.8 ꢁ 7.5 % ID gꢀ1 at 6 h p.i. The % ID gꢀ1 uptake at the tumour
site was maintained up to 24 h post injection (21.6 ꢁ 1.8 %
IDgꢀ1). The tumour-to-blood, the tumour-to-liver, and the
tumour-to-kidney ratios for [Cu-64-DAHA-EDTA]-B72.3 were
3.3, 1.0, and2.3, respectively, at6 h p.i. The tumour/kidney values
improved (2.2) at 24h.
Initially, the rapid blood clearance of [Cu-64-DAHA-EDTA]-
B72.3 was thought to be due to instability of the [Cu-64-DAHA-
EDTA]-B72.3 in vivo; however, analysis of the blood indicated
the Cu-64-DAHA-EDTA was releasedintact. Approximately4 %
of Cu-64-DAHA-EDTA was released from [Cu-64-DAHA-
EDTA]-B72.3 at 16h p.i. At 24h this increased to 16 %. Linder
and Hazegh-Azam report that free Cu (i.e. Cu-64) in the portal
blood and general circulation is generally taken up by serum
protein, albumin, and transcuprein.[15] These protein-bound cop-
pers ions are then rapidly deposited in the liver. If the Cu-64/Cu2þ
has been releasedfromthe DAHA-EDTA one wouldexpect tosee
a continual increase in liveruptake over the 24h period. Infact the
liver-to-tumour ratio remains steady while the total amount
of radioactivity in the liver decreases over the 24 h period (see
Tables 2 and 3).
Table 3. Ratio of uptake of [Cu-64-DAHA-EDTA]-B72.3 in tumour
versus key organs
Tumour to organ
Time
6 h
1 h
2 h
4 h
16 h
24 h
Blood
Liver
0.9
0.7
1.1
2.0
1.1
2.0
1.9
0.7
1.5
3.3
1.0
2.3
3.2
0.9
2.1
5.6
1.3
2.2
Kidney
The biodistribution and localisation of other radiolabelled
B72.3 antibodies with other BFCs show a broad spectrum of
characteristics. Roselli et al. reported that tumour uptake of
around 20–30 % ID gꢀ1 at 24 h was achieved for the In-111
labelled B72.3[9b] Its tumour-to-blood ratio was reported to be
around 2 and its tumour-to-liver ratio was around 0.6. The
tumour uptake increased to a maximum of around 40 % ID gꢀ1
after 48 and 120 h. The tumour-to-blood ratio for the Cu-64-
SarAr-B72.3 agent conducted in our laboratory was 1.4 at 24 h;
however, the tumour to liver and kidney ratios were substantially
better at 1.9 and 4.0, respectively at 24 h. At 48h, the Cu-64-
SarAr-B72.3 tumour further increased to 38.4 ꢁ 4.8 % ID gꢀ1
indicating a more stable compound in the blood than [Cu-64-
DAHA-EDTA]-B72.3. The 99mTc labelled B72.3 reported by
Rosenzweig et al. used the diamidedimercaptide BFC for radi-
olabelling.[9j] Its maximum tumour uptake was only 8 % ID gꢀ1
after 24 h. At this time point the tumour-to-blood and tumour-to-
liver ratios were found to be 2.6 and 5.4 respectively. Brown et al.
also reported the biodistribution of 99mTc labelled B72.3 and
showed the tumour uptake is also low at 7.3 and 10.8% ID gꢀ1 at
6 and 24 h, respectively.[9f] At 6 h, the tumour-to-blood ratio was
0.5and tumour-to-liver ratio was1.4, while at 24h the valueswere
improved; the tumour-to-blood ratio was 1.4 and the tumour-to-
liver ratio was 2.5. The difference in the tumour-to-kidney and
-liver ratios of these radiolabelled B72.3 agents appears to reflect
the natural biological clearance path of the radiometal ion used.
Achieving optimum imaging quality with PET imaging
agents requires that the agent localises to the target and clears
rapidly from the blood. The [Cu-64-DAHA-EDTA]-B72.3
described in this study, localised to target sites within 6 h and
cleared rapidly from the blood, to attain a tumour-to-blood ratio

618
M. Ramli et al.
of ,3.3 by 6 h, which is higher than any other labelled whole
B72.3 product (in the same animal model) at this time to our
knowledge. If the [Cu-64-DAHA-EDTA]-B72.3 behaves in a
similar manner in humans one could envisage imaging on the
day of injection. Unfortunately the slow clearance of the [Cu-64-
DAHA-EDTA]-B72.3 from the liver suggests this agent would
not be as effective as other agents for monitoring tumours in the
liver. The use of DAHA may be attractive for other biomarkers
that have preferential clearance through the kidneys.
and 13C NMR spectra were collected using a Bruker Avance
DPX 400 Spectrometer. Positive fast atom bombardment (FAB)
mass spectra were determined on a JEOL DX-300 mass spec-
trometer (MS: parent ion (FAB): M/C, LHþ). Melting points
are reported uncorrected. Diffuse reflectance FTIR spectra were
recorded on a BioRad Digilab FTS-60 in 1 % KBr. A microplate
reader was from Dynatech.
