Towards the development of a targeted albumin-binding radioligand: Synthesis, radiolabelling and preliminary in vivo studies

Article abstract of DOI:10.1016/j.nucmedbio.2021.01.001

Introduction: The compound named 4-[10-(4-(2,5-dioxo-2,5-dihydro-1H-pyrrol-1-yl)butanamido)decyl]-11-[10-(β,D-glucopyranos-1-yl)-1-oxodecyl]-1,4,8,11-tetraazacyclotetradecane-1,8-diacetic acid is a newly synthesised molecule capable of binding in vivo to

Full text of DOI:10.1016/j.nucmedbio.2021.01.001

Nuclear Medicine and Biology 94–95 (2021) 53–66
Contents lists available at ScienceDirect
Nuclear Medicine and Biology
journal homepage: www.elsevier.com/locate/nucmedbio
Towards the development of a targeted albumin-binding radioligand:
Synthesis, radiolabelling and preliminary in vivo studies
Cathryn Helena Stanford Driver a, Thomas Ebenhan a, Zoltan Szucs c, Mohammed Iqbal Parker d,
b
Jan Rijn Zeevaart a,e, , Roger Hunter
⁎
a
South African Nuclear Energy Corporation, Radiochemistry and NuMeRI PreClinical Imaging Facility, Elias Motsoaledi Street, R104 Pelindaba, North West 0240, South Africa
Department of Chemistry, University of Cape Town, Cape Town, South Africa
b
c
MTA Atomki, Debrecen, Hungary
d
Department of Medical Biochemistry and Institute for Infectious Disease and Molecular Medicine, University of Cape Town Medical School, University of Cape Town, Cape Town, South Africa
Preclinical Drug Development Platform, North West University, Potchefstroom, South Africa
e
a r t i c l e i n f o
a b s t r a c t
Article history:
Introduction: The compound named 4-[10-(4-(2,5-dioxo-2,5-dihydro-1H-pyrrol-1-yl)butanamido)decyl]-11-
[10-(β,D-glucopyranos-1-yl)-1-oxodecyl]-1,4,8,11-tetraazacyclotetradecane-1,8-diacetic acid is a newly synthe-
sised molecule capable of binding in vivo to albumin to form a bioconjugate. This compound was given the
name, GluCAB(glucose-chelator-albumin-binder)-maleimide-1. Radiolabelled GluCAB-maleimide-1 and subse-
quent bioconjugate is proposed for prospective oncological applications and works on the theoretical dual-
targeting principle of tumour localization through the “enhanced permeability and retention (EPR) effect” and
glucose metabolism.
Methods: The precursor, GluCAB-amine-2, and subsequent GluCAB-maleimide-1 was synthesised via sequential
regioselective, distal N-functionalisation of a cyclam template with a tether containing a synthetically-derived
β-glucoside followed by a second linker to incorporate a maleimide moiety for albumin-binding. GluCAB-
amine-2 was radiolabelled with [64Cu]CuCl2 in 0.1 M NH4OAc (pH 3.5, 90 °C, 30 min), purified and converted
post-labeling in 0.01 M PBS to [64Cu]Cu-GluCAB-maleimide-1. Serum stability and protein binding studies were
completed according to described methods. Healthy BALB/c ice (three groups of n = 5) were injected intrave-
nously with [64Cu]Cu-TETA, [64Cu]Cu-GluCAB-amine-2 or [64Cu]Cu-GluCAB-maleimide-1 and imaged using
microPET/CT at 1, 2, 4, 8 and 24 h post-injection. Biodistribution of the compounds were determined ex vivo
after 24 h using gamma counting.
Received 6 October 2020
Received in revised form 14 December 2020
Accepted 2 January 2021
Keywords:
Cancer
Radiopharmaceuticals
Macrocyclic ligands
Copper-64
Albumin-binding
Results: GluCAB-maleimide-1 was synthesised in five consecutive steps with an overall yield of 11%. [64Cu]Cu-
GluCAB-amine-2 (97% labelling efficiency) was converted to [64Cu]Cu-GluCAB-maleimide-1 (93% conversion;
90% radiochemical purity). Biodistribution analysis indicated that the control compounds were rapidly and al-
most completely excreted as compared to [64Cu]Cu-GluCAB-maleimide-1 that exhibited a prolonged biological
half-life (6–8 h). Both, [64Cu]Cu-GluCAB-maleimide-1 and -amine-2 were excreted through the hepatobiliary sys-
tem but a higher hepatic presence of the albumin-bound compound was noted.
Conclusions, advances in knowledge and implications for patient care: This initial evaluation paves the way for fur-
ther investigation into the tumour targeting potential of [64Cu]Cu-GluCAB-maleimide-1. An efficient targeted
radioligand will allow for further development of a prospective theranostic agent for more personalized patient
treatment which potentially improves overall patient prognosis, outcome and health care.
© 2021 Elsevier Inc. All rights reserved.
1. Introduction
necessary therefore to understand cancer-cell tumourigenesis and the
tumour micro-environment. Tumours develop as a result of cancerous
The principle of targeted cancer diagnosis or treatment involves the
use of drug-delivery systems that improve the efficiency of drug uptake
by the tumour [1]. In designing a selective cancer targeting agent, it is
cells that are fast-growing and which over-express many cell-surface re-
ceptors, each associated with the provision of metabolites for cell
growth and division [2]. New cancer diagnostics or therapeutics can be
designed to have a high affinity for one of these up-regulated cancer
cell-surface receptors leading to the compound's accumulation within
the tumour cell (‘Active’ targeting). Active targeting ligands include
monoclonal antibodies (mAbs) as well as antibody fragments, proteins,
⁎
Corresponding author at: Preclinical Drug Development Platform, North West Univer-
sity, Potchefstroom, South Africa.
E-mail address: janrijn.zeevaart@necsa.co.za (J.R. Zeevaart).
https://doi.org/10.1016/j.nucmedbio.2021.01.001
0969-8051/© 2021 Elsevier Inc. All rights reserved.

C.H.S. Driver, T. Ebenhan, Z. Szucs et al.
Nuclear Medicine and Biology 94–95 (2021) 53–66
peptides, and small molecules that target a number of different antigens
and receptors.
polymethylene linker, as well as two carboxymethylene groups at the
other cyclam nitrogens for radionuclide coordination and stability.
This ligand is theoretically capable of being labelled with a radionuclide,
and thereafter, the expectation was that the covalent addition of circu-
lating, endogenous human serum albumin (HSA) [17] to the maleimide
via Michael addition of a cysteine thiol group, would furnish the active
bioconjugate.
The protein carrier and passive targeting agent selected, human
serum albumin (HSA), has a molecular weight of 66.5 kDa (large enough
for the EPR effect), is the most abundant serum protein (35–50 g/L,
50–60%), and has a half-life of around 19 days, making it very stable in
context [18,19]. HSA has been used as a protein carrier in the develop-
ment of new targeting agents [17], such as FDA-approved Abraxane
(paclitaxel-albumin nanoparticle) for the treatment of breast cancer
[20]; and for improving the biological half-life, pharmacokinetics and
overall therapeutic efficacy of radiopharmaceuticals, such as 177Lu-
DOTA-TATE and 177Lu-PSMA [21–23]. Small molecule binding to albu-
min can either be covalent – a Michael addition of the thiol of
cysteine-34 to a maleimide [17,19,21] – or reversible, non-covalent –
physical interactions of 4-(p-iodophenyl)butyric acid and Evans Blue
(azo dye) with the protein [22].
Adding a glucose moiety to the molecule can be pharmacologically
advantageous; firstly, the glucose increases the hydrophilicity and
therefore the aqueous solubility of the ligand and secondly, it plays a
supportive role as an active targeting agent on a cellular level. The con-
cept of conjugating glucose as the active targeting agent was based on
the well-known expression of glucose transporters (GLUT) on the cell
surface, and up-regulation on most cancer cells owing to their high glu-
cose requirement for provision of energy and metabolic needs [24,25].
The most widely used radiopharmaceutical that exploits this property
for cancer diagnosis is [18F]Fluorodeoxyglucose ([18F]FDG) [26] and
based on this same principle of GLUT and glycolysis targeting, a few
other glucose analogue targeting agents have also been developed
[27–30]. Most recently, a glucose-targeted probe (1380 g/mol) for pho-
toacoustic and fluorescent imaging was proven to have cell uptake
through GLUT [31].
For the GluCAB design the tetra-amine macrocycle cyclam was iden-
tified as the chelating agent as it is cheap and readily available and
cyclam derivatives have been widely used for the complexation of a
number of different metals with very good in vivo stability for a range
of medicinal applications [32–35]; with furthermore, these cyclam de-
rivatives are often used as the chelator of choice for complexation of
copper-64 for PET imaging purposes.
The tumour microenvironment is characterized by interstitial hyper-
tension, hypoxia, low extracellular pH, decreased lymphatics, and the
angiogenesis of blood vessels with defective architecture [3,4]. These ab-
normalities result in the leakage of blood plasma components into tu-
mour tissue, where they are then retained. The latter explains why
certain macromolecular drugs (>40 kDa for the renal excretion thresh-
old [5]) preferentially accumulate within the tumour, a phenomenon
noted by Matsamura and Maeda and termed the “enhanced permeabil-
ity and retention (EPR) effect” [6]. New cancer diagnostics or therapeu-
tics can be also be designed to exploit the tumour microenvironment
and the EPR effect (‘passive’ targeting) [7,8]. Passive targeting agents in-
clude polymers, liposomes, nanoparticles, and some larger proteins. Ac-
tive and passive targeting can be used in combination to improve the
efficacy of drug delivery to the tumour site and uptake by the cancer
cells.
A theranostic radiopharmaceutical contains a radionuclide and can
provide both a diagnostic as well as a therapeutic option depending on
the radioisotope that is coordinated to it [9]. The design of a targeted
theranostic radiopharmaceutical for oncological purposes requires care-
ful consideration of the physical decay properties of the radioisotope to
be used, the specific in vivo targeting mechanism for the tumour and the
clearance of the compound from other tissues so that the radioisotope,
the targeting molecule, the chelating agent and the linker are all com-
patible with each other for maximum in vivo stability [10]. Taking
these factors into consideration, the general aim of this project was
therefore to develop a new dual-target bioconjugate greater than
40 kDa in size that could be radiolabelled with a therapeutic or a diag-
nostic isotope for potential theranostic purposes.
The diagnostic isotope selected was copper-64 (t1/2 = 12.7 h), which
decays though positron emission (17%, 655 keV), electron capture/
Auger-emissions (44%) and beta emission (39%, 573 keV) [11,12].
Copper-64 and its other isotopes (60Cu, 61Cu, 62Cu, 67Cu) have been
shown to complex to many chelating agents with high stability and
have been used in many radiopharmaceutical applications [11–15].
While copper-64 and the other copper isotopes are mostly used for di-
agnostic purposes, copper-67 (t1/2 = 61.8 h) emits beta (β−) particles
of 580 keV (100%) (penetration range of 2.1 mm) with gamma rays of
92 and 184 keV being useful for therapeutic applications. Copper-64
and copper-67 together form a theranostic isotope pair that can be
used for diagnosis and therapy [12,16]. This 64Cu/67Cu-theranostic is a
favourable option for a theranostic agent, however, the limitation with
using copper-67 is its high production cost and limited availability. Re-
cently, copper-64 has been investigated as a single theranostic isotope
for both its proven diagnostic use and potential therapeutic capabilities
(beta and Auger-emissions), however use as a therapeutic isotope is
limited by the high dosage required to achieve therapeutic value [11].
The compound proposed for development as a potential theranostic
radioligand is shown in Fig. 1 and was designed to attach to a larger pas-
sive targeting agent such as albumin to form a bioconjugate. This ligand
will be referred to as GluCAB (Glucose-Chelator-Albumin-binder).
GluCAB-maleimide-1 (1) comprises a distally N-di-functionalised
cyclam chelating agent, functionalised through sequential N-alkylation
with a glucose moiety and a maleimide, each contained at the end of a
The development of GluCAB as a new targeted 64Cu/67Cu-radioligand
for potential theranostic applications was planned in a phased approach.
The first phase encompassed the ligand synthesis and radiolabelling
while the subsequent phases would involve in vitro stability along with
in vivo metabolism in healthy animals (phase 2), and in vivo tumour
targeting in a xenograft animal model (phase 3). Herein is described the
synthesis of the desired GluCAB-maleimide-1 followed by techniques
for radiolabelling of this compound with copper-64 (phase 1). In addition,
phase 2, for a preliminary understanding of the in vivo characteristics of
[
64Cu]Cu-GluCAB-maleimide-1 and its albumin-binding potential, is also
reported through the use of small animal PET/CT imaging followed by
post mortem biodistribution.
Fig. 1. The structure of GluCAB-maleimide-1.
54

