Efficient bioconjugation of 5-fluoro-5-deoxy-ribose (FDR) to RGD peptides for positron emission tomography (PET) imaging of αvβ 3 integrin receptor

Article abstract of DOI:10.1039/c3ob40550h

The utility of 5-fluoro-5-deoxyribose (FDR) as an efficient bioconjugation agent for radiolabelling of the RGD peptides c(RGDfK) and c(RGDfC) is demonstrated. The bioconjugation is significantly superior to that achieved with 2-fluoro-2-deoxyglucose (FDG) and benefits from the location of the fluorine at C-5, and that ribose is a 5-membered ring sugar rather than a 6-membered ring. Both features favour ring opening to the aldehydic form of the sugar to promote smooth oxime ligation with aminooxy ether functionalised peptides. [ ¹⁸F]FDR was prepared in this study by synthesis from fluoride-18 using an automated synthesis protocol adapting that used routinely for [ ¹⁸F]FDG. c(RGDfK) was functionalised with an aminooxyacetyl group (Aoa) via its lysine terminus, while c(RGDfC) was functionalised with an aminooxyhexylmaleimide (Ahm) through a cysteine-maleimide conjugation. Bioconjugation of [¹⁸F]FDR to c(RGDfC)-Ahm proved to be more efficient than c(RGDfK)-Aoa (92% versus 65%). The unlabelled (¹⁹F) bioconjugates c(RGDfK)-Aoa-FDR and c(RGDfC)-Ahm-FDR were prepared and their in vitro affinity to purified integrin α_vβ₃ was determined. c(RGDfK)-Aoa-FDR showed the greater affinity. Purified hot bioconjugates c(RGDfK)-Aoa-[¹⁸F]FDR and c(RGDfC)-Ahm-[¹⁸F]FDR were assayed by incubation with MCF7, LNCaP and PC3 cell lines. In both cases the conjugated RGD peptides showed selectivity for PC3 cells, which express α_vβ₃ integrin, with the c(RGDfK)-Aoa-[¹⁸F]FDR demonstrating better binding, consistent with its higher in vitro affinity. The study demonstrates that [ ¹⁸F]FDR is an efficient bioconjugation ligand for RGD bioactive peptides. The Royal Society of Chemistry 2013.

Full text of DOI:10.1039/c3ob40550h

Organic &
Biomolecular Chemistry
View Article Online
View Journal | View Issue
PAPER
Efficient bioconjugation of 5-fluoro-5-deoxy-ribose
(FDR) to RGD peptides for positron emission
Cite this: Org. Biomol. Chem., 2013, 11,
4551
tomography (PET) imaging of αvβ3 integrin receptor†
Sergio Dall’Angelo,a Qingzhi Zhang,b Ian N. Fleming,c Monica Piras,a
Lutz F. Schweiger,a David O’Hagan*b and Matteo Zanda*a
The utility of 5-fluoro-5-deoxyribose (FDR) as an efficient bioconjugation agent for radiolabelling of the
RGD peptides c(RGDfK) and c(RGDfC) is demonstrated. The bioconjugation is significantly superior to that
achieved with 2-fluoro-2-deoxyglucose (FDG) and benefits from the location of the fluorine at C-5, and
that ribose is a 5-membered ring sugar rather than a 6-membered ring. Both features favour ring opening
to the aldehydic form of the sugar to promote smooth oxime ligation with aminooxy ether functionalised
peptides. [18F]FDR was prepared in this study by synthesis from fluoride-18 using an automated synthesis
protocol adapting that used routinely for [18F]FDG. c(RGDfK) was functionalised with an aminooxyacetyl
group (Aoa) via its lysine terminus, while c(RGDfC) was functionalised with an aminooxyhexylmaleimide
(Ahm) through a cysteine–maleimide conjugation. Bioconjugation of [18F]FDR to c(RGDfC)-Ahm proved to
be more efficient than c(RGDfK)-Aoa (92% versus 65%). The unlabelled (19F) bioconjugates c(RGDfK)-Aoa-
FDR and c(RGDfC)-Ahm-FDR were prepared and their in vitro affinity to purified integrin αvβ3 was deter-
mined. c(RGDfK)-Aoa-FDR showed the greater affinity. Purified “hot” bioconjugates c(RGDfK)-Aoa-[18F]FDR
and c(RGDfC)-Ahm-[18F]FDR were assayed by incubation with MCF7, LNCaP and PC3 cell lines. In both
cases the conjugated RGD peptides showed selectivity for PC3 cells, which express αvβ3 integrin, with the
c(RGDfK)-Aoa-[18F]FDR demonstrating better binding, consistent with its higher in vitro affinity. The study
demonstrates that [18F]FDR is an efficient bioconjugation ligand for RGD bioactive peptides.