High specific activity 64Cu (up to 6.5 ꢂ 1014 Bq gꢀ1) was
obtained from the National Medical Cyclotron (NMC) as a
by-product of the 67Ga production process.
For application in radioimmunotherapy, the [Cu-64-DAHA-
EDTA]-B72.3 shows high target-to-non-target ratios for organs
such as blood and bone marrow. The total body dose is
0.011 mGy MBqꢀ1 (see Table 3), which is lower than that
reported for I-131-B72.3, Y-90-DTPA-B72.3, Cu-64-SarAr-
B72.3, and Cu-67-SarAr-B72.3 at 0.709, 0.652, 0.027, and
0.137 mGy MBqꢀ1, respectively.[3] This reduced dose to non-
target tissues in the body is largely due to the rapid blood
clearance of the [Cu-64-DAHA-EDTA]-B72.3. The rate of
localisation of radioactivity and the [Cu-64-DAHA-EDTA]-
B72.3 stability at the target site over the 6–24 h period, match
the half-life (12.7 h) of the Cu-64 well. It is important to
remember that calculations of the dosimetry data were made
with the results from a biodistribution study in mice. Differences
in metabolic rates will affect the biodistribution in humans,
altering the final dosimetry of the product. Consequently, the
dosimetry results obtained here should be used only as a guide
for the expected biodistribution in humans.
Radioactivity was measured in a CRC-15R Capintec or a
Wallac Wizard 1470 gamma counter. Cu-64 in 0.1 N HCl with
specific activity that ranged from 2.00 to 7.80 ꢂ 108 MBq gꢀ1
was supplied by the National Medical Cyclotron, ANSTO.
B72.3 antibody (7 mg mLꢀ1) was purchased from Bioquest,
Sydney and concentrated to 54.8 mg mLꢀ1 using a Centricon
30 concentrator from Amicon. EDC, BSA, mucin, 5-amino
salicylic acid, and Tween 20 were purchased from Sigma.
Conjugated rabbit anti-mouse immunoglobulin (RAM-HP)
was purchased from DAKO. A 20000 MCO concentrator from
Sartorious, 20 ꢂ 20 cm2 silica gel impregnated glass fibre sheets
(Gelman Sciences), and a 0.22 mm filter from Millipore were
used. A Waters HPLC was fitted with a 1000 mL sample
loop (Rheodyne), a 60 ꢂ 21.2 mm Bio-Sep-Sec S3000 Guard
(Phenomenex), and a 600 ꢂ 21.1 mm Bio-Sep-Sec S3000 Col-
umn (Phenomenex) was used for purification of the conjugates.
The HPLC was fitted with two detectors, UV-vis (set to 280 nm)
and a single channel radioactive monitor from EG & Ortec. Data
acquisition was conducted using Value Chrom software from
Bio-Rad Laboratory. A second HPLC (Waters) was used for
quality control purposes. This was system fitted with a 50 mL
sample loop (Rheodyne), a 75 ꢂ 7.8 mm Bio-Sep-Sec S3000
Guard (Phenomenex), a 300 ꢂ 7.8 mm Bio-Sep-Sec S3000
Column (Phenomenex), and a UV-vis monitor (280 nm). Data
acquisition was conducted using Millennium 245 series software.
The outlet was connected to a fraction collector (Pharmacia).
Protein concentrations were verified using a Bio-Rad Protein
assay kit.
For metabolite studies the Sephadex-G-25 was prepared in
the following manner. Sepadex-G-25 resin was swollen in
0.01 M PBS (which contained 0.05 % sodium azide) at pH 7.2
for at least for 3 h or overnight. The swollen Sephadex-G-25
resin was degassed by vacuuming and gentle shaking for 30 min.
The resin was loaded into a glass tube; 55 L ꢂ 1.2 cm outer
diameter fitted with a frits glass base, to a height of 48 cm and the
top of the resin was covered with a thin layer of glass wool. The
column was then stabilised by passing through 3 ꢂ 125 mL of
0.01 M PBS at pH 7.2 onto the column. Prior to its use the
column was blocked by loading 1 mL of 10 % BSA followed by
elution with 3 ꢂ 125 mL of 0.01 M PBS at pH 7.2.
Conclusion
The DAHA-EDTA was readily and quantitatively radiolabelled
with Cu-64 under mild conditions. The Cu-64-DAHA-EDTA
complex conjugation to whole B72.3 antibody was optimised to
produce a reasonably high specific activity [Cu-64-DAHA-
EDTA]-B72.3 product. Unfortunately, radiolabelling of proteins
requiresa two-step processandmoreextensive purificationbefore
use. Its biodistribution in tumour bearing nude mice demonstrated
unusually fast clearance from the blood and good localisation at
the tumour site compared with similar products reported in the
literature. The DAHA-EDTA proved to be an effective chelator
for radiolabelling whole antibody and is worthy of further
investigation for radiolabelling other biomarkers.
Experimental
Materials and Equipment
All materials used were of ACS reagent grade and used without
further purification. Ethylenediaminetetraacetic dianhydride,
2-amino-4-nitrophenol, 10 % palladium charcoal, sodium
borohydride, sodium azide, and sodium hydroxide were pur-
chased from Sigma–Aldrich. EDTA, acetic anhydride, pyridine,
diethyl ether, and acetone were from Ajax Chemical Company,
acetonitrile and methanol purchased from Univar, and ethanol
from BDH. HPLC grade acetonitrile was purchased from
Merck. Sephadex G-25 (20–80 mm) for gel filtration and bovine
serum albumin (BSA) were purchased from Aldrich.