C.H.S. Driver, T. Ebenhan, Z. Szucs et al.
Nuclear Medicine and Biology 94–95 (2021) 53–66
2. Results and discussion
group participation by the 2-benzoyl group on the α-face. A C-10 alkyl
chain was used for the linker in order to minimize steric interference be-
tween the targeting agent-chelator and the biomolecule carrier (albu-
min) during biological targeting. In a previous study that investigated
the effects of the linkers of albumin-binding compounds on the in vitro
and in vivo properties, it was found that an alkyl linker performed
slightly better than a larger, more hydrophilic PEG linker [38]. The syn-
thesis of the second linker, Boc-protected amine (6), for alkylative at-
tachment to the chelator, was achieved according to a series of
standard functional group conversion reactions as indicated in Scheme
2B. 10-amino-1-decanol (11) was obtained through a Gabriel reaction
by first treatment of 10-bromo-1-decanol with potassium phthalimide
followed by subsequent hydrolysis of the phthalimide (10) formed
with hydrazine hydrate. The resultant amine was then protected as a
t-butyl carbamate (12) followed by conversion of the hydroxyl group
to bromide (6) under Appel conditions. The yields of each step were
high (>85%) throughout.
GluCAB-maleimide-1 was synthesised according to the
retrosynthetic analysis shown in Scheme 1. The synthesis required the
sequential N-alkylation of di-(tert-butyl acetate) cyclam (5) using teth-
ered glycosyl bromide (4) (active targeting moiety) followed by Boc-
protected amine (6) as functionalised linkers and subsequent functional
group inter-conversions, and will be discussed in the text. The preferred
order of alkylation between linkers (4 and 6) and the challenges faced
during synthesis will further be discussed.
2.1. Synthesis of linkers
Synthesis of the C-1, O-alkylated glycosyl bromide (4) is shown in
Scheme 2A. It involved using the one-pot glycosylation method of
Murakami et al. [36] in which perbenzoylated glucose (8) was con-
verted, almost exclusively, to its α-glycosyl iodide (9) using iodine
and hexamethyldisilazane (in situ producer of iodotrimethylsilane
(Me3SiI)) under Lewis acid catalysis with zinc chloride. The formation
of the glycosyl iodide α-anomer using Me3SiI has been previously
proven and studied mechanistically [37]. Iodide (9) was not purified
but substituted directly with 10-bromodecanol via activation with
1 eq. of ZnCl2 to form the β-anomer of glycoside (4) (59% yield over
the two steps).
2.2. Functionalisation of cyclam
Distal, sequential dialkylation of cyclam (5) with bromide-linkers
(4) and (6) required prior protection of two of the distal nitrogens of
cyclam with alkoxycarbonylmethylene groups. However, alkylation
studies with t-butyl bromoacetate and a base resulted in a complex mix-
ture containing six alkylated cyclam compounds, amongst which were
three difunctionalised regioisomers that included the desired 1,8-N,N′-
di-functionalised product. The ratio of mono, di and trisubstituted prod-
The β-glycoside (4) formed diastereoselectively via exclusive β-
addition of 10-bromodecanol onto the intermediate oxocarbenium ion
(from iodide ionization with the Lewis acid ZnCl2), due to neighbouring
Scheme 1. Retrosynthetic analysis of GluCAB-maleimide-1 and the synthesis of GluCAB-amine-2 (a) Glycosyl bromide (4), K2CO3, MeCN, 60 °C, 16 h (58%), (b) carbamate (6), K2CO3,
MeCN, 80 °C, 16 h (74%), (c) NaOMe, MeOH, 1 h (71%), (d) TFA:CH2Cl2 (1:1), 16 h (69%)
55

C.H.S. Driver, T. Ebenhan, Z. Szucs et al.
Nuclear Medicine and Biology 94–95 (2021) 53–66
Scheme 2. Synthesis of linkers - glycosyl bromide (4) and Boc-protected amine (6) A) (a) BzCl, pyridine, 0 °C – RT, (89%), (b) I2, (Me3Si)2, ZnCl2, CH2Cl2, 16 h, RT (c) 10-bromodecanol,
ZnCl2, Molecular sieves 4 Ǻ, CH2Cl2 (59% over steps a and b). B) (a) Potassium phthalimide, DMF, 100 °C, 20 h, (97%) (b) Hydrazine hydrate, EtOH, reflux, 24 h (88%), (c) (Boc)2O,
DCM:MeOH 4:1, 3 h, (89%), (d) CBr4, PPh3, DCM, (92%)
ucts could be slightly manipulated by adjusting the equivalents of t-
butyl bromoacetate but a mix of the products was still obtained. The
maximum amount of disubstituted product could be obtained by the ad-
dition of 3 equivalents of the t-butyl bromoacetate over 6 h in incre-
ments of 0.5, using potassium carbonate as base (2.5 equiv). These
three regioisomers could, with difficulty, then be purified from each
other using column chromatography (CH2Cl2:MeOH 5:1), but the yield
of each (1,4-N,N′ = 16%; 1,8-N,N′ = 41%; 1,11-N,N′ = not determined)
was very low.
A method developed by Pandya et al. [39] (Scheme 3) reported the
selective synthesis of cyclam (5) starting from 1,4,8,11-tetraazatricyclo
[9.3.1.14,8]hexadecane (tricyclic bis-aminal cyclam (13)) via a quater-
nary ammonium salt intermediate. Cyclam (13), synthesised by alkyl-
ation of cyclam with formaldehyde (2 eq), comprises the bridging of
two of the six-membered rings in cyclam as aminals in stable chair con-
formations, which makes the outward-facing lone pairs more available
for alkylation (at positions 1 and 8). Following quaternization, the meth-
ylene bridges are base-hydrolysed to afford symmetrical trans-
disubstituted cyclam (5).
mechanism proposed in Supporting Information Scheme S1. This reac-
tion was repeated numerous times with similar results and could there-
fore represent a novel method for mono-alkylation of a cyclam
macrocycle. 1,8-N,N′-disubstituted cyclam (5) was then obtained by
reacting (13) with excess t-butyl bromoacetate (6 eq) in acetonitrile
for 5 days at 70 °C and precipitation of a crude product with diethyl
ether. The 1,8-N,N′-disubstituted cyclam (5) was recrystallized from
acetonitrile with a small percentage (<5%) of methanol in a 20% yield.
The 1H and 13C NMR spectral data (see Supporting Information
Fig. S1) was in accordance with that reported in the literature for a
1,8-N,N′-disubstituted cyclam [40]. Briefly,1H NMR indicated the re-
quired peaks for the cyclam macrocycle and the two t-
butyloxycarbonylmethylene groups distally: one methylene singlet
(4H, 3.30 ppm); one t-butyl group (18H, 1.43 ppm); multiplets at
1.72 ppm (4H), 2.60 ppm (4H), 2.69 ppm (8H) and 2.74 ppm (4H) for
the symmetrically equivalent methylene groups of a distally disubsti-
tuted cyclam macrocycle. Similarly, 13C data showed only 5 cyclam res-
onances. The data for 1,8-N,N′-cyclam (5) was also distinguishable [40]
from the previously isolated and identified 1,4-N,N′-regioisomer (see
Supporting Information, Fig. S2), as well as the 1,11-isomer, based on
symmetry considerations. Although low-yielding, this procedure was
found to be reproducible, furnished high-purity (5) and could be scaled
up. GluCAB-amine-2 was then synthesised via standard alkylation and
deprotection strategies as shown in Scheme 1.
Following the method [39] as described for the alkylation of (13) to
form (5), the product isolated proved rather to be the mono-alkylated
cyclam (14) (65% yield) in which the aminal bridge at the site of alkyl-
ation was deprotected in situ as a result of water present in the reaction
(originating from crystal structure of recrystallized (13)) via the
Scheme 3. Selective synthesis of trans-disubstituted cyclam (5) according to Pandya et al. [39] (a) Formaldehyde (2 eq), H2O, RT, 2 h (b) BrCH2CO2t -Bu (4 eq), MeCN, 60 °C, 24 h (c) 3 M
NaOH, RT, 3 h and the mono-alkylated cyclam (14) obtained (d) BrCH2CO2t -Bu (4 eq), MeCN, 60 °C, 48 h, (e) 3 M NaOH, 16 h, (65%)
56