Received 19th March 2013,
Accepted 15th May 2013
DOI: 10.1039/c3ob40550h
www.rsc.org/obc
peptides, proteins and other bio-macromolecules by 18F–C
bond-forming reactions is generally difficult, whereas intro-
Introduction
There is a significant research focus presently aimed at duction of 18F can be more conveniently achieved by the use of
tagging bioactive peptides with the fluorine-18 (18F) isotope, to 18F-labelled “prosthetic groups” which can be linked to the
monitor in vivo distribution by positron emission tomography peptide/protein by “click”-chemistry, and other “click-type”
(PET).1 This imaging modality has the advantage of requiring reactions (oxime-bond formation, thiol-maleimide chemistry,
low dose administration due to the sensitivity of detection. etc.).3 2-[18F]Fluoro-2-deoxyglucose ([18F]FDG) is the most com-
The 18F isotope has a relative long half-life (t1/2 = 110 min) in monly used radio tracer synthesised routinely in the clinical
the context of appropriate PET active nuclides (11C = 20.4 min, environment as it is an excellent diagnostic in oncology and
13N = 10 min, 15O = 2.1 min), but the relatively short half-life other pathologies.4 [18F]FDG can be used for 18F-fluoroglucosy-
demands that the synthesis of [18F]-ligands is rapid. This lation of hydroxylamine-functionalised peptides via oxime-
restricts the diversity of candidate molecules for [18F]-labelling bond formation.5 However, in practice [18F]FDG is a rather
of candidate PET tracers.2 Direct radio-fluorination of poor prosthetic group for this purpose, because the position of
the electronegative fluorine atom at C-2 inhibits ring opening
of the sugar to its aldehydic form, necessary for oxime or
aJohn Mallard Scottish PET Centre, School of Medical Sciences, University of
imine formation. To circumvent this problem other strategies
Aberdeen, Foresterhill, Aberdeen, AB25 2ZD, UK. E-mail: m.zanda@abdn.ac.uk;
have emerged for tagging the [18F]-isotope to peptides. The
Tel: (+44) 0 1224 437153
bUniversity of St Andrews, School of Chemistry and Centre for Biomolecular Sciences,
North Haugh, St Andrews, Fife, KY16 9ST, UK. E-mail: do1@st-andrews.ac.uk;
Fax: +44(0)1334 463800; Tel: +44(0)1334 467171
most common strategy has involved the preparation of
[18F]-fluorobenzoate derivatives6 or [18F]-fluorobenzaldehyde-
derived oximes.7 These motifs have emerged as there are suit-
able synthetic methods for the rapid introduction of 18F into
the para position of the aryl moiety. However these motifs
suffer from higher hydrophobicity which can alter the overall
cAberdeen Biomedical Imaging Centre, Lilian Sutton Building, Foresterhill, Aberdeen,
AB25 2ZD, UK
†Electronic supplementary information (ESI) available: Experimental methods
and crystallographic data. See DOI: 10.1039/c3ob40550h
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem., 2013, 11, 4551–4558 | 4551

View Article Online
Paper
Organic & Biomolecular Chemistry
physicochemical properties of the peptide. More recently
ligands such as the NOTA-aluminium ligand sequester 18F
directly avoiding covalent C–18F bond formation although
clearly the NOTA ligand remains attached to the peptide.8 The
challenge is to tag the protein without significantly modifying
its bioactivity. We have recently introduced [18F]FDR as a bio-
conjugation ligand, and explored ligation to linear peptides.9
Angiogenesis, the growth of new blood vessels from pre-
existing vessels, is one of the hallmarks of growing tumours.