Ligand Synthesis and Characterisation
DAHA-EDTA was synthesised in a manner similar to that
described previously (see Fig. 1).[11,12]
All buffers were prepared using Millipore triply de-ionised
water. Silica gel impregnated, instant thin-layer chromatogra-
phy paper (ITLC-SG), and 0.22 mm filters were purchased from
Gelman Scientific. Deuterated NMR solvents, D2O, NaOD and
d6-DMSO, were purchased from Cambridge Isotope Laborato-
ries. Elemental analyses were performed by the Microanalytical
N,N0-Bis(carboxymethyl)-N,N0-bis(2-hydroxy-5-
nitroacetanilido)-1,2-diaminoethane (DNHA-EDTA)
To a solution of 2-amino-4-nitrophenol (1.74 g, 1.59 ꢂ 10ꢀ2
mol) in dry acetonitrile (100 mL) was added EDTA anhydride
(2.0 g, 7.81 ꢂ 10ꢀ3 mol). The suspension was refluxed under
nitrogen overnight. The warm reaction mixture was then filtered
and the resultant bright yellow solid was washed with 70 %
1
Service, The University of Queensland, Australia. H NMR
spectra were collected using a JEOL JNM-GX400 spectrometer

DAHA-EDTA for Cu-64 Radiolabelling for PET Imaging
619
O2N
(10 mL, 28 % NH3) in ethanol (8 mL) was added to the suspen-
sion. The reduction was continued at room temperature for
30 min under N2 (g) or, until the solution became clear in colour.
Concentrated hydrochloric acid was added dropwise to the
solution until the evolution of gas ceased. The final pH of the
solution was ,6.5. The palladium catalyst was removed by
filtration and the filtrate reduced under vacuum. Water was
azeotroped using copious amounts of ethanol/acetone. The
purple/brown residue was suspended in a minimum volume of
ethanol (20 mL) and insoluble salts were removed by filtration.
Yield of the isolated product was 0.85 g (95 %). Mp 5048C.
N
N
O
O
NH2
ꢀ
O
O
O
O
OH
NO2
O2N
HOOC
NH
COOH
NH
N
N
l
(d6-DMSO) 7.52 (s, 2H, OH). 6.90–6.81 (m, 4H, Ar), 3.92, 3.76
(s, 4H, 4H, CH2CONH, CH2COOH), 3.28 (s, 4H, NCH2CH2).
max/cmꢀ1 1734.0 NHCO, 1684.0 COOH, 2998 (br) OH. dH
HO
OH
O
O
DNHA-EDTA
N,N0-Bis(carboxymethyl)-N,N0-bis(2-
hydroxyacetanilido)-1,2-diaminoethane (DHA-EDTA)
Pd/C
The parent species DHA-EDTA of DAHA-EDTA was pre-
pared and used to evaluate the serum stability of the Cu-64
radioabelled complex. To a solution of 2-aminophenol (1.74 g,
1.59 ꢂ 10ꢀ2 mol) in dry acetonitrile (100 mL) was added EDTA
anhydride (2.0 g, 7.81 ꢂ 10ꢀ3 mol). The suspension was
refluxed under nitrogen overnight. The warm reaction mixture
was then filtered and the resultant yellow solid was washed with
70 % ethanol followed by copious amounts of methanol or
ethanol and then acetone and dried under vacuum. The yield
for the yellow powder was 3.52 g (95 %). m/z (FAB) 475 (MHþ).
Mp 204–2058C. lmax/cmꢀ1 1708.2 NHCO, 1635.8 COOH,
3314.1 OH. dH (d6-DMSO) 8.02 (s, 2H, OH), 6.91–6.72 (m,
8H, Ar), 3.46, 3.38 (s, 4H, 4H, CH2CONH, CH2COOH), 2.86 (s,
4H, NCH2CH2). dC (D2O/NaOD) 53.97, 59.49, 59.72 (CH2)
117.89, 120.11, 125.31, 128.2 (CH, Ar), 125.2, 151.46
(Cq),174.57, 180.05 (C=O). Anal Calc. for C22H26N4O8ꢃ3/
2H2O: C 52.69, H 5.82, N 11.17. Found: C 52.84, H 5.59,
N 11.26 %.
NaBH4
H2N
NH2
HOOC
NH
COOH
NH
N
N
HO
OH
O
O
DAHA-EDTA
Fig. 1. Schematic for the synthesis of DAHA-EDTA.
ethanol followed by copious amounts of methanol or ethanol
then acetone. Yield 3.4 g (81 %). m/z (FAB) 564 (MHþ).
Mp 222–2268C. lmax/cmꢀ1 1728.4 NHCO, 1693.9 COOH,
3299.3 OH. dH (d6-DMSO) 9.88 (s, 2H, OH), 9.06, 7.89, 6.94
(m, 6H, Ar), 3.56, 3.52 (s, 4H, 4H, CH2CONH, CH2COOH),
2.83 (s, 4H, NCH2CH2). Anal. Calc. for C22H24N6O12ꢃH2O:
C 45.37, H 4.50, N 14.43. Found: C 45.42, H 4.39, N 14.08 %.