C.H.S. Driver, T. Ebenhan, Z. Szucs et al.
Nuclear Medicine and Biology 94–95 (2021) 53–66
Briefly, cyclam (5) was alkylated using glycoside (4) (0.45 eq – limiting
reagent to minimize formation of di-glycosylated product) to form
mono-glycosylated cyclam (15) (58%) (full synthesis in Supporting In-
formation), followed by alkylation using carbamate (6) (2.0 eq) to
yield cyclam (3) (74%). Both alkylations used potassium carbonate
(3.0 eq) in acetonitrile with heating (60 or 80 °C) for 16 h. Alkylation
with glycoside (4) was performed first so as to introduce UV activity
into the structure, which assisted with chromatography and isolation.
Similarly, the order of deprotecting the protecting groups of cyclam
(3) (benzoate esters, tert-butyl esters and tert-butyl carbamate) was
completed in a way that ensured the easiest purification. Deprotection
of the benzoate esters of (3) to yield cyclam (16) (Supporting Informa-
tion) was completed under nucleophilic conditions using sodium
methoxide, towards which the tert-butyl esters and carbamate were
stable. Compound (16) could be purified using silica-based column
chromatography. Further Boc- and tert-butyl ester deprotection of
(16) was carried out under acidic conditions using TFA to afford
GluCAB-amine-2 (2), for which full characterisation was carried out
(see the Supporting Information). The glycosyl functionality of
GluCAB-amine-2 was found to be stable under the acidic Boc-
deprotection conditions using TFA, in which only a very small amount
(<10%) of by-product was fortunately observed. The purification of
free GluCAB-amine-2 using column chromatography (silica gel 60,
DCM:MeOH:NH4OH 7:2.5:0.5), however, proved challenging as a result
of polarity, which caused the compound to adhere to the stationary
phase. Consequently, while the reaction conversion seemed high
(~90% according to TLC), the reaction yield was fairly low (<50%). The
stationary phase was then changed to an Alumina-N matrix in which
elution with DCM:MeOH:NH4OH (8:1.7:0.3; 7:2.5:0.5; 6.5:3:0.5) re-
moved the impurities, after which the product could be eluted with
40% H2O in MeOH. Once concentrated, the product was desalted on
a solid phase extraction (SPE) C18 light cartridge using water, and
eluted from the cartridge using 50% MeOH in H2O. The product was
dried under an air stream, re-dissolved in water (1 mL), frozen and
lyophilised to yield a colourless solid powder pellet. The yield of
GluCAB-amine-2 obtained from this purification was improved to
69%, leading to an overall yield, over the four steps from cyclam 5,
of 21%.
Insertion of the two linkers in the N4 and the N11 position of
GluCAB-amine-2 ensured placement of the glucose targeting moiety
distally away from the albumin carrier (that would eventually be at-
tached to the terminal amine position) allowing for minimal interfer-
ence between the two positions. This structure also placed the two
coordinating carboxylic acid groups opposite to each other to form a
more stable, octahedral type geometry when coordinating the radioiso-
tope [40].
2.3. Maleimide insertion
The concept of a GluCAB bioconjugate requires the in vivo binding of
the synthesised ligand to albumin. Towards this end the primary amine
of GluCAB-amine-2 was converted into
a maleimide (GluCAB-
maleimide-1), thereby providing a Michael acceptor for capture of the
free Cys-34 thiol group of albumin as described previously [17,19,21].
GluCAB-maleimide-1 was synthesised through a mixed anhydride pep-
tide coupling strategy in which 4-maleimidobutyric acid NHS-ester (18)
(synthesised from 4-maleimidobutyric acid (17)) (Scheme 4) dissolved
in DMF was reacted with GluCAB-amine-2 in an aqueous medium
(model reactions and different peptide coupling strategies are displayed
in Supporting Information Schemes S2 and S3). The advantage of this
NHS-coupling method over other mixed anhydride methods was that
the aqueous medium was more suitable for miscibility and reaction of
the highly polar GluCAB-amine-2. However, equally, this method suf-
fered from sensitivity of the NHS-ester (18) towards hydrolysis in an
aqueous medium, especially at an acidic (pH < 7) or basic pH
(pH > 8). The NHS-ester (18) was found to be most stable when dis-
Scheme 4. Synthesis of GluCAB-maleimide-1 (a) NHS, DIC, DCM, 2 h, RT (72%) (b) 0.01 M PBS, pH 7.4, 1 h, RT (55%) (mz[+H] = 955.6) (hydrolysed maleamic acid product 1a (mz[+H] =
973.6)).
57

C.H.S. Driver, T. Ebenhan, Z. Szucs et al.
Nuclear Medicine and Biology 94–95 (2021) 53–66
solved in an organic solvent, which resulted in optimized reaction con-
ditions involving the addition of small increments of the NHS-ester in
DMF to the amine in a 0.01 M PBS buffer (pH 7.4) at room temperature.
The conversion from starting material (GluCAB-amine-2) to a single
product (GluCAB-maleimide-1) could be observed on TLC. This product
was very soluble in water because of its high polarity and needed to be
purified using a Sep-Pak C-18 Plus light cartridge as described in the
methods. During purification, it appeared that some of the maleimide
product was degrading (hydrolysing) and only approximately 50% of
GluCAB-maleimide-1 product was isolated. HPLC-MS analysis of (1)
after 1 h in water indicated two peaks (8.5 and 9.0 min) in the UV-
trace (210 nm). The 8.5 min peak (mz[+H] = 973.6) proved to be the
maleamic acid compound (1a) (Scheme 4) while the 9.0 min peak
(mz[+H] = 955.6) corresponded to the GluCAB-maleimide-1.
Overall, the maleimide functionality proved to be very sensitive to-
wards hydrolysis to the open maleamic acid form, especially under acidic
conditions (pH < 6.5), at higher temperatures, and in the presence of nu-
cleophiles. This sensitivity, therefore, had to be carefully considered dur-
ing any proposed radioisotope complexation, as maleimide hydrolysis
during this process and subsequent administration of the maleamic acid
compound in vivo was feared would have a serious effect on the
compound's biological efficacy due to inefficient albumin-binding.
. The formation of these bonds; however, requires a large amount of en-
ergy for the metal ion to overcome its kinetic inertness and bind to the
N- and O-donors with dissociation from its chloride ligands to form an
octahedral complex [44]. The temperature, therefore, is crucial for pro-
viding enthalpy of activation. The pH of the reaction also had a slight ef-
fect on complex formation in which the best complexation was
observed at pH 3.5, while at pH 9 the LE decreased by 15%. A possible
reason for this is that at the higher pH values, there are more negatively
charged ions present in the reaction solution to compete for the posi-
tively charged copper ions, making the binding rate to the macrocycle
slower and less efficient.
These optimized conditions (high temperature and low pH) for
radiolabelling of GluCAB-amine-2 with copper-64 preclude complex
formation using GluCAB-maleimide-1 because of incompatibility with
the maleimide functional group. The maleimide incorporation using
NHS-ester (18) as described, would therefore need to occur post-
labelling.
2.4.2. Post-labelling conversion of [64Cu]Cu-GluCAB-amine-2 to [64Cu]Cu-
GluCAB-maleimide-1
[
64Cu]Cu-GluCAB-amine-2 was converted to [64Cu]Cu-GluCAB-
maleimide-1 by reaction with a 4-maleimidobutyric acid NHS-ester
(18) for 1 h in 0.01 M PBS at pH 7.4 and an incubation temperature of
40 °C. Pleasingly, the maleimide product (retention time = 13.5 min;
90% radiochemical purity; conversion of 93%) proved to be stable in
the PBS buffer during the reaction process with no noticeable hydrolysis
of the maleimide to the maleamic acid (retention time = 12.5 min)
(Fig. 2). The UV-VIS retention time of the cold copper complexes
matched the retention time of the radiolabelled complexes (data not
shown). The time required for the reaction and purification (~1.5–2 h)
was sufficiently lower than the half-life of copper-64 (12.7 h); thus,
the process was suitable for a post-labelling conversion.
2.4. Radiolabelling
2.4.1. Preparation of [64Cu]CuCl2 and radiolabelling of GluCAB-amine-2
Copper-64 was obtained through irradiation of natural CuO (69.17%
abundance of 63Cu) in the SAFARI-1 research reactor at Pelindaba, Necsa
via the 63Cu(n,γ)64Cu nuclear reaction [41]. This method produces
copper-64 of a lower specific activity as compared to the more widely
used method of enriched 64Ni-target irradiation with low energy pro-
tons (11–14 MeV; 64Ni(p, n)64Cu) [42,43] but it is less expensive and
more easily accessible than the cyclotron-produced copper-64. A
[
64Cu]CuCl2 solution was prepared with a specific (molar) activity of
2.5. In vitro studies
1450 GBq/g (115 MBq/μmol).
As previously mentioned copper-64 is known to form kinetically
inert and thermodynamically stable complexes with a variety of cyclam
chelators. These complexes form under a wide range of reaction condi-
tions that need to be compatible with the functional groups attached to
the macrocycle. Earlier, we had established that the maleimide function-
ality of GluCAB-maleimide-1 had revealed a high sensitivity towards hy-
drolysis under acid conditions and in aqueous solutions at high
temperatures. Therefore, in pursuit of the 64Cu-labelled GluCAB
radioligand, complexation of GluCAB-amine-2 with copper-64 was in-
vestigated to first determine optimal conditions, with the possible in-
tention to incorporate the maleimide subsequently.
2.5.1. Stability studies and protein binding of [64Cu]Cu-GluCAB-amine-2
and [64Cu]Cu-GluCAB-maleimide-1
Preliminary stability studies in 0.01 M PBS buffer were performed,
firstly to test the tendency of the maleimide towards hydrolysis and sec-
ondly, to establish whether copper was lost from the chelator. For [64Cu]
Cu-GluCAB-maleimide-1, it was found that approximately 14% of the
maleimide had hydrolysed after 24 h (0.01 M PBS (pH 7.4); room tem-
perature) (Supporting Information, Fig. S3) with only a 1.5% re-
occurrence in free copper-64 species.
Protein binding studies provide an indication of how well, if at all, a
compound will bind to proteins in the blood, and especially to HSA, con-
sidering that this protein constitutes 60% of all serum proteins. Binding
of the GluCAB compound to HSA was seen to be an essential part of
the targeting mechanism and [64Cu]Cu-GluCAB-maleimide-1 was de-
signed in such a manner (maleimide functionality) that a high percent-
age of protein binding was anticipated as reported in literature [19]. A
protein binding study was carried out comparing [64Cu]Cu-GluCAB-
maleimide-1 and [64Cu]Cu-GluCAB-amine-2 (no maleimide functional-
ity), (Fig. 3). The radioactivity present in the proteins (precipitated and
centrifuged out from serum) was measured following incubation (0, 1, 2
and 24 h) with each compound at 37 °C. Upon addition to the serum,
To this end, GluCAB-amine-2 was reacted with the prepared [64Cu]
CuCl2 (115 MBq/μmol; 0.33 eq) in a 0.1 M NH4OAc solution at different
pH's, temperatures, and incubation times. An overview of the experi-
mental series is given in the Supporting Information Table S1. The best
labelling efficiency (LE) for complexation of GluCAB-amine-2 with
copper-64 was found to be 98% with an average LE of 95% under the
same conditions (pH 3.5; 95 °C; 30 min) irrespective of radioisotopic ac-
tivity concentration. Radio-HPLC analysis of the reaction indicated that
the free [64Cu]CuCl2 was only minimally retained on the reverse-phase
HPLC column, eluting with a retention time of 3–4 min, while the
slightly less hydrophilic [64Cu]Cu-GluCAB-amine-2 eluted at around
11.5 min. The LE decreased significantly (p < 0.05) upon incubation at
higher pH values (pH 5 (86%) and 9 (80%)), for shorter reaction times
(20 min (61%) and 10 min (39%)) or at lower temperatures (45 °C
(5%), 70 °C (29%) or 80 °C (81%)).
[
64Cu]Cu-GluCAB-maleimide-1 was immediately 85% protein bound
which further increased to a maximum of 90% within 2 h that was
sustained through to 24 h. The small percentage that remained ‘un-
bound’ could be attributed to the nature of the radiolabelling process
in which 3% radioactivity remained as free [64Cu]CuCl2 and the residual
as GluCAB-amine-2. Conversely, [64Cu]Cu-GluCAB-amine-2 showed a
very low (30%) protein binding upon addition to the serum, which
reached 50% protein binding after 24 h. This can be considered non-
specific protein association, as no maleimide group was present. Possi-
ble non-specific interactions included charge interactions (between
The temperature of the reaction had the most detrimental impact on
the radiolabelling since at 45 °C hardly any complexation occurred (<5%
LE). Copper (II) is a medium hard/soft Lewis acid and forms strong
bonds with the nitrogen of amino groups of macrocycles such as those
in cyclam, especially when there are carboxylate pendant arms [33,34]
58