The integrin protein αvβ3 (also known as the vitronectin recep-
tor) plays an important role in regulating angiogenesis and
tumour metastasis. Indeed, there is good correlation between
αvβ3 expression on tumour endothelial cells and the extent of
angiogenesis.10 The Arg-Gly-Asp (RGD) tripeptide motif binds
to the αvβ3 receptor, therefore the RGD sequence has been
extensively used as a targeting function for αvβ3 integrin over-
expressing tumours.11 Several RGD peptides have been identi-
fied that bind to αvβ3 with high affinity and specificity, includ-
ing c(RGDfK) and c(RGDfC).12 RGD-peptides radio-labelled
with 68Ga,13 64Cu,14 and 18F nuclides have been investigated as
PET tracers in pre-clinical and clinical studies, mainly for
oncological, bone and cardiovascular pathologies.15 Several
strategies have been pursued for the incorporation of 18F into
RGD peptides. The most important examples of [18F]RGD PET
tracers include: (1) [18F]-galacto-RGD, which is a c(RGDfK) gly-
copeptide bearing a [18F]-2-fluoropropionate tag;16 (2) [18F]-
fluciclatide (or 18F-AH111585) which is a rigid bicyclic RGD oli-
gopeptide incorporating a PEG linker radiolabelled with a
para-[18F]-benzaldehyde-derived oxime;17 (3) cyclic and linear
RGD peptides bearing an oxime-bound [18F]FDG tag;18 (4)
cyclic RGD peptides displaying [18F]-silylated prosthetic
groups; (5) 18F-(RGD)2 peptides which are a class of dimeric
cyclic RGD peptide tracers generally labelled with para-[18F]-
benzoate tags;19 (6) [18F]RGD-K5 and related RGD tracers
bearing a “click-type” [18F]-fluoroalkyl-triazole tag.20 Interest-
ingly, a direct comparison between “click”-radiofluorination
strategy and oxime- or amide-bond forming 18F-prosthetic
groups was performed, showing that oxime-bond formation is
at least as convenient as “click”-chemistry for the radiofluori-
nation of RGD peptides.21
Scheme 1 Functionalisation of cRGDfC via aminooxy conjugation with various
sugars: (i) H2O, rt 20 min; (ii) NH4OAc or NaOAc buffer (250 mM, pH 4.6).
Scheme 2 Preparation of c(RGDfK)-Aoa and conjugation with FDR: (i) Boc-
amino-oxy-acetic acid (2 eq.), HATU (2 eq.), sym-collidine (4 eq.), rt, 2 h; (ii) TFA–
CH2Cl2 (1 : 1), rt, 2 h; (iii) FDR (1.3 eq.) in sodium acetate buffer (0.5 M, pH 4.6),
rt, 10 min.
peptides to FDR, they were each functionalised with an amino-
oxy functionality via a suitable linkage.9 For c(RGDfC) a con-
ventional Michael addition of the cysteine residue to an
aminooxyhexylmaleimide (Ahm)22 was employed (Scheme 1).
The c(RGDfK) peptide was prepared by solid phase synthesis
and functionalised by acylation of its lysine residue with Boc-
protected aminooxyacetic acid (Aoa) before removal of protect-
ing groups and cleavage from the resin under acid conditions
(Scheme 2).23 Conjugation of c(RGDfC) and c(RGDfK) with
FDR was then carried out in NaOAc or NH4OAc buffer
(100–250 mM, pH 4.6), using the previously described opti-
mised conditions (Schemes 1 and 2).9
A series of experiments was carried out with c(RGDfC),
comparing the conjugation ability of several sugars, and in
particular comparing 5- and 6-membered ring sugars. Sugars
exist in solution as an equilibrium mixture of both their cyclic
and acyclic aldehydic forms. The less favoured aldehydic form
is required for reaction with the aminooxy group to form the
conjugated oxime. It is emerging as a feature of FDR that it
conjugates rapidly, and in this study conjugations were
explored with c(RGDfC)-Ahm in both NH4OAc and NaOAc
(250 mM) solutions at room temperature. Fig. 1 illustrates the
The oxime-bioconjugation of [18F]FDR to two cyclic RGD
peptides is reported in this paper. An enzyme-linked immuno-
sorbent assay (ELISA) was used to investigate whether FDR
conjugation to these RGD peptides affected their ability to
bind to immobilised αvβ3 protein. The ability of [18F]FDR
labelled RGD peptides to specifically bind its target was
assessed in cells which express different levels of αvβ3 protein,
in the presence and absence of an excess of non-radiolabelled
RGD peptide.