Radiolabelling DAHA-EDTA with Cu-64:
(Cu-64-DAHA-EDTA)
N,N0-Bis(carboxymethyl)-N,N0-bis(2-hydroxy-5-
aminoacetanilido)-1,2-diaminoethane (DAHA-EDTA)
Preliminary work in our laboratory investigated the optimum
buffering conditions for the complexation of Cu-64 with
DAHA-EDTA and the parent DHA-EDTA. Buffers ranging
from pH 4–7 (0.1 M sodium acetate at pH 4, 5, and 6 and 0.1 M
sodium phosphate at pH 6 and 7) were tested. Greater than 95 %
of complexation of the Cu-64 was achieved for all buffer con-
ditions investigated.[11]
An aliquot of Cu-64 in 0.1 N HCl solution was first evapo-
rated to dryness and then re-suspended with an aliquot of Milli-
water and an equivalent aliquot of 0.1 M phosphate buffer at
pH 6. To a 2.8 ꢂ 10ꢀ8 mol Cu-64/Cu solution (109.9 mL,
394 MBq of Cu-64 in 1 to 1 of H2O and 0.1 M phosphate at
pH 6) was added a 3.4 ꢂ 10ꢀ8 mol DAHA-EDTA solution
(3.8 mL of 20 mg mLꢀ1 of DAHA-EDTA in 0.1 M phosphate
buffer at pH 6). The mixture was vortexed and then incubated at
,378C for 10 min. The percentage of complexation or radiola-
belling of Cu-64 to DAHA-EDTA was determined by instant
thin-layer chromatography (ITLC). The procedure involved
depositing a 0.2–1 mL aliquot of the mixture on activated silica
impregnated glass fibre strips (1 cm from the bottom). The strips
were dried and then developed in Milli-Q water and a trace
amount of NH4OH (typically, 10 mL of Milli-Q water þ 150 mL
of 32 % of NH4OH). Once the solvent reached 0.5 cm of the
top of the strips, they were dried in air. The ITLC strips were
Method A: An aqueous solution of sodium borohydride
(0.24 g in 2 mL H2O) was slowly added to an aqueous, nitrogen-
purged solution of Pd/C catalyst (50 mg, 1 mL). To this suspension
was added DNHA-EDTA (1.0 g) in an ethanol/NaOH solution
(10 mL/1 mL, 8 % NaOH). Care was taken to add the latter slowly
as the reaction could be vigorous. The suspension was stirred at
room temperature for 20 min, or until the solution became clear.
Acid (conc. HCl) was added dropwise till excess sodium borohy-
dride was quenched (i.e. effervescence ceased). Pd/C was
removed by filtration and the volume of the filtrate reduced under
vacuum to dryness. The resultant pale pink powder was dissolved
in a minimum of methanol and insoluble salts were removed by
filtration. The product was isolated after removal of methanol.
Yield: 0.85 g (95 %). m/z (FAB) 509 (MHþ). lmax/cmꢀ1 1740.1
NHCO, 1683.3 COOH, 3300 (br) OH. dH (d6-DMSO) 9.33 (s, 2H,
OH), 8.41–8.28 (m, 6H, Ar), 5.42, 5.35 (s, 4H, 4H, CH2CONH,
CH2COOH), 3.80 (s, 4H, NCH2CH2). Anal. Calc for
C22H28N6O8Na2ꢃNaClꢃ5H2O: C 34.88, H 5.05, N 11.09. Found:
C 34.55, H, 4.86, N 10.57 %.
Method B: Sodium borohydride (0.16 g) was added to an
aqueous, nitrogen purged solution of 10 % Pd/C (50 mg, 2 mL
H2O). A concentrated ammonia solution of DNHA-EDTA

620
M. Ramli et al.
OH
O
then cut into 10 portions and each portion was counted using a
Wallac Wizard 1470 gamma counter. Rf 1.0 for Cu-64-DAHA-
EDTA and 0.0 for Cu-64/Cu2þ. The percentage of radioactivity
in each fraction was calculated. The radiochemical purity of the
Cu-64-DAHA-EDTA was found to be .95 %.
MC ꢂ Metal chelator
Ab
C
ꢂ Antibody
Ab
NHꢀ Clꢁ
N
N
C
Serum Stability of Cu-64-DAHA-EDTA
H
The stability of the Cu-64-DAHA-EDTA complex (or release of
Cu-64 from the complex) in human serum cannot be determined
due to the Cu-64-DAHA-EDTA complex naturally forming salt
adducts (via peripheral aniline-nitrogen groups) with the pro-
tein. To overcome this problem the Cu-64 complex of the parent
species, DHA-EDTA, was formed and assessed as a surrogate
for Cu-64-DAHA-EDTA.