C.H.S. Driver, T. Ebenhan, Z. Szucs et al.
Nuclear Medicine and Biology 94–95 (2021) 53–66
Fig. 2. Radio-HPLC chromatograms of free [64Cu]CuCl2 (green peak – RT = 3.4 min); [64Cu]Cu-GluCAB-amine-2 (pink peak – RT = 11.5 min; 97% radiochemical purity) and post labeling
conversion to [64Cu]Cu-GluCAB-maleimide-1 (blue peak – RT = 13.5 min; 90% radiochemical purity, 93% conversion).
of Copper-64 macrocyclic complexes. This study noted that 64Cu-
the protonated amino group and carboxylic acid functionalities), hydro-
gen bonding, and other hydrophobic interactions (between ligand and
serum protein).
labelled macrocycles (TETA, NOTA, DOTA, cyclam) exhibited very good
complex stability with only 1.7–4.2% transchelation to serum proteins
but a [64Cu]Cu-NOTA conjugated to a single-domain antibody indicated
40% transchelation. The resistance to transchelation noted for the
cyclam macrocyles indicates that the % protein binding noted for the
64Cu-GluCAB complexes should be as a result of the maleimide group
and some non-specific protein association, as indicated, with minimal
copper released; however, the increase of transchelation upon
conjugation of the copper-complex to a biomolecule warrants further
investigation into the kinetic stability of the GluCAB complexes in vivo.
Serum stability studies were completed for both the [64Cu]Cu-
GluCAB-maleimide-1 and the [64Cu]Cu-GluCAB-amine-2. Radio-HPLC
analysis seemed to indicate high stability of both 64Cu-complexes to
de-metallation as well as ligand robustness at 37 °C over the 24 h
since no degradation products or increase in the amount of free
copper-64 was observed (Supporting Information, Fig. S4).
A concern however, regarding the % protein binding noted for
GluCAB-amine-2 and the stability of the complexes as determined by
absence of free copper-64, is the possible release of copper ions from
the cyclam chelator with transchelation and binding of the copper to
other serum proteins. Zarschler et al. [45] defined this challenge and
sought to establish in vitro assays for determining the in vivo stability
[
64Cu]Cu-TETA which indicated high serum stability and low protein
binding (4.2 2.3% resistance to serum protein transchelation) [45] is
to be used as a negative control for in vivo studies as this has been
proven to be rapidly metabolised and cleared in vivo [46].
Fig. 3. Percentage binding of [64Cu]Cu-GluCAB-maleimide-1 and [64Cu]Cu-GluCAB-amine-2 to proteins in serum over a 24 h period at 37 °C (mean SD; n = 3).
59

C.H.S. Driver, T. Ebenhan, Z. Szucs et al.
Nuclear Medicine and Biology 94–95 (2021) 53–66
2.6. In vivo studies
According to current theories [47] on drug metabolism, the most
common routes of drug elimination is renal excretion (glomerular filtra-
tion of small, hydrophilic, non-protein bound drugs or tubular secretion
of larger, protein-bound drugs); or hepatobiliary excretion (more lipo-
philic compounds). In the study of the metabolism and excretion of
drug-radioisotope complexes, it has been noted that the clearance path-
way, specifically of bifunctional chelate conjugates, is significantly influ-
enced by the lipophilicity and net charge of the complex [46]. The more
lipophilic a complex is the higher the liver accumulation will be and the
slower the systemic clearance. Similarly, a conjugate with a positively-
charged complex exhibits slow excretion through the liver but con-
versely, a negatively-charged complex is more rapidly and efficiently
cleared through the kidneys. The routes of excretion noted from the
MIP PET images for all three compounds are in line with the above men-
tioned aspects on compound metabolism and excretion. [64Cu]Cu-TETA
is a relatively small, hydrophilic molecule with a net negative complex
charge while [64Cu]Cu-GluCAB-maleimide-1 and [64Cu]Cu-GluCAB-
amine-2 are more lipophilic with a neutral complex charge. As identi-
fied by the amount of activity in the specific organs involved with excre-
tion (kidneys, bladder, liver, intestines), [64Cu]Cu-TETA was excreted
through the renal system while [64Cu]Cu-GluCAB-maleimide-1 and
2.6.1. MicroPET/CT imaging studies in healthy Balb/c mice
The prospective capability of GluCAB for tumour targeting can only
be evaluated in vivo since the first mechanism of uptake relies on bind-
ing of the GluCAB-maleimide-1 to albumin and extravasation into the
tumour through the EPR effect, which necessitates a living biological
system. Valuable understanding of the tracer administration, metabo-
lism, excretion and biodistribution of the compound needed to be
achieved before further actual tumour targeting studies in mice could
be attempted. This concept of altered biodistribution of [64Cu]Cu-
GluCAB-maleimide-1 (and consequently the GluCAB bioconjugate)
was therefore determined in healthy Balb/c mice using microPET/CT im-
aging followed by post mortem radioanalysis of excised organs and tis-
sue. This was compared to the behavior of two other compounds;
[
64Cu]Cu-TETA (negative control since rapid metabolism and excretion
was anticipated) and [64Cu]Cu-GluCAB-amine-2 (lacking the maleimide
functionality; predicting a faster excretion rate despite non-specific
serum protein binding). Similar compound concentrations of [64Cu]
Cu-TETA (100 μL; 1.2
0.22 MBq; 39 nmol/mouse), [64Cu]Cu-
GluCAB-amine-2 (150 μL; 5.6 MBq, 31 nmol/mouse) and [64Cu]Cu-
GluCAB-maleimide-1 (100 μL; 1.0 0.09 MBq, 31 nmol/mouse) were
administered for comparison of the groups (n = 5); no adverse side ef-
fects were noted. Subsequently, microPET/CT imaging was conducted
over a 24 h period for which maximum intensity projection (MIP)
allowed for a visual comparison between the biodistribution of the
three compounds (Fig. 4).
[
64Cu]Cu-GluCAB-amine-2 were eliminated to different extents via the
hepatobiliary system.
Although [64Cu]Cu-TETA and [64Cu]Cu-GluCAB-amine-2 have simi-
lar blood clearance rates, the difference between their excretion can po-
tentially be explained by: 1) the difference in molecular weight of the
compounds (432.5 g/mol versus 790.1 g/mol) making the smaller
TETA compound more suitable for rapid glomerular filtration; 2) the dif-
ference in charge between the two complexes (negative versus neutral),
which influences the hydrophilicity of the compounds; 3) the tethers
(C-10-aliphatic chains) attached to the chelator for [64Cu]Cu-GluCAB-
amine-2 that renders this compound more lipophilic and therefore
prone to hepatobiliary metabolism; and 4) the nominal non-specific
binding of [64Cu]Cu-GluCAB-amine-2 to serum proteins (as herein de-
termined) that may provoke more hepatic involvement.
The MIP images represent the expected
[
64Cu]Cu-TETA
biodistribution (Fig. 4A) as reported in literature [46], which is domi-
nated by a rapid tracer clearance from the blood pool and a major sys-
temic clearance by way of renal excretion (i.e. high activity present
only in the bladder within 1–2 h post injection). Furthermore, at 24 h
post-injection (p.i), only minimal activity was still evident in the animal.
By comparison, the [64Cu]Cu-GluCAB-amine-2 microPET/CT (Fig. 4B)
visualised the occurrence of activity within the hepatobiliary system,
suggesting a different route of excretion compared to [64Cu]Cu-TETA.
However; just as [64Cu]Cu-TETA, [64Cu]Cu-GluCAB-amine-2 had cleared
early-on in significant amounts and within 24 h p.i negligible
radioactivity concentration in all major organs (except for intestines)
was observed. MIP images of the mice administered with [64Cu]Cu-
GluCAB-maleimide-1 (Fig. 4C) showed that this compound had a
prolonged biological half-life and blood circulation as indicated by the
high overall systemic activity (a blue colour in the image) and clear vis-
ibility of the heart even at 8 h p.i as compared to [64Cu]Cu-GluCAB-
amine-2. These images also clearly indicated a high accumulation of ra-
dioactivity in the liver, intestines, and bladder from 1 to 8 h p.i. This ac-
tivity was mostly cleared after 24 h but with some hepatic radioactivity
activity still notable.
Further analyses included the image-guided quantification from
volumes-of-interest (VOI), yielding standard uptake values (SUV
(g/mL)); e.g. VOI and SUV indicative of the decay-corrected com-
pound concentration in the heart muscle (representing blood pool)
for each time point, allowing for an estimation of their biological
half-life. The heart activity concerning [64Cu]Cu-TETA and [64Cu]
Cu-GluCAB-amine-2 calculated from the 1 h PET image was 0.02
and 0.04 g/mL, respectively. This was deemed fairly low and
reflected the representation from the images. Therefore, the biolog-
ical half-lives of these two compounds were assumed to be <5 min,
but could not be exactly determined. However, analysis of the heart
region for [64Cu]Cu-GluCAB-maleimide-1 (1 h = 1.44 g/mL) indi-
cated the expected exponential decrease of the circulating activity
over the 24 h period (24 h = 0.10 g/mL) and estimated the biologi-
cal half-life as about 6–8 h. Other notable SUV for [64Cu]Cu-GluCAB-
maleimide-1 included the liver (max. 2.74 g/mL at 1 h to min.
0.29 g/mL at 24 h) and intestines (max. 0.79 g/mL at 4 h to min.
0.07 g/mL at 24 h).
Similar to [64Cu]Cu-GluCAB-amine-2, [64Cu]Cu-GluCAB-maleimide-
1, is a lipophilic and neutrally charged complex that shows metabolism
through the liver but with the difference of having a tethered maleimido
functional group. This study confirmed a markedly prolonged biological
half-life of [64Cu]Cu-GluCAB-maleimide-1 (6–8 h) compared to -amine-
2 (<5 min). This may be sufficient proof of the effectiveness of the
maleimide functionality for binding to albumin and therefore substan-
tially extending an unwanted-short tracer residence time in the blood
circulation. The albumin-binding of [64Cu]Cu-GluCAB-maleimide-1 in
turn affected the rate of liver metabolism, which seemed markedly re-
duced as inferred from the prolonged and high accumulation of activity
in the liver and intestines over the 24 h period. The high activity seen in
the liver could be enhanced by the prolonged plasma retention of the
compound and the fact that the liver is a highly perfused organ but albu-
min is known to be eliminated through renal (6%), hepatobiliary (10%)
and other catabolic (84%) mechanisms [48], and therefore the activity
which was seen in the liver and intestines is not unexpected for a cova-
lently bound GluCAB-albumin bioconjugate. While reversible albumin-
binding [22] allows for a slow release and accumulation of the drug at
the target site as a result of a longer circulation time, the release of the
small molecule implies that renal excretion will mostly be evident. Co-
valent albumin-binding can both, improve on the drug circulation
time as well as facilitate tumour delivery and retention there in through
the EPR-effect [21] but the drug is not easily released and the excretion
mechanism would be that of a larger, lipophilic construct through the
liver.
The accumulation of activity in the liver of the [64Cu]Cu-GluCAB-
maleimide-1 injected mice could also have been as a result of a few pos-
sible other processes: firstly, a loss of the copper(II) ions from the chela-
tor complex in the blood stream followed by subsequent binding of the
ionic copper to serum proteins and delivery to the liver for further
60