Results and discussion
Synthesis of RGD-FDG conjugates
Two cyclic RGD peptides, c(RGDfC) and c(RGDfK) were chosen
for αvβ3 integrin recognition. In order to conjugate these
4552 | Org. Biomol. Chem., 2013, 11, 4551–4558
This journal is © The Royal Society of Chemistry 2013

View Article Online
Organic & Biomolecular Chemistry
Paper
Fig. 2 Analytical radio-RP-HPLC traces of c(RDGfK)-Aoa-[18F]FDR (blue trace)
and c(RDGfC)-Ahm-[18F]FDR (red trace) products.
an Oasis HLB cartridge. The solution was reanalysed by
analytical radio-HPLC, which demonstrated that each product
was radiochemically pure as illustrated in Fig. 2. The conju-
gation and purification of the [18F]FDR-RGD adducts took
∼50 min.
Fig. 1 Comparison of the reaction conversions (HPLC) of cRGDfC-Ahm with
various sugars (1 : 1 molar ratio) in NH4OAc buffer (200 mM, pH 4.6) at 25 °C.
relative conversions of oxime product formation of c(RGDfC)-
Ahm with FDR, D-ribose, 2-FDG, 6-FDG and D-glucose. All
sugars were incubated in 1 : 1 molar ratio with c(RGDfC)-Ahm
and each at a final concentration of 22 mM. NH4OAc solution
was used as it offers a suitable buffer solution for direct MS
analysis. FDR was the most efficient sugar explored with conju-
gation conversions of up to 84% in 10 min. This is followed by
D-ribose (51% in 10 min), a comparison which illustrates the
influence of the fluorine in place of the OH at C-5. All of the
pyranoses (6-membered sugars) reacted considerably slower
than the ribose sugars, in the order 6-FDG > D-glucose > 2-FDG
consistent with the thermodynamically more stable 6-mem-
bered rings.9 A positive fluorine effect is observed comparing
6-FDG to D-glucose, while the fluorine in 2-FDG will disfavour
oxocarbenium ion formation and inhibits ring opening to the
aldehydic form. Conjugation with 2-FDG is very sluggish under
the conditions generating negligible conjugation product
(ca. 2–5%) over a 1 h reaction. The reactions of FDR benefits
from an increased temperature (40 °C, 1 : 1 molar ratio, 90%
conversion, 10 min) and an excess of FDR (1.5 equiv., 25 °C),
resulted in a conversion up to 93% in 10 min. It is clear from
the histogram (Fig. 1) that FDR is the most efficient bioconju-
gation reagent of those explored.
αvβ3 Integrin affinity evaluation
The cyclic RGD peptide conjugates were submitted to assays to
measure their affinity for the αvβ3 receptor. Competitive
affinities were calculated analyzing the interactions between
immobilized αvβ3 and the commercially available c[RGDfK-
(Biotin-PEG-PEG)] peptide in the presence of the RGD conju-
gates. The linear heptapeptide GRGDSPK (Sigma Chem Co.
Ltd) was used as a reference competitive peptide and IC50
values for each compound were normalised with respect to
this peptide IC50 to allow comparisons between different
experiments. Competitive inhibition data in Table 1 show that
all of the conjugated c(RGD) peptides have affinities for αvβ3 in
the nanomolar range. The highest affinity (68 nM) was dis-
played by c(RGDfK)-Aoa-FDR, slightly higher than that of the
non-conjugated peptide c(RGDfK). Overall, the αvβ3 binding
affinity data in Table 1 indicate that FDR is an efficient pros-
thetic group that does not affect the αvβ3 targeting capacity of
RGD peptides. The relatively low affinity of c(RGDfC)-Ahm-FDR
compared to c(RGDfK)-Aoa-FDR might be the result of
increased hydrophobicity introduced by the relatively longer
aliphatic linkage.
Isotopically labelled (18F) conjugations of the RGD peptides
were carried out with an excess of both c(RGDfC)-Ahm and
c(RGDfK)-Aoa (ca. 2 mg) using [18F]FDR (15–75 MBq) at pH
4.6 (NaOAc buffer) at room temperature. c(RGDfC)-Ahm (92%
conversion, n = 2) reacted more rapidly than c(RGDfK)-Aoa
(65% conversion, n = 2) at 15 min under the same conditions.