NHꢀ Clꢁ
N
N
O
C
C
Ab
H2N
MC
The Cu-64-DHA-EDTA was prepared in a similar manner to
that described for Cu-64-DAHA-EDTA above. The Cu-64-
DHA-EDTA complex (50 mL) was incubated in 2.5 mL of
human serum (filtered at 0.2 mm) and incubated at 378C. At
set intervals (2, 4, 24, 48, and 72 h) 50 mL samples were removed
and separated by size exclusion chromatography(Sephadex G-25,
Sigma) using a gravity feed column (55ꢂ 0.7cm) pre-blocked
with 1000 mL of 10% BSA and equilibrated with 0.01 mM PBS at
pH 7.2. The UV profile of the eluent was monitored at 280 nm
(UV-1 monitor, Pharmacia) for serum proteins and 2 mL fractions
of the eluent were collected to assess the amount of radioactivity
using a Wallac Wizard 1470 gamma counter. The total activity
collected was compared with 50mL standards (in triplicate) to
ensure that there was no activity left on the column. The activity
associated with each fraction was measured and the percentage of
total radioactivity loaded on the column calculated. The percent-
age of activity associated with the fractions containing serum
proteins and as free complex were calculated to determine the
amount of Cu-64 released from the complex.
Ab
OH
NH
O
H
H
N
NHꢀ Clꢁ
N
ꢀ
MC
C
O
OH
Fig. 2. Schematic of the conjugation of Cu-64-DAHA-EDTA to the
antibody.
using 0.1 M phosphate buffer at pH 6. The mixture was
vortexed and then incubated at 378C for 1 h. To quench the
reaction, a 0.05 to 0.10 molar ratio of EDTA to Cu in 10 mM
PBS at pH 7.2 was added. The mixture was then transferred to a
20000 MCO concentrator or ultra-filter, previously wetted
with 100 mL of 0.2 M phosphate buffer at pH 8. A further
200 mL of 0.2 M phosphate buffer at pH 8 was added to the
filter. The filter was then microcentrifuged for 15 min at
13148 g. The protein was washed a further 2–3 times with the
same buffer until 85–90 % radiochemical purity of [Cu-64-
DAHA-EDTA]-B72.3 was achieved. The [Cu-64-DAHA-
EDTA]-B72.3 was then filtered using a 0.22 mm filter.
Radiolabelling of B72.3 Antibody with Cu-64-DAHA-EDTA
Radiolabelling of the B72.3 antibody is based on the formation
of an amide linkage between an activated endogenous carbox-
ylic functional group (using EDC) on the antibody and amine
substituents of DAHA-EDTA. A schematic of the conjugation
reaction is given in Fig. 2.
Final purification of [Cu-64-DAHA-EDTA]-B72.3 was
achieved by size exclusion chromatography (using
a
600 ꢂ 21.1 mm Bio-Sep-Sec S3000 Column) fitted to HPLC.
The column was then eluted with 0.01 M PBS at pH 7.2 at a
3 mL minꢀ1 flow rate. The retention time of [Cu-64-DAHA-
EDTA]-B72.3 was found to be 40.3 min. The fractions contain-
ing the radioimmunoconjugate were pooled and concentrated
using a 20000 MCO ultra filter. The final product recovered was
then assessed for radiochemical purity, protein concentration,
and immunoreactivity.
The radiochemical purity of the [Cu-64-DAHA-EDTA]-
B72.3 product was determined using size exclusion chromato-
graphy (300 ꢂ 7.8 mm Bio-Sep-Sec S3000) fitted to HPLC.
The column was then eluted with 0.01 M PBS at pH 7.2 at a
1 mL minꢀ1 flow rate. Fractions of 0.5 mL were then retrieved.
The fractions were then counted using a Wallac Wizard 1470
gamma counter. The percentage radiochemical purity was
calculated as a ratio of radioactivity associated with the protein
to the total radioactivity loaded on the column.
The conjugation reaction was optimised for the concentration
of protein and molar ratios of EDC and Cu-64-DAHA-EDTA.
Conditions varied with protein concentration of 1 to 5 mg mLꢀ1
,
EDC-to-antibody molar ratios of 100–1000, and Cu-64-DAHA-
EDTA-to-antibody ratios of 1–100. For example, an aliquot of
B72.3 antibody (diluted 1 : 1 with 0.1M phosphate buffer at pH 6)
was added to a 500 mol ratio of 20 mgmLꢀ1 of EDC in 0.1M
phosphate buffer at pH 6. To the mixture was added a 5–100 mol
ratio of Cu-64-DAHA-EDTA (depending on the specific activity
of Cu-64). If the Cu-64 was of specific activity higher than
2 ꢂ 108 MBq gꢀ1, Cu-64-DAHA-EDTA was added at a 5–10
molar ratio to antibody. When the specific activity of the Cu-64
was lower than 2 ꢂ 108 MBq gꢀ1, the molar ratio of Cu-64-
DAHA-EDTA to antibody ranged from 20–100. In each case
the reaction mixtures were incubated at 378C for 1 h.