C.H.S. Driver, T. Ebenhan, Z. Szucs et al.
Nuclear Medicine and Biology 94–95 (2021) 53–66
Fig. 4. Maximum intensity projection microPET/CT images (anterior view) of BALB/c mice following injection with (A) [64Cu]Cu-TETA (100 μL; 1.2 0.22 MBq; 39 nmol/mouse), (B) [64Cu]
Cu-GluCAB-amine-2 (150 μL; 5.6 MBq, 31 nmol/mouse) and (C) [64Cu]Cu-GluCAB-maleimide-1 (100 μL; 1.0 0.09 MBq, 31 nmol/mouse). Image acquisition was done under isoflurane
anaesthesia (2%) at 1, 2 and 24 h (A/B) and at 1, 2, 4, 8 and 24 h (C) post injection (n = 3). (CT scan = 5 min; PET scan = 20 min); B = bladder; H = heart; L = liver; I = intestines.
Normalised scale bar 0–10 kBq/mL.
concentration (SUV) for [64Cu]Cu-TETA, [64Cu]Cu-GluCAB-amine-2 and
[
last image acquisition, the post mortem biodistribution analysis quantifies
the precise amount of radioactivity per organ allowing herein the robust
comparison of the 64Cu-labelled compounds as illustrated in Fig. 5.
As already indicative from 24 h MIP image in Fig. 4A, very low 64Cu-
activity amounts were obtained for [64Cu]Cu-TETA; indeed all organs
processing; or secondly, but less likely, transchelation of the copper
64Cu]Cu-GluCAB-maleimide-1 provided first such results. Following the
from the GluCAB complex to a copper-dependent enzyme, e.g., superox-
ide dismutase (SOD), that is highly abundant in liver cells [45]. 64Cu-
labelled cyclam and TETA have been proven to show enhanced resis-
tance to both serum protein and SOD transchelation [45]; however, it
has also been noted that a chelator-biomolecule complex is less kineti-
cally stable in vivo and therefore these processes and liver accumulation
for [64Cu]Cu-GluCAB-maleimide-1 need to be further investigated.
showed negligible TETA levels ex vivo (on average 0.20
0.06%ID/g)
with the only notable radioactivity found in the kidneys (0.37 and
0.40%ID/g). No notable difference between the activity in the organs of
mice injected with [64Cu]Cu-TETA and [64Cu]Cu-GluCAB-amine-2 was
observed - except for a 6-fold higher [64Cu]Cu-GluCAB-amine-2
2.6.2. Biodistribution studies
The image-guided quantification matched the illustrated distribution
behavior of the compounds with the normalised organ activity
61

C.H.S. Driver, T. Ebenhan, Z. Szucs et al.
Nuclear Medicine and Biology 94–95 (2021) 53–66
Fig. 5. Comparison of ex vivo biodistribution at 24 h after injection of [64Cu]Cu-TETA (n = 5), [64Cu]Cu-GluCAB-amine-2 (n = 1) and [64Cu]Cu-GluCAB-malemide-1 (n = 5) presented as
mean %ID/g SD; two-tailed Student's t-test comparing radiolabelled TETA with GluCAB-malemide-1 returned p < 0.05 (*), p < 0.01 (**) or p < 0.001 (***).
completed. The bifunctional chelator GluCAB-maleimide-1, has been
presence within the large intestine (2.31%ID/g) and 2.5-fold higher
64Cu]Cu-GluCAB-amine-2 hepatic presence (0.91%ID/g). This confirms
successfully synthesised, radiolabelled and preliminarily investigated
for its in vivo albumin-binding capabilities. Towards this end, GluCAB-
amine-2 was successfully radiolabelled with copper-64 and subse-
quently converted to its radiolabelled GluCAB-maleimide-1 derivative,
which proved to be very stable in serum. In vivo studies for [64Cu]Cu-
GluCAB-maleimide-1 clearly estimated a prolonged biological half-life
(6–8 h) as compared to the pharmacokinetics of [64Cu]Cu-GluCAB-
amine-2 (lacking the maleimide functionality), and although an ele-
vated liver uptake was noted, an adequate compound excretion was
seen within 24 h. These results have translational merit for continuation
to phase 3 of the investigation – the evaluation of [64Cu]Cu-GluCAB-
maleimide-1 towards tumour targeting in diseased animal models.
However, this study had some limitations which need to be considered
such as 1) the high temperature and low pH conditions required for che-
lator radiolabelling that are not compatible with the sensitive
maleimide functional group; and 2) the increased liver uptake - possibly
as a result of the complex charge, complex instability in vivo or SOD
transchelation within the liver. Further studies are under way address-
ing these limitations, − including in vitro studies to confirm targeting
of GLUT and the dual-targeting principle of the conjugate – and once ac-
complished, tumour targeting using [64Cu]Cu-GluCAB-PET/CT imaging
can be established.
[
the excretion patterns and conclusions drawn from the microPET/CT
image analysis. The [64Cu]Cu-TETA and [64Cu]Cu-GluCAB-amine-2 com-
pounds were rapidly eliminated from the system through the renal and
hepatobiliary pathways, respectively. The biodistribution analysis of
[
64Cu]Cu-GluCAB-maleimide-1 mice recorded the maximum 64Cu-
activity in the plasma (10.7
1.8%ID/g) and the liver (9.6
1.4%ID/
g) with the activity in all the other organs being <4%ID/g. Therefore,
the activity of [64Cu]Cu-GluCAB-maleimide-1 in all the organs was
significantly different (p < 0.05 or less) as compared to that of [64Cu]
Cu-TETA and [64Cu]Cu-GluCAB-amine-2 except for the large intestines
indicating similar [64Cu]Cu-GluCAB-amine-2 levels. These results were
in-line with observations from the image-guided biodistribution for
[
64Cu]Cu-GluCAB-maleimide-1, which is seen to remain in circulation
(i.e. bound to plasma proteins, such as HSA) as indicated by a ~ 45 to
55-fold higher presence in plasma (10.7 1.8%ID/g) as compared to
64Cu]Cu-TETA (0.25 0.16%ID/g; p < 0.001) and [64Cu]Cu-GluCAB-
[
amine-2 (0.19%ID/g). Consequently, all other organs also show signifi-
cantly higher activity levels as compared to TETA, in particular the
liver uptake of 9.60
1.36%ID/g for GluCAB-maleimide-1, which was
25-fold higher than TETA (p < 0.001), therefore confirming the
hepatobiliary excretion pathway expected for larger protein based com-
pounds. The liver, being a highly perfused organ, also reflects any re-
maining blood pool 64Cu-activity, which is possible for other organs as
4. Experimental section
well; the %ID/g of the heart (3.15
lung (3.21 1.03) and twice as high to that of the spleen (1.33
2.13) was similar to that of the
4.1. Synthesis and characterisation of GluCAB-amine-2 and GluCAB-
maleimide-1
0.30). Also, notable uptake (with reference to muscle tissue (0.57
0.10%ID/g)), was seen in ovaries and stomach. Interestingly, while the
64Cu-activity in the large intestines was similar between [64Cu]Cu-
GluCAB-maleimide-1 and -amine-2, the [64Cu]Cu-GluCAB-maleimide-
1 levels were 4-fold higher than -amine-2 within the small intestinal
tract (Fig. 5).
The full chemical synthesis of all the compounds described in the
text can be found in the Supporting Information. Described below is
the synthetic method (and characterisation) for compounds (1) and
(2) via the protected intermediate compound (3) and synthesised com-
ponents (4), (5) and (6) (Scheme 1). All commercial chemical reagents
and solvents were purchased from Sigma Aldrich Chemical Co. Ltd. or
Merck (South Africa). Thin layer chromatography was carried out on
Silica-gel 60 F254 plates (Art. 5554; Merck). Column chromatography
was done using Silica-gel 60 from (Merck 7734, 0.040–0.063 mm).
Melting points were determined on a Reichert-Jung Thermovar hot-
3. Conclusion
Herein, the first and second phase of a potentially new radioligand
designed for tumour-targeting and theranostic capabilities has been
62