This is consistent with the more nucleophilic aminooxy group
of c(RGDfC)-Ahm relative to c(RGDfK)-Aoa, whose aminooxy
group bears an electron-withdrawing carbonyl in β-position.
The reaction mixtures were purified by semi-prep RP-HPLC
and the radioactive fractions were collected by eluting the
radioactivity with a 1 : 1 water–ethanol solution (1.5 mL) from
Table 1 Inhibition of cyclo[RGDfK(Biotin-PEG-PEG)] binding to immobilised
αvβ3. Data are the average of two independent experiments performed in tripli-
cate. Q = normalized activities as ratio IC50(peptide)/IC50(GRGDSPK)
Compound
IC50 (μM)
Q
Ref peptide (GRGDSPK)
c(RGDfC)
c(RGDfC)-Ahm
c(RGDfC)-Ahm-FDR
c(RGDfK)
2.27
1
0.085
0.222
0.403
0.138
0.068
0.037
0.098
0.178
0.061
0.030
c(RGDfK)-Aoa-FDR
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem., 2013, 11, 4551–4558 | 4553

View Article Online
Paper
Organic & Biomolecular Chemistry
Fig. 3 Characterisation of αvβ3 expression on MCF7, LNCaP and PC3 cells. PC3
(solid line), LNCaP (dashed line) and MCF7 (alternating dotted/dashed line) cells
were incubated with an anti-αvβ3 receptor antibody and an Alexa-fluor488 sec-
ondary antibody. Cells were also incubated in the presence of secondary anti-
body only (dotted line). The relative level of αvβ3 expression was determined by
flow cytometry analysis as described in the General Experimental section. For
each treatment the number of cells (counts) was plotted against fluorescence
intensity at A488. Result is representative of duplicate samples.
Cell incubation experiments
PC3, LNCaP and MCF7 cells were chosen for the binding
experiments as they are known to express high, low and no
αvβ3 protein, respectively.24 Flow cytometry analysis (Fig. 3)
confirmed that our cell line clones expressed αvβ3 levels con-
sistent with published findings.
Binding of [18F]-labelled RGD peptides to αvβ3 was assessed
for each of these cell lines. Binding of c(RGDfK)-Aoa-[18F]FDR
and c(RGDfC)-Ahm-[18F]FDR was higher in PC3 cells compared
to the other cells (Fig. 4), consistent with the αvβ3 expression
levels (Fig. 3). Inclusion of cold c(RGDfK) (10 μM) to competi-
tively block specific binding to αvβ3 decreased [18F]RGD
binding to PC3 cells by approximately 60–70%, but had no
effect in the other two cell lines. These findings indicate that
the majority of bound radioactivity in PC3 cells can be attribu-
ted to specific binding to αvβ3, whereas the low level binding
to LNCaP and MCF7 is non-specific. Interestingly, in PC3 cells
c(RGDfK)-Aoa-[18F]FDR exhibited a higher level of binding
compared to c(RGDfC)-Ahm-[18F]FDR. This increase is consist-
ent with the higher in vitro affinity (Table 1), and suggests that
c(RGDfK)-Aoa-[18F]FDR is the better prospect for imaging
angiogenesis. It is concluded that these [18F]RGD conjugates
bind to αvβ3 in cells against a background of some non-specific
binding.
Fig. 4 Differential binding of 18F-labelled FDR-conjugated RGD peptides to
cultured cells that express different αvβ3 levels. MCF7, LNCaP and PC3 cells were
incubated with 18F-labelled c(RGDfK)-Aoa-[18F]FDR (a) or 18F-c(RGDfC)-Ahm-
[
18F]FDR (b) in the presence (+) or absence (−) of 10 μM c(RGDfK) peptide for
1 h, as described in methods and materials. 18F-RGD binding is expressed as
counts per minutes (cpm), as a function of the radioactivity added per plate and
normalised for protein content. Each datapoint is the average standard devi-
ation of at least two independent data points.
c(RGDfK)-Aoa-[18F]FDR and c(RGDfC)-Ahm-[18F]FDR were pre-
pared by conjugation with [18F]FDR.9
Cellular binding of c(RGDfK)-Aoa-[18F]FDR and c(RGDfC)-
Ahm-[18F]FDR was assessed in MCF7, LNCaP and PC3 cell
lines. In both cases the conjugated RGD peptides showed
selectivity for PC3 cells, which highly express αvβ3 integrin.