More specifically a typical conjugation reaction/labelling
procedure is as follows: To 2.8 ꢂ 10ꢀ9 mol (415 mg, 15.2 mL) of
B72.3 (27.4 mg mLꢀ1) in 0.1 M phosphate buffer at pH 6 was
added 1.4 ꢂ 10ꢀ6 mol (13.3 mL) of EDC (20 mg mLꢀ1 in 0.1 M
phosphate buffer at pH 6). To this mixture was added
2.8 ꢂ 10ꢀ8 mol (114.7 mL) of Cu-64-DAHA-EDTA. The final
concentration of antibody was then adjusted to 3 mg mLꢀ1
Thespecific activityof purified[Cu-64-DAHA-EDTA]-B72.3
was calculated as a ratio of radioactivity associated with radi-
olabelled antibody to concentration of protein present. The radio-
activity was measured using a calibrated Capintec CRC-15R dose
calibrator and protein concentration using a micro BioRad Protein

DAHA-EDTA for Cu-64 Radiolabelling for PET Imaging
621
Assay on Immulon-1 96 well plates. The absorbance at 595 nm
was measured on a Dynatech MR 7000 ELISA reader. The
concentration of the sample was calculated from the standard
curve of absorbance versus protein concentration.
The final product was stored at 4 and 378C for up to 48 h and
then analysed by size-exclusion HPLC. The [Cu-64-DAHA-
EDTA]-B72.3 radiochemical purity was maintained (.95 %) at
24 h at both temperatures. However, at 48 h breakdown of the
radiolablled conjugate was evident with up to 10 % loss of
radioactivity from antibody.
flanks of mice aged 4 weeks. The tumours were then left to grow
for 2 weeks to ,7 ꢂ 7 mm2 in size before use for animal studies.
Each mouse was injected with no more than 0.35 MBq or
35 mg of protein of [Cu-64-DAHA-EDTA]-B72.3. After injec-
tion with [Cu-64-DAHA-EDTA]-B72.3, at time the point
required, the mice were anaesthetised by using CO2, then
sacrificed and dissected. The blood and selected organs were
removed, weighed and then counted using a Wallac Wizard 1470
gamma counter. The percentage of injected dose for blood and
selected organs as well as the percentage of injected dose per
gram of blood of selected organs were calculated. Selected data
for % ID gꢀ1 are given in Table 2. Table 3 gives the calculated
ratio of tumour to blood, liver, and kidney at the various time
points.
Immunoreactivity of [Cu-64-DAHA-EDTA]-B72.3
The immunoreactivity of [Cu-64-DAHA-EDTA]-B72.3 was
measured using an enzyme linked immunoabsorbance assay
(ELISA) in a similar manner to that described previously.[3,10]
Two plates were prepared at the same time. The first one was for
standard antibody (unmodified antibody) and the second one
was for test antibody (modified antibody, i.e. [Cu-64-DAHA-
EDTA]-B72.3). Serial dilutions of mucin from 4 mg mLꢀ1 to
2 ng mLꢀ1 in 0.05 M PBS at pH 7.2 were added to the first four
rows of a 2 Immulon 4, 96 well plate with 0.05 M PBS at pH 7.2
(100 mL). Glycine (i.e. control for non-specific binding) was
plated in parallel to the second four rows of the plates. The plates
were covered with cling film and incubated overnight at 48C.
Each well was then washed two times with 200 mL of 0.05 %
Tween 20 in 0.05 M PBS pH 7.2. The wells were incubated with
200 mL of 0.05 % Tween 20 in 0.05 M PBS at pH 7.2 for at least
2 h at 378C and then washed three times with 200 mL of 0.05 %
Tween in 0.05 M PBS at pH 7.2. Antibodies (standard antibody
and test antibody), 100 mL prepared at 4 mg mLꢀ1 in 0.05 %
Tween 20 in 0.05 M PBS at pH 7.2, were added to each well and
then incubated for 1.5 h at 378C. Each well was then washed
three times with 200 mL of 0.05 % Tween 20 in 0.05 M PBS at
pH 7.2. A 1/2000 dilution of the second antibody, peroxide-
conjugated rabbit anti-mouse immunoglobulins (RAM-HP), in
0.05 % Tween 20 in 0.05 M PBS at pH 7.2, was added to each
well. The plates were then incubated for 1 h at 378C. The excess
RAM-HP was removed and the well washed a further three
times with 200 mL of 0.05 % Tween 20 in 0.05 M PBS at pH 7.2
and once with 200 mL of 0.05 M PBS at pH 6.8. 5-Amino sal-
icylic acid (5-AS) was prepared by dissolving 100 mg in 100 mL
of 10 mM PBS at pH 6.8 and adding 100 mL of 1% freshly
prepared H2O2. An aliquot (100 mL) of the 5-AS was added to
each well and allowed to react at room temperature before
reading the absorbance at 410 nm. Specific binding of the
antibody was determined by subtracting the absorbance for
glycine wells from the total absorbance obtained for each well.
A saturation curve, correlating absorbance with antigen con-
centration, was plotted. A comparison of saturation curve,
generated for standard antibody and test antibody was used to
assess the relative immunoreactive fractions.
Determination of Metabolites in Blood Plasma
Blood was withdrawn from each of the sacrificed mice into
heparinised syringes and then transferred to a pre-weighed
microcentrifuge tube. The blood samples were weighed and then
counted using a Wallac Wizard 1470 gamma counter. The
counted blood samples were centrifuged for 6 min at 6708 g on a
microcentrifuge. The supernatant (plasma) from five mice at a
certain time point was pooled into a pre-weighed micro-
centrifuge tube. The fractions were then counted using a Wallac
Wizard 1470 gamma counter.