C.H.S. Driver, T. Ebenhan, Z. Szucs et al.
Nuclear Medicine and Biology 94–95 (2021) 53–66
stage microscope. Infrared spectra were recorded on a Perkin Elmer
Spectrum 100 FT-IR Spectrometer using NaCl disks for oils or KBr/com-
pound discs for solids. High-resolution mass spectra were obtained on a
Agilent 6530 Accurate-Mass Q-TOF LC/MS with electrospray ionization
(ESI) using an Agilent 1290 HPLC fitted with Agilent Eclipse Plus C18
RRHD 1.8 μm 2.1 × 50 mm column. 1H NMR and 13C NMR were recorded
on a Varian Mercury 300 MHz (75.5 MHz for 13C), a Varian Unity
(400 MHz for 13C) or a Bruker Advance III with Ultra Shield 400 Plus
magnet. All spectra were recorded in deuterated chloroform or deuter-
ated water and all chemical shifts were recorded in ppm with reference
to the resonance of the residual solvent used as internal standard. For
radiolabelling, all solvents used were HPLC grade from Merck and all
buffer solutions were prepared using common procedures with Milli-
Q grade water (>18 MΩ/cm). The activity of the [64Cu]CuCl2 solution
was measured using a Capintec CRC-15R gamma detector (Capintec
Inc., Florham Park, NJ, USA). All radiolabelled compounds were analysed
by radio-HPLC analysis on Varian Prostar 325 UV–Vis HPLC apparatus
(Varian Inc., Walnut Creek, CA, USA) fitted with a radiometric GABI
Star gamma detector (raytest GmbH, Straubenhardt, Germany) using a
Zorbax Stable Bond-C18 column (5 μm; 4.6 × 250 mm) with a gradient
elution over 20 min of 95–5% A in B (Solvent A = H2O (0.1% TFA); Sol-
vent B = MeCN (0.1% TFA)). Purification of the radiolabelled products
was done using a pre-conditioned Sep-Pak Light C-18 cartridge (Waters
Corporation, Milford, MA, USA). For in vitro and in vivo studies, samples
were centrifuged using a Hettich EBA 20 (A. Hettich GmbH & Co,
Tuttlingen, Germany) and automated gamma counting was done
using Hidex AMG (LabLogic, Turku, Finland). Serum separation was
done using BD Vacutainer® SST™ serum separator tubes (Becton, Dick-
inson and Company (Franklin Lakes, NJ, USA)).
29.3 (×4), 29.2 (×2), 28.2 (×2)(CH3), 28.1(CH3), 27.2 (x2), 25.8, 25.7,
24.8, 23.7 (CH2Alk/NCH2CH2CH2N)]; νmax/cm−1: 1733 (C_O); HRMS
(ESI MALDI-TOF): m/z Calculated for C81H119N5O16 (M)+; 1417.8652
found – mass too large, could not be determined.
4.3. 1-[10-(β,D-glucopyranos-1-yl)-1-oxodecyl]-8-(10-aminodecyl)-4,11-
diacetic acid-1,4,8,11-tetraaza-cyclotetradecane (2)
Removal of the benzoyl protecting groups from the glycoside of com-
pound (3) was done as follows: Sodium methoxide (25% in MeOH)
(0.5 mL) was added to a solution of (3) (0.100 g, 0.07 mmol) in anh.
MeOH (4 mL) and stirred for 1 h. The reaction was quenched by addition
of Dowex H+ to a pH of 5 and stirred for 15 min. Following work-up and
purification (see supporting information), 1-[10-(β,D-glucopyranos-1-
yl)-1-oxodecyl]-4,11-Bis-(t-butoxycarbonylmethyl)-8-(10-(t-
butoxycarbonylamino)decyl)-1,4,8,11-tetraazacyclotetradecane (16)
was obtained as a clear oil (0.050 g, 71%).
Cyclam (16) (0.050 g, 0.05 mmol) was dissolved in a mixture of
CH2Cl2:TFA (1:1; 4 mL) and stirred for 16 h at RT. The solvent was evap-
orated under a stream of air and the oily residue redissolved in 10%
NH4OH in water (1 mL). The product was purified with column chroma-
tography using Alumina N as the stationary phase. The mobile phase for
elution of the by-products consisted of a gradient elution with DCM:
MeOH:NH4OH (8:1.7:0.3; 7:2.5:0.5; 6.5:3:0.5) while the product was
eluted with MeOH:H2O (6:4; 4:6). The combined product fractions
were concentrated under a stream of air overnight. The residue was
redissolved in water (0.5 mL) and desalted by loading onto a Sep-Pak
Light C-18 cartridge, washing the cartridge with water (2 mL) and
then eluting the product with 50% MeOH in water (2 mL). The solvent
was evaporated off under a stream of argon, the residue redissolved in
water (1 mL) and freeze-dried (Christ Alpha I-5, Type 1050,
Medizinische Apparatebau, Harz, Germany) overnight to yield the title
compound, GluCAB-amine-2 as a lyophilised powder (0.027 g, 69%).
δH (D2O, 300 MHz): 4.34 (1H, d, J = 7.8 Hz, H-1Glu), 3.80 (2H, m,
OCH2, H-6aGlu), 3.63–3.56 (4H, m, OCH2, H-6bGlu, \\NCH2),
3.49–3.25 (6H, m, H-2/3/4/5Glu, \\NCH2), 3.20–2.85(16H, bm,
\\NCH2), 2.70–2.60 (6H, bm,\\NCH2), 1.79 (4H, bm, NCH2CH2CH2N),
1.60–1.50 (8H, bm, CH2),1.29–1.15 (24H, m, CH2-alk)δC(D2O,
100 MHz): 173.8 (C_O), 173.7 (C_O), 102.0 (C-1Glu), 75.8 (C-3Glu),
75.7 (C-2Glu), 73.1 (C-5Glu), 70.6 (OCH2), 69.6 (C-4Glu), 60.7 (C-
6Glu), 54.6 (CH2COO), 54.5 (CH2COO), [54.0, 52.5, 51.6, 51.5, 50.6
(×2), 50.0 (×2), 49.6, 49.4 (C\\N)], 41.4 (CH2NHCO), [30.8, 30.7, 30.6
(×2), 30.4 (×2), 30.3 (×2), 30.0, 29.2, 28.7, 28.0, 27.8, 27.6, 25.8, 25.7,
22.5, 21.3 (CH2Alk/NCH2CH2CH2N)]; νmax/cm−1: 3345 (OH/NH),
1688 (C_O)HRMS (ESI MALDI-TOF): m/z Calculated for
C40H79N5O10 (M + H)+; 790.5860;found 790.5887.
4.2. 1-[10-(2,3,4,6-O-tetrabenzoyl-β,D-glucopyranos-1-yl)-1-oxodecyl]-
4,11-Bis-(t-butoxycarbonylmethyl)-8-(10-(t-butoxycarbonylamino)-
decyl)-1,4,8,11-tetraazacyclotetradecane (3)
Cyclam (5) (0.290 g, 0.49 mmol) was reacted with glycosyl bromide
(4) (0.280 g, 0.34 mmol) and K2CO3 (0.204 g, 1.47 mmol) in MeCN
(20 mL) to yield 1-[10-(β,D-glucopyranos-1-yl)-1-oxodecyl]-4,11-Bis-
(t-butoxycarbonylmethyl)-1,4,8,11-tetraazacyclotetradecane (15) as a
clear oil following purification (0.233 g, 58%)(See supporting informa-
tion for full reaction and characterisation).
Alkyl bromide (6) (0.115 g, 0.34 mmol) was dissolved in MeCN
(1 mL) and added to a solution of cyclam (15) (0.200 g, 0.17 mmol) in
MeCN (10 mL) along with K2CO3 (0.071 g, 0.51 mmol). The solution
was stirred at 80 °C for 16 h followed by filtration through a small celite
pad which was then washed with MeCN (2 mL). The filtrate was con-
centrated down to leave an oily reside which was purified using column
chromatography (CH2Cl2:MeOH:NH4OH, 8.8:1.1:0.1). Some product
remained on the column and was flushed out with CH2Cl2:MeOH:
NH4OH, 7:2.5:0.5. Compound (3) was obtained as a clear oil (0.176 g,
74%); Rf = 0.56 (CH2Cl2:MeOH:NH4OH, 8.8:1.1:0.1); δH (CDCl3,
300 MHz): 8.01–7.81 (8H, m, ArH), 7.55–7.25 (12H, m, ArH), 5.89 (1H,
t, J = 9.6 Hz, H-3Glu), 5.66 (1H, t, J = 9.6 Hz, H-4Glu), 5.50 (1H, dd,
J = 7.8, 9.6 Hz, H-2Glu), 4.83 (1H, d, J = 7.8 Hz, H-1Glu), 4.62 (1H, dd,
J = 3.2, 12.0 Hz, H-6aGlu), 4.50 (2H, dd, J = 5.2, 12.0 Hz, H-6bGlu,
\\NH), 4.14 (1H, m, H-5Glu), 3.89 (1H, dt, J = 6.4, 9.6 Hz, OCH2), 3.53
(1H, dt, J = 6.8, 9.6 Hz, OCH2), 3.38(4H, bs,\\NCH2), 3.19–2.94 (18H,
bm,\\NCH2), 2.78–2.68 (4H, bm,\\NCH2), 1.98 (4H, bm, NCH2CH2CH2-
N), 1.67 (4H, bm, CH2), 1.50 (2H, m, CH2), 1.43 (27H, s, CH3), 1.34–1.01
(26H, m, CH2-alk); δC (CDCl3, 100 MHz): 170.8 (C_O), 170.8 (C_O),
[166.2, 165.9, 165.2, 165.0 (C_O)], 156.5 (C_O), [133.4, 133.2, 133.2,
133.1 (ArC)], [129.9 (×2), 129.8 (×2), 129.6 (×2), 129.0 (×2), 128.6,
128.5 (×2), 128.4 (×2), 128.3, 128.1 (×2) (ArC)], 101.3 (C-1Glu), 81.1
(C-tBu), 81.1 (C-tBu), 80.5 (C-tBu), 73.0 (C-3Glu), 72.2 (C-2Glu), 72.0
(C-5Glu), 70.4 (OCH2), 69.9 (C-4Glu), 63.3 (C-6Glu), 56.1
(CH2COO),56.1 (CH2COO), [53.9, 51.8, 51.3 (×2), 51.2 (×2), 50.7 (×2),
49.3 (×2)(C\\N)],40.9 (CH2NHCO), [29.8, 29.7, 29.4 (×2), 29.4 (×2),
4.4. 4-[10-(4-(2,5-dioxo-2,5-dihydro-1H-pyrrol-1-yl)butanamido)decyl]-
11-[10-(β,D-glucopyranos-1-yl)-1-oxodecyl]-1,4,8,11-
tetraazacyclotetradecane-1,8-diacetic acid (1)
Maleimido-butyric acid NHS-ester (18) (0.005 g, 0.017 mmol) was
dissolved in DMF (0.05 mL) and added to a solution of GluCAB-amine-
2 (0.010 g, 0.011 mmol) in 0.01 M PBS (0.40 mL; pH 7.4). The reaction
was stirred for 1.5 h at room temperature followed by evaporation of
solvent under an air stream. The residue was redissolved in water
(0.20 mL) and loaded onto a Sep-Pak Light C-18 cartridge pre-
conditioned with EtOH (4.0 mL) and water (2.0 mL). The cartridge
was fractionally eluted with water (2 × 2 mL), 10% EtOH/water (2
× 1 mL), 20% EtOH/water (4 × 0.5 mL) and 30% EtOH/water (10
× 0.5 mL). The fractions of 30% EtOH/water were combined and dried
under argon to yield the GluCAB-maleimide-1 (0.006 g, 55%) δH (D2O,
300 MHz): 6.78 (2H, s, CHMal), 4.32 (1H, d, J = 7.8 Hz, H-1Glu), 3.86
(2H, m, OCH2a, H-6aGlu), 3.65–3.56 (4H, m, OCH2b, H-6bGlu,
\\NCH2),3.52 (2H, m, OCCH2CH2CH2NMal), 3.49–3.25 (6H, m, H-2/3/
4/5Glu, \\NCH2), 3.20–2.85(16H, bm, \\NCH2), 2.70–2.60 (6H, bm,
63