The c(RGDfC)-Aoa-[18F]FDR gave the better binding. In con-
clusion, the study demonstrates that [18F]FDR is an easily
accessible and efficient conjugation ligand for RGD peptides.
Its role could clearly be extended to PET radiolabelling studies
with other bespoke bioactive peptides.
Conclusions
[18F]FDR proved to be the most efficient bioconjugation agent
of several monosaccharides explored for conjugation to the
RGD peptides. The combination of the electronegative fluorine
located at the 5-position and 5 rather than a 6-membered ring
Experimental
General
monosaccharide, promotes ring opening to the reactive alde- All reagents and solvents were of highest grade from commer-
hyde form. Thus efficient conjugations occur in a few minutes cial sources, unless otherwise specified. Peptide c(RGDfC) was
at ambient temperature and at pH 4.6 to generate oximes. The purchased from Peptides International, USA and China-
conjugates c(RGDfK)-Aoa-FDR and c(RGDfC)-Ahm-FDR were Peptides. FDR and N-(6-aminoxyhexyl)maleimide hydrochlo-
assessed for their affinity to purified integrin αvβ3. In the event ride (Ahm) were prepared as previously described.9 6-Deoxy-6-
c(RGDfK)-Aoa-FDR showed the greater affinity. In each case fluoro-glucopyranose (6-FDG) was prepared by hydrolysis of
4554 | Org. Biomol. Chem., 2013, 11, 4551–4558
This journal is © The Royal Society of Chemistry 2013

View Article Online
Organic & Biomolecular Chemistry
Paper
the fully acetylated precursor. Normal phase column chrom-
For radiotracer binding experiments, cells were seeded at
atography was performed using silica gel 60 (40–63 μm). NMR 0.5 × 105 cells in 60 mm plates and cultured for 2–3 days to
spectra were recorded on Bruker Advance 300, 400 or 500 approximately 90% confluency. Medium was aspirated and
instruments. 1H and 13C NMR spectra were recorded using replace with 4 mL of fresh RPMI supplemented with 1 mM
deuterated solvent as the lock and residual solvent as the MnCl2. [18F]-labelled RGD peptides (approximately 0.5 MBq)
internal standard. 19F NMR spectra were referenced to CFCl3 was added per plate and incubated at 37 °C for 1 h in the pres-
as the external standard. Chemical shifts are reported in parts ence or absence of 10 μM cold cRGDfK peptide. Cells were
per million (ppm) and coupling constants ( J) are given in washed 5 times with ice-cold PBS, to remove unincorporated
Hertz (Hz). The abbreviations for the multiplicity of the radioactivity, then detached from plates with 0.5 mL trypsin/
proton, carbon and fluorine signals are as follows: s singlet, d EDTA, prior to addition of 0.5 mL medium and transfer into
doublet, dd doublet of doublets, ddd doublet of doublet of microfuge tubes. [18F]RGD binding to cells was measured
doublets, triplet, dt double triplets, q quartet, m multiplet, br using a well counter with a Nuclear Instruments (Oakland, UK)
s broad singlet. When necessary, resonances were assigned interface until at least 1000 counts were accumulated. Cells
using two-dimensional experiments (COSY, TOCSY, HMBC were then pelleted by centrifugation (200 g for 5 min), and
and HSQC). Mass analyses were recorded on a Micromass LCT protein content measured using the BCA protein assay. [18F]-
TOF mass spectrometer using ESI in positive mode. HPLC ana- RGD binding in each plate was expressed as a function of the
lyses/semi-preparations were performed using a Prostar 325 or radioactivity added per plate and normalised for protein
Shimadzu prominence or Agilent 1200 HPLC system with content.
reverse phase column as indicated in individual experiment.