An aliquot of plasma (500 mL) was loaded into a BSA-pre-
blocked and pre-equilibrated (0.01 M PBS at pH 7.2) Sephadex
G-25 (48 ꢂ 1.2 cm) column. The column was then eluted with
0.01 M PBS at pH 7.2 with a 1.85 mL minꢀ1 flow rate and the
eluent monitored using a UV-1 Pharmacia Monitor at 280 nm.
Fractions of 1.85 mL were collected and then counted using a
Wallac Wizard 1470 gamma counter.
Dosimetry of [Cu-64-DAHA-EDTA]-B72.3
The dosimetry of [Cu-64-DAHA-EDTA]-B72.3 was deter-
mined assuming its biological distribution in mice is similar to
humans. Calculations were conducted in a similar manner to that
reported previously.[3,16] Briefly, using the biodistribution
results obtained for [Cu-64-DAHA-EDTA]-B72.3 in the mice,
the radioactivity measured for each organ or tissue was cor-
rected for the natural decay of the radioisotope. The residence
time (mCi h) of the radioactivity in each organ was determined
by integrating the percentage injected dose curve for that organ
(using Restime software developed by Eric Hetherington,
ANSTO).[3] As no time points were monitored beyond 24 h the
residence time of activity in each organ beyond 24 h was con-
sidered to be equivalent to the rate of decay of the radionuclide
(termed extension activity). Both residence time activity and
extension activity were added to determine the total residence
time of the product in each organ. As the dose to each organ is a
combination of the dose from the tissue of that organ as well as
any blood that may be present within it, the total residence time
was corrected for blood content in selected organs according to
the method of Weber et al..[16] The total residence time was
also corrected for activity that was distributed to the tail of the
nu/nu mice by a proportional increase within every organ. The
MIRDOSE 3 program was used to determine the dose delivered
to each organ in a 70 kg human adult as a consequence of the
cumulated activity in those organs. The dose to the whole body
was corrected for additional dose resulting from localisation
of the product to the tumour mass. Data are summarised in
Tables S2 and S3 (Supplementary Material).
Animal Studies
All animal studies were approved by the Animal Care and Ethic
Committee (ACEC) ANSTO under protocol No. 99/143. There
were 5 Nu/nu mice (nude mice) for each time point. The studied
time points were 1, 2, 4, 6, 16, and 24 h. Five 2–3 week-old mice
were implanted with 100 mL of 10 ꢂ 106 cell mLꢀ1 cell sus-
pension in 0.01 M PBS at pH 7.2. The implant was left to grow
for 1 week. The mice were later sacrificed and tumour tissue was
harvested, diced to 2 ꢂ 2 mm2, and kept in chilled sterile media.
The diced cells of tumour were then implanted into the hind

622
M. Ramli et al.
Supplementary Material
(b) Z. Cai, C. J. Anderson, J. Labelled Comp. Radiopharm. 2014, 57,
224. doi:10.1002/JLCR.3165
[8] A. Thor, N. Ohuchi, C. A. Szpak, W. W. Johnston, J. Schlom, Cancer
Res. 1986, 46, 3228.
[9] (a) R. T. Maguire, R. F. R. Schmelter, V. L. Pascucci, J. J. Conklin,
Antibody, Immunoconjugates, Radiopharm. 1998, 2, 257.
(b) D. E. Milenic, M. Roselli, M. W. Brechbiel, C. G. Pippin, T. J.
McMurray, J. O. Carrasquillo, D. Colcher, R. Lambrecht, O. Gansow,
J. Schlom, Nucl. Med. 1998, 25, 471.
The typical HPLC profile of crude and purified final [Cu-64-
DAHA-EDTA]-B73.2 (Fig S1), complete set of biodistribution
data (Table S1), and tumour-to-key organ uptake ratios and the
dosimetry data (Tables S2 and S3) are available on the Journal’s
website.
Acknowledgements
(c) S. M. Larson, J. A. Carrasqillo, D. C. Colcher, K. Yokoyoma, J. C.
Reynols, A. Bacharach, A. Raubitcheck, L. Pace, R. D. Finn, M.
Rotman, M. Stabin, R. D. Neuman, P. Sugarbaker, J. Schlom, J. Nucl.
Med. 1992, 32, 1661.
The authors thank Beverley Izard and Kerynne Belbin for assistance with
animal studies and Mr Eric Hetherington who conducted the dosimetry cal-
culations. Ramli and Smith (two generations of Ph.D. students of Professor L.
F. Lindoy) would also like to express thanks to Professor Leonard F. Lindoy
for his guidance, support, and the inspiration to pursue research.
(d) D. Colcher, M. F. Minelli, M. Roselli, R. Muraro, J. Schlom,
Cancer Res. 1988, 48, 4597.
(e) D. Colcher, A. M. Keenan, M. S. Larson, R. Schlom, Cancer Res.
1984, 44, 5744.
References
[1] J. Park, J. Y. Kim, Curr. Top. Med. Chem. 2013, 13, 458. doi:10.2174/
1568026611313040006
(f) B. A. Brown, C. B. Deaborn, C. A. Drozynski, H. Sands, Cancer
Res. 1990, 50(Suppl), 835s.