C.H.S. Driver, T. Ebenhan, Z. Szucs et al.
Nuclear Medicine and Biology 94–95 (2021) 53–66
\\NCH2), 2.17 (2H, m, OCCH2CH2CH2NMal), 1.87 (2H, m, OCCH2CH2-
CH2NMal), 1.79 (4H, bm, NCH2CH2CH2N), 1.62–1.50 (8H, bm,
CH2),1.29–1.15 (24H, m, CH2-alk);δC (D2O, 100 MHz): 174.8
(C_O),173.8 (C_O), 173.7 (C_O), 172.4 (C_OMal), 134.5 (C_C),
102.2 (C-1Glu), 75.9 (C-3Glu), 75.8 (C-2Glu), 73.3 (C-5Glu), 70.8
(OCH2), 69.9 (C-4Glu), 60.7 (C-6Glu), 54.6 (CH2COO),54.5 (CH2COO),
[54.1, 52.6, 51.6, 51.6, 50.5 (×2), 50.1 (×2), 49.7, 49.4 (C\\N)], 40.4
(CH2NHCO),38.1 (C-4), 34.2 (C-2), [30.8, 30.7, 30.6 (×2), 30.4 (×2),
30.3 (×2), 30.0, 29.2, 28.7, 28.0, 27.8, 27.6, 25.8, 25.7, 25.6, 22.5, 21.3
(CH2Alk/NCH2CH2CH2N)]; HRMS (ESI MALDI-TOF): m/z Calculated for
in 0.01 M PBS (900 μL) for in vivo application. The production of the
maleimide compound included the following steps: After purification,
the dried [64Cu]Cu-GluCAB-amine-2 was redissolved in 0.01 M PBS
(400 μL) and 4-maleimidbutyric acid NHS-ester (9 μL, 100 μg/μL in
DMF, 1.2 eq) was added to the reaction vial. The reaction was stirred
for 1 h at room temperature followed by SPE purification as indicated
and drying under a stream of argon with heating (60 °C). The residue
was re-dissolved in 0.01 M PBS (900 μL) thereby providing [64Cu]Cu-
GluCAB-maleimide-1 (93% conversion, 90% overall yield).
The radiolabelling for the TETA (control compound) was done in a
similar manner: [64Cu]CuCl2 solution (60 μL; 0.72 μmol, 91.5 MBq)
was added to a solution of TETA in 0.1 M NH4OAc buffer (150 μL;
1.61 μmol; ligand:metal 2:1). The pH of the solution was adjusted
with 2 M NaOH (1 μL) to pH 3.5, the vial capped and the solution heated
for 30 min at 90 °C. The product was purified, dried and redissolved in
0.01 M PBS. A labelling efficiency of 95% and radiochemical purity of
99% was obtained.
C
48H86N6O13 (M + H)+; 955.6286; found 955.6291.
4.5. Isotope preparation
All solutions and buffers were prepared using common procedures
using Milli-Q grade water (>18 MΩ/cm). Copper-64 was obtained by ir-
radiation of a CuO (Sigma-Aldrich; 99.9999% purity) powder target
(1 mg) for 36 h in the ‘in-core’ position with a neutron flux of 1
× 1014 n/cm2/s (8 h cooling period) at the SAFARI-1 research reactor
(Necsa, Pelindaba, South Africa). The irradiated copper oxide powder
was dissolved in c. HCl (50 μL) and diluted with water (950 μL) to pro-
duce a solution of [64Cu]CuCl2. Gamma spectrometry (Canberra Germa-
nium Detector – GC2518 (24% deficiency) (Mirion Technologies,
Canberra; Inc. CT, USA)) confirmed presence of the characteristic
511 keV and 1345 keV, emissions of copper-64.
4.10. Stability and protein binding studies
The in vitro stability and serum protein binding studies of [64Cu]Cu-
GluCAB-amine-2 and [64Cu]Cu-GluCAB-maleimide-1 were completed
according to a method previously described [49]. Briefly, the stability
was evaluated by adding the desired compound (50 μL, ~10 MBq) to
serum (1 mL) and incubating the solution at 37 °C. Samples (100 μL)
were drawn at 0, 1, 2, 18 and 24 h after addition of compound and the
proteins precipitated by the addition of cold acetonitrile (500 μL). The
samples were centrifuged for 10 min at 6000 rpm and the supernatant
collected, diluted with MilliQ water and analysed by radio-HPLC (condi-
tions as above). The protein binding was evaluated by addition of the
desired compound (25 μL, ~5 MBq) to serum (450 μL) and incubation
of the solution at 37 °C. Samples (100 μL) were drawn at 0, 1, 2 and 24
h after administration and the proteins precipitated by the addition of
cold acetonitrile (500 μL). The samples were centrifuged for 10 min at
6000 rpm and the supernatant was separated from the protein pellet
which was then washed again with acetonitrile (100 μL). The activity
of the pellet and supernatant (combined with the wash) was then mea-
sured using automated gamma counting and decay-corrected for all
samples. The protein binding is expressed as a percentage of the total ac-
tivity measured.
4.6. Radiolabelling
To establish the optimal conditions for radiolabelling of GluCAB-
amine-2 with 64Cu the radiolabelling performance was compared at
pH 3.5, 5.0 and 9.0, at temperatures of 45, 70, 80 and 90 °C and reaction
times of 10, 20 and 30 min, respectively. Purification of the radiolabelled
products was done using a pre-conditioned (4 mL EtOH followed by
2 mL H2O) Sep-Pak Light C-18 cartridge by loading the reaction solution
on the cartridge, washing with water (2 mL) and eluting the labelled
product with 50% EtOH/H2O (1 mL).
4.7. 64Cu-radiolabelling - optimized method
A [64Cu]CuCl2 solution (100 μL; ~10–12 MBq; molar equivalent li-
gand:metal = 3:1) was added to the GluCAB-amine-2 in 0.1 M
NH4OAc buffer (100 μL; 10 mg/mL; pH 5.5). The pH of the solution
was adjusted with 0.1 M NaOH (60 μL) to pH 3.5, the vial capped and
the solution heated for 30 min at 90 °C (Table S1). After the solution
had cooled it was purified on a SPE cartridge as indicated. The reaction
solution was dried under a stream of argon with heating (60 °C), the res-
idue was re-dissolved in 0.01 M PBS (400 μL) and radio-HPLC analysis
performed.
4.11. Small animal PET/CT imaging
The animal studies were planned and conducted in accordance with
the national regulation for the use of laboratory animals for research
purposes (SANS 10386). The research was approved by the nationally
registered North-West University Animal Care, Health and Safety Re-
search Ethics Committee, (NWU-AnimCare; NWU-00379-16-A5). Fe-
male balb/c mice (age: 6–8 weeks; weight: 26
3 g) obtained from
4.8. Post-labelling conversion of [64Cu]Cu-GluCAB-amine-2 to [64Cu]Cu-
GluCAB maleimide-1
the North West University, Preclinical Drug Development Platform
(PCDDP) Vivarium (Potchefstroom, South Africa) were housed in indi-
vidually ventilated cages (Techniplast IVC, Buguggiate, Italy) (ad libitum
access to water and food, 12 h light/dark cycles and sterile corn-cob bed-
ding). The animals were acclimatised at the Pre-Clinical Imaging Facility
(Pelindaba, South Africa) for 1 week prior to commencement of the
study. Imaging studies were performed using the small animal PET/CT
camera (nanoScanPET/CT, Mediso Medical Imaging Systems, Budapest,
Hungary). Mice were randomised and allocated to one of three groups
(n = 5) as follows: A) [64Cu]Cu-TETA, B) [64Cu]Cu-GluCAB-amine-2 or
C) [64Cu]Cu-GluCAB-maleimide-1. The compounds, as indicated per
group, were administered intravenously via the lateral tail vein ([64Cu]
Cu-TETA: 100 μL; 1.2 0.22 MBq; 39 nmol/mouse, [64Cu]Cu-GluCAB-
amine-2: 150 μL; 5.6 MBq, 31 nmol/mouse and [64Cu]Cu-GluCAB-
4-maleimidobutyric acid NHS-ester (4.5 μL, 100 μg/μL in DMF, 1.3 eq)
was added to the [64Cu]Cu-GluCAB-amine-2 reaction vial (400 μL,
0.01 M PBS) and stirred for 1 h at room temperature. The reaction solu-
tion (400 μL) was purified on a SPE cartridge as described above, dried
under argon with slight heating (60 °C) and redissolved in 0.01 M PBS.
4.9. 64Cu-radiolabelling for in vivo studies
A [64Cu]CuCl2 solution (60 μL; 0.72 μmol, 91.5 MBq) was added to a
solution of GluCAB-amine-2 in 0.1 M NH4OAc buffer (177 μL;
2.24 μmol;10 mg/mL). The pH of the solution was adjusted to pH 3.5
using 2 M NaOH (4 μL), the vial capped and the solution heated for
30 min at 90 °C. Purification was done as mentioned above. The dried
maleimide-1: 100 μL; 1.0
0.09 MBq, 31 nmol/mouse). The animals
were anaesthetised using a mix of isoflurane in oxygen (3–4% for induc-
tion and 2–2.5% for maintenance). MicroPET/CT image acquisition (CT
[
64Cu]Cu-GluCAB-amine-2 (labelling efficiency of 97%) was redissolved
64

C.H.S. Driver, T. Ebenhan, Z. Szucs et al.
Nuclear Medicine and Biology 94–95 (2021) 53–66
scan: time – 5 min, method – semi-circular, zoom – max FOV, binning –
1:4 and image reconstruction: voxel size –medium, filter type - Cosine;
PET scan: time – 20 min, coincidence mode – 1-5, prone-head first posi-
tion) were performed at 1, 2 and 24 h after injection for group A and B
and at 1, 2, 4, 8 and 24 h after injection for Group C using the Nucline
software (Mediso Ltd., Budapest, Hungary). Qualitative and quantitative
analysis was performed on decay-, scatter-, randoms- and attenuation-
corrected images using InterView Fusion software (Mediso Ltd., Buda-
pest, Hungary). PET images were normalised (low 0.0; high 10.0) to
allow for comparison between different time points and between the
three compounds. Image-guided tissue quantification, i.e. activity con-
centration in blood pool (heart) as well as in other organs was done
by creating tissue-specific, CT-guided volume-of-interest (‘VOI’s) areas
and each VOI was used to determine the mean standardized uptake
value (SUV) (g/mL).
Appendix A. Supplementary data
Details provided include a complete synthesis and characterisation
of all compounds indicated in the main manuscript as well as radio-
HPLC chromatograms for serum stability analysis of the 64Cu-labelled
compounds. Supplementary data to this article can be found online at
https://doi.org/10.1016/j.nucmedbio.2021.01.001.
References
[1] Ting G, Chang C-H, Wang H-E, Lee T-W. Nanotargeted radionuclides for cancer nu-
clear imaging and internal radiotherapy. J Biomed Biotechnol. 2010:953537.
[2] Van Zutphen S, Reedijk J. Targeting platinum anti-tumour drugs: overview of strat-
egies employed to reduce systemic toxicity. Coord Chem Rev. 2005;249:2845–53.
[3] Giancotti FC. Deregulation of cell signaling in cancer. FEBS Lett. 2014;588:2558–70.
[4] Acharya S, Sahoo SK. PLGA nanoparticles containing various anticancer agents and
tumour delivery by EPR effect. Adv Drug Deliv Rev. 2011;63:170–83.
[5] a) Maeda H, Bharate GY, Daruwalla J. Polymeric drugs for efficient tumor-targeted
drug delivery based on EPR-effect. Eur J Pharm Biopharm. 2009;71:409–19.
b) Fang J, Nakamura H, Maeda H. The EPR effect: unique features of tumor blood ves-
sels for drug delivery, factors involved, and limitations and augmentation of the
effect. Adv Drug Deliv Rev. 2009;63:136–51.
4.12. Biodistribution
Following the imaging studies, the biodistribution of the
64Cu-compounds at 24 h p.i was determined as follows: anaesthetised
animals were euthanised by decapitation and blood was immediately
collected into a pre-weighed SST™ tubes; serum was yielded by centri-
fugation (5 min; 6000 rpm). Other tissues and organs were collected
into pre-weighed 6 mL plastic vials and the radioactivity of all the sam-
ples was measured using automated gamma counting. A standard dilu-
tion of the injected activity was performed measuring three volume
fractions (100, 200 and 500 μL) to determine the counts/MBq. This
value was then used to express the organ activity as a decay-corrected,
percentage of the injected dose per gram of tissue (%ID/g).
[6] a) Matsumura Y, Maeda H. A new concept for macromolecular therapeutics in can-
cer chemotherapy: mechanism of tumoritropic accumulation of proteins and the
antitumor agent smancs. Cancer Res. 1986;46:6387–92.
b) Maeda H, Matsumura Y. Tumoritropic and lymphotropic principles of macromo-
lecular drugs. Crit Rev Ther Drug Carrier Syst. 1986;6:193–210.
c) Maeda H. SMANCS and polymer-conjugated macromolecular drugs: advantages
in cancer chemotherapy. Adv Drug Deliv Rev. 1986;46:169–85.
[7] a) Rosenblum D, Joshi N, Tao W, Karp JM, Peer D. Progress and challenges towards
targeted delivery of cancer therapeutics. Nat Commun. 2018;9:1410.
b) Kydd J, Jadia R, Velpurisiva P, Gad A, Paliwal S, Rai P. Targeting strategies for the
combination treatment of cancer using drug delivery systems. Pharmaceutics.
2018;9:46.
[8] a) Bertrand N, Wu J, Xu X, Kamaly N, Farokhzad OC. Cancer nanotechnology: the
impact of passive and active targeting in the era of modern cancer biology. Adv
Drug Deliv Rev. 2014;66:2–25.
b) Cho K, Wang X, Nie S, Chen Z, Shin D. Therapeutic nanoparticles for drug delivery
in cancer. Clin Cancer Res. 2014;14:1310–13016.
[9] Ramogida CF, Orvig C. Tumour targeting with radiometals for diagnosis and therapy.
Chem Commun. 2013;49:4720–39.
[10] Liu S. Bifunctional coupling agents for radiolabelling of biomolecules and target-
specific delivery of metallic radionuclides. Adv Drug Deliv Rev. 2008;60:1347–70.
[11] Gutfelin B, Souza SAL, Valentini G. Copper-64: a real theranostic agent. Des Devel
Ther. 2018;12:3235–45.
4.13. Statistics
The protein binding studies were done in triplicate and the data re-
ported as the mean
standard deviation (SD). The standard uptake
values (SUV) determined from microPET/CT analysis and the
biodistribution data determined from gamma counting are presented
as the mean values and standard deviation (SD). The significance be-
tween two mean values of interest was determined by using the two-
tailed, equal variance Student's t-test. The levels of significance (p) for
validating the alternative hypothesis are established at p < 0.05, <0.01
or <0.001.
[12] Sun X, Anderson CJ. Production and applications of Copper-64 radiopharmaceuticals.
Methods Enzymol. 2004;386:237–61.
[13] Jalilian AR, Osso J. The current status and future of theranostic Copper-64 radiophar-
maceuticals. Iran J Nucl Med. 2017;25:1–10.
[14] Paurova M, David T, Cisarova I, Lubal P, Hermann P, Kotek J. Optimization of the se-
lectivity and rate of copper radioisotope complexation: formation and dissociation
kinetic studies of 1,4,8-trimethylcyclam-based ligands with different coordinating
pendant arms. New J Chem. 2018;42:11908–29.
[15] Wadas TJ, Wong EH, Weisman GR, Anderson CJ. Copper chelation chemistry and its
role in copper radiopharmaceuticals. Cur Pharm Design. 2007;13:3–16.
[16] Szymanski P, Fraczek T, Markowicz M, Mikiciuk-Olasik E. Development of copper
based drugs, radiopharmaceuticals and medical materials. Biometals. 2012;25:
1089–112.
[17] Liu Z, Chen X. Simple bioconjugate chemistry serves great clinical advances: albumin
as a versatile platform for diagnosis and precision therapy. X. Chen Chem Soc Rev.
2016;45:1432–56.
[18] Elzoghby AO, Samy WM, Elgindy NA. Albumin-based nanoparticles as potential con-
trolled release drug delivery systems. J Control Release. 2012;157:168–82.
[19] a) Kratz F. Albumin as a drug carrier: design of prodrugs, drug conjugates and nano-
particles. J Control Release. 2008;132:171–83.
Declaration of competing interest
The authors declare no competing financial interest.
Acknowledgements
The authors acknowledge financial support by the South African Nu-
clear Energy Corporation (Necsa). Further financial support and infra-
structure is acknowledged for pre-clinical studies from the Nuclear
Medicine Research Infrastructure, South Africa (NuMeRI) NPC as funded
by the Department of Science and Innovation (DSI), South Africa. M.I.P.
was jointly supported by the SAMRC with funds received from the Na-
tional Department of Health, South Africa and the MRC UK with funds
from the UK Government's Newton Fund and GSK. The authors would
like to thank: Department of Chemistry (University of Cape Town) tech-
nical staff for assistance with NMR analysis and the Hunter Research
Group for organic synthesis support; Ms M. Buys and Ms B.
Marjanovic-Painter (Necsa) for assistance with isotope gamma analysis
and HPLC/MS analysis; The staff at the Pre-Clinical Drug Development
Platform (PCDDP), NWU, Potchefstroom for assistance with in vivo stud-
ies and associated veterinary services; Ms. D van Wyk (NuMeRI) and Dr.
J. Kleynhans (NuMeRI) for assistance during microPET/CT imaging.
b) Kratz F, Ehling G, Kauffmann HM, Unger C. Acute and repeat-dose toxicity studies
of the (6-maleimidocaproyl)hydrazone derivative of doxorubicin (DOXO-EMCH),
an albumin-binding prodrug of the anticancer agent doxorubicin. Hum Exp
Toxicol. 2008;26:19–35.
[20] Miele E, Spinelli GP, Miele E, Tomao F. Albumin-bound formulation of paclitaxel
(Abraxane® ABI-007) in the treatment of breast cancer. Int J Nanomedicine. 2009;
4:99–105.
[21] Jacobson O, Kiesewetter DO, Chen X. Albumin-binding Evans blue derivatives for di-
agnostic imaging and production of long-acting therapeutics. Bioconjug Chem. 2016;
27:2239–47.
[22] Lau J, Jacobson O, Niu G, Lin K-S, Bénard F, Chen X. Bench to bedside: albumin
binders for improved cancer radioligand therapies. Bioconjug Chem. 2019;30:
487–502.
[23] Müller C, Farkas R, Borgna F, Schmid RM, Benešová M, Schibli R. Synthesis,
radiolabeling, and characterization of plasma protein-binding ligands: potential
tools for modulation of the pharmacokinetic properties of (radio)pharmaceuticals.
Bioconjug Chem. 2017;28:2372–83.
[24] Upadhyay M, Samal J, Kandpal M, Singh OV, Vivekanakdan P. The Warburg effect: in-
sights from the past decade. Pharmacol Ther. 2013;137:318–30.
65