HPLC analyses of radioactive compounds where performed
using a Shimadzu Co. (Kyoto, Japan) HPLC system equipped
with a SPD-M20A Prominence DAD UV detector (Shimadzu,
Japan) and NaI radio-detector (Berthold Technologies, Bad
Wildbad, Germany) using a Phenomenex Luna C-18(2) analyti-
cal column (250 × 4.6 mm, 5 μm, 100 A) equipped with the
corresponding guard column, 0.1% TFA in ACN (solvent A)
and 0.1% TFA in water (solvent B), flow rate of 1 mL min−1
and the following linear gradient: 100% solvent B for 4 min,
then from 0% to 60% of solvent A in 10 min and finally chan-
ging to 95% solvent A to flush the column. Semi-prep HPLC
purification of radioactive compounds were performed using a
Gynkotek HPLC system consisting of a gradient pump (P580),
column oven (STHS8S) and variable UV detector (UVD340S)
coupled in series with a BIOSCAN NaI detector (B-FC-3200)
and equipped with a Phenomenex Luna C-18(2) semi-prep
column (250 × 100 mm, 5 μm, 100 A). The dose calibrators
used to measure doses were CAPINTEC CRC 15R and CAPIN-
TEC CRC 15PET.
Cold bioconjugation of c(RGDfC)-Ahm with various sugars
A fresh solution of Ahm (179 μL, 44 mM, 7.88 μmol) in
degassed water was added to an Eppendorf tube containing
c(RGDfC) (5.0 mg, 7.84 μmol). The mixture was mixed well
with the pipette tip and incubated at 25 °C for 20 min. HPLC
monitoring indicated that the reaction reached to completion.
An aliquot of the resultant solution of c(RGDfC)-Ahm (20 μL,
44 mM, 0.88 μmol) was then mixed with a sugar (FDR, 2-FDG,
6-FDG, D-glucose or D-ribose) (20 μL, 44 mM, 0.88 μmol) in
NH4OAc buffer (pH = 4.6, 500 mM). The mixture was incu-
bated at 25 °C and an aliquot of the sample (2 μL) was taken at
the interval of 10 min, 30 min and 60 min respectively and
diluted with water (78 μL) for HPLC analysis. Conversion rate
to c(RGDfC)-Ahm-sugar was calculated based on the inte-
gration of the peak areas of the starting c(RGDfC)-Ahm and
the product c(RGDfC)-Ahm-sugar. HPLC was performed on
a Prostar 325 HPLC system with HiChrom C18 column
(250 × 4.6 mm, 10 μ). Mobile phase consisted of 5% aceto-
nitrile water with 0.1% formic acid (A) and 95% acetonitrile
with 0.1% formic acid (B). The program ran a linear gradient
from 0% B to 60% B in 20 min followed by returning to initial
to equilibrate the column. Peaks were detected at 220 nm with
Binding affinity tests. Binding affinity of cold bioconjugates
to purified αvβ3 integrin was determined by ELISA as described
previously.25
a flow rate of 1.0 mL min−1
.
Cell experiments
Preparative scale synthesis of c(RGDfC)-Ahm
MCF7, LNCaP and PC3 cells were purchased from ATCC and
confirmed as authentic and contamination-free. Cell charac- A fresh solution of Ahm (783 μL, 40 mM, 31.3 μmol) in
terisation for αvβ3 expression on MCF7, LNCaP and PC3 was degassed water was added to a screw cap glass vial containing
determined by flow cytometry analysis essentially as previously c(RGDfC) (20.0 mg, 31.3 μmol). The mixture was incubated at
described.24 An anti-αvβ3 primary antibody (LM609, Millipore) 25 °C for 30 min and then diluted with water. The resultant
was used at 1 : 100 dilution and detected with an Alexa- c(RGDfC)-Ahm was purified by reverse phase semi-prep HPLC
fluor488 labelled goat anti-mouse secondary antibody using a Phenomenex Kingsorb C18 column (250 × 10 mm,
(ThermoFisher scientific) at 1 : 500 dilution. Stained cells were 5 μ). Mobile phase consisted of A (H2O + 0.1% FA or 0.05%
detected on a FACS calibur flow cytometer (Becton Dickinson). TFA) and B (acetonitrile + 0.1% FA or 0.05% TFA). The
Alexa-fluor secondary antibody was incubated with cells in the program ran a linear gradient from 10% B to 40% B in 20 min,
absence of primary antibody to determine background cell 40% B to 90% B in 30 min and returning to initial condition
fluorescence.
at 3.5 mL min−1
.
The fraction (tR 