[2] (a) Nuclear Science Advisory Committee (NSAC), The 2015 Long
Range Plan For the DOE-NP Isotope Program. Available at: http://
science.energy.gov/,/media/np/nsac/pdf/docs/2015/2015_NSACI_
Report_to_NSAC_Final.pdf
(g) K. D. Brandt, D. K. Johnson, Bioconjug. Chem. 1992, 3, 118.
doi:10.1021/BC00014A005
(h) M. Roselli, J. Schlom, O. A. Gansow, M. W. Brechbiel,
S. Mirzadeh, C. G. Pippin, D. E. Milenic, D. Colcher, Nucl. Med. Biol.
1990, 18, 389.
(b) A. N. Asabella, G. L. Cascini, C. Altini, D. Paparella,
A. Notaristefano, G. Rubini, BioMed Res. Int. 2014, 2014, 786463.
doi:10.1155/2014/786463
(i) W. B. Webster, S. J. Harwood, R. G. Caroll, M. A. Morrissey,
J. Nucl. Med. 1992, 33, 498.
(c) S. V. Smith, Expert Opin. Drug Discovery 2007, 2, 659. doi:10.1517/
17460441.2.5.659
(d) S. V. Smith, J. Inorg. Biochem. 2004, 98, 1874. doi:10.1016/
J.JINORGBIO.2004.06.009
(j) H. S. Rosenzweig, G. N. Ranadive, T. Seskey, M. W. Epperly,
W. D. Bloomer, Nucl. Med. Biol. 1994, 21, 171. doi:10.1016/0969-
8051(94)90006-X
[10] N. Di Bartolo, A. M. Sargeson, S. V. Smith, Org. Biomol. Chem. 2006,
[3] N. Di Bartolo, S. V. Smith, E. Hetherington, A. M. Sargeson, Aust. J.
Chem. 2009, 62, 1261. doi:10.1071/CH09369
4, 3350. doi:10.1039/B605615F
[11] P. F. Schmidt, Radiolabelling of Antibodies for Use in Diagnosis or
Treatment of Disease 1997, Ph.D. thesis, University of Sydney.
[12] (a) S. V. Smith, R. M. Lambrecht, P. F. Schmidt, F. T. Lee,
T. M. Donlevy, N. M. Di Bartolo, U.S. Patent 5807535 A 1998.
(b) S. V. Smith, R. M. Lambrecht, P. F. Schmidt, F. T. Lee, Patent EP
590766A3 1994.
[4] (a) M. N. Naomm, K. H. Ahmed, N. F. Naji, Int.J. Phys. Res. 2013, 3, 13.
(b) F. Szelecse´nyi, Z. Kova´cs, K. Nagatsu, M. Znhang, K. Suzuki,
Radiochim. Acta 2014, 102, 465. doi:10.1515/RACT-2013-2145
[5] (a) N. I. Ayzatskiy, N. P. Dikiy, A. N. Dovbnya, Yu. V. Lyashko, V. I.
Nikiforov, B. I. Shramenko, A. Eh. Tenishev, A. V. Torgovkin, V. L.
Uvarov, in Cyclotrons and their Applications: 18th International
Conference 2007, 243–245.
[13] Oak Ridge Institute for Science and Education, PO Box 117, Oak Ridge,
TN 37831, USA.
(b) N. A. Smith, D. L. Bowers, D. A. Ehst, Appl. Radiat. Isot. 2012, 70,
2377. doi:10.1016/J.APRADISO.2012.07.009
[6] (a) V. I. Nikiforov, V. L. Uvarov, Nucl. Instrum. Methods Phys. Res. B
2011, 269, 3149. doi:10.1016/J.NIMB.2011.04.102
[14] (a) C. F. Meares, M. K. Moil, H. Diril, D. L. Kukis, M. J. McCall, S. V.
Deshpande, S. J. DeNardo, D. Snook, A. A. Epenetos, Br. J. Cancer
1990, 62(Suppl.), 21.
(b) M. K. Moi, S. J. DeNardo, C. F. Meares, Cancer Res. 1990,
(b) S. Howard, V. N. Starovoitova, Appl. Radiat. Isot. 2015, 96, 162.
doi:10.1016/J.APRADISO.2014.12.003
(c) V. N. Starovoitova, P. L. Cole, T. L. Grimm, J. Radioanal. Nucl.
Chem. 2015, 305, 127. doi:10.1007/S10967-015-4039-Z
[7] (a) C. J. Anderson, R. Ferdani, Cancer Biother. Radiopharm. 2009, 24,
379. doi:10.1089/CBR.2009.0674
50(Suppl. 3), 789s.
[15] M. C. Linder, M. Hazegh-Azam, Am. J. Clin. Nutr. 1996, 63, 797s.
[16] (a) D. A. Weber, K. F. Eckerman, L. T. Dillman, J. C. Ryman, MIRD:
Radionuclide Data and Decay Schemes 1989 (The Society of Nuclear
Medicine, Inc.: Reston, VA).
(b) M. G. Stabin, J. Nucl. Med. 1996, 37, 538.