C.H.S. Driver, T. Ebenhan, Z. Szucs et al.
Nuclear Medicine and Biology 94–95 (2021) 53–66
[25] Wahl RL. Targeting glucose transporters for tumor imaging: “sweet” idea, “sour” re-
sult. J Nucl Med. 1996;37:1038–41.
[26] Smith TA. FDG uptake, tumour characteristics and response to therapy: a review.
[38] Siwowska K, Haller S, Bortoli F, Benesova M, Groehn V, Bernhardt P, et al. Preclinical
comparison of albumin-binding Radiofolates: impact of linker entities on the in vitro
and in vivo properties. Mol Pharm. 2017;14:523–32.
Nucl Med Comm. 1998;19:97–105.
[27] Morais GR, Falconer RA, Santos I. Carbohydrate-based molecules for molecular imag-
ing in nuclear medicine. Eur J Org Chem. 2013;2013:1401–14.
[28] Feng H, Wang X, Chen J, Cui J, Gao T, Gao Y, et al. Nuclear imaging of glucose metab-
olism: beyond 18F-FDG. Contrast Media Mol Imaging. 2019;2019:7954854.
[29] Zhang M, Zhang Z, Blessington D, Li H, Busch TM, Madrak V, et al. Pyrophoephorbide
[39] a) Pandya DN, Kim JY, Park JC, Lee H, Phapale PB, Kwak W, et al. Revival of TE2A: a
better chelator for Cu(II) ions than TETA? Chem Commun. 2010;46:3517–9.
b) Pandya DN, Dale AV, Kim JY, Lee H, Ha YS, An GI, et al. New macrobicyclic chelator
for the development of ultrastable 64Cu-radiolabeled bioconjugate. Bioconjug
Chem. 2010;23:330–5.
[40] Pandya DN, Bhatt N, Dale AV, Kim JY, Lee H, Ha YS, et al. New bifunctional chelator
for 64Cu-immuno-positron emission tomography. Bioconjug Chem. 2013;24:
1356–66.
2-deoxyglucosamide:
a new photosensitizer targeting glucose transporters.
Bioconjug Chem. 2003;14:709–14.
[30] Schibli R, Dumas C, Petrig J, Spadola L, Scapozza L, Garcia-Garayoa E, et al. Synthesis
and in vitro characterization of organometallic rhenium and technetium glucose
complexes against glut 1 and hexokinase. Bioconjug Chem. 2005;16:105–12.
[31] Weber J, Bollepalli L, Belenguer AM, Di Antonio M, De Mitri N, Joseph J, et al. An
activatable cancer-targeted hydrogen peroxide probe for photoacoustic and fluores-
cence imaging. Cancer Res. August 27 2019. https://doi.org/10.1158/0008-5472.
CAN-19-0691.
[41] Vimalnath KV, Rajeswari A, Chirayil V, Sharad PL, Jagadeesan KC, Joshi PV, et al. Stud-
ies on preparation of 64Cu using (n,γ) route of reactor production using medium flux
research reactor in India. J Radioanal Nucl Chem. 2011;290:221–5.
[42] International Atomic Energy Agency. Cyclotron produced radionuclides: emerging
positron emitters for medical applications: 64Cu and 124I. IAEA radioisotopes and ra-
diopharmaceuticals reports no. 1. Vienna: IAEA; 2016. p. 3–7.
[43] Xie Q, Zhu H, Wang F, Meng X, Ren Q, Xia C, et al. Establishing reliable Cu-64 produc-
tion process: from target plating to molecular specific tumour micro-PET imaging.
Molecules. 2017;22:641.
[32] a) Liang X, Sadler PJ. Cyclam complexes and their applications in medicine. Chem
Soc Rev. 2004;33:246–66.
b) Ke CY, Mathias CJ, Green MA. Folate-receptor-targeted radionuclide imaging
[44] Silversides JD, Allan CC, Archibald SJ. Copper(II) cyclam-based complexes for radio-
pharmaceutical applications: synthesis and structural analysis. Dalton Trans. 2007;
9:971–8.
[45] Zarschler K, Kubeil M, Stephan H. Establishment of two complementary in vitro as-
says for radiocopper complexes achieving reliable and comparable evaluation of
in vivo stability. RSC Adv. 2014;4:10157–64.
agents. Adv Drug Deliv Rev. 2004;56:1143–60.
[33] Hunter TM, Paisey SJ, Park H-S, Cleghorn L, Parkin A, Parsons S, et al. Configurations
of metallocyclams and relevance to anti-HIV activity. J Inorg Biochem. 2004;98:
713–9.
[34] Hoffmann TJ, Smith CJ. True radiotracers: Cu-64 targeting vectors based upon
bombesin peptide. Nucl Med Biol. 2009;36:579–85.
[46] Jones-Wilson TM, Deal KA, Anderson CJ, McCarthy DW, Kovacs Z, Motekaitis RJ, et al.
The in vivo behaviour of copper-64-labeled azamacrocyclic complexes. Nucl Med
Biol. 1998;25:523–30.
[35] Galibert M, Jin Z-H, Furukawa T, Saga T, Fujibayashi Y, Dumy P, et al. RGD-cyclam
conjugates: synthesis and potential application for positron emission tomography.
Bioorg Med Chem Lett. 2010;20:5422–5.
[47] Bardal SK, Waechter JE, Martin DS. Applied Pharmacology; Chapter 2: Pharmacoki-
[36] Murakami T, Sato Y, Shibakami M. Stereoselective glycosylations using benzoylated
glucosyl halides with inexpensive promoters. Carbohydr Res. 2008;343:1297–308.
[37] a) Thiem J, Meyer B. Synthesen mit Iod- und Bromtrimethylsilan in der
Saccharidchemie. Chem Ber. 1980;113:3075–85.
netics. St Louis, MO: Elsevier/Saunders; 2011; 17–34.
[48] Levitt DG, Levitt MD. Human serum albumin homeostasis: a new look at the roles of
synthesis, catabolism, renal and gastrointestinal excretion, and the clinical value of
serum albumin measurements. Int J Gen Med. 2016;9:229–55.
b) Gervay J, Nguyen TN, Hadd MJ. Mechanistic studies on the stereoselective forma-
tion of glycosyl iodides: first characterization of β-D-glycosyl iodides. Carbohydr
Res. 1980;300:119–25.
[49] Beaino W, Anderson CJ. PET imaging of very late antigen-4 in melanoma: compari-
son of 68Ga- and 64Cu-labeled NODAGA and CB-TE1A1P-LLP2A conjugates. J Nucl
Med. 2014;55:1856–63.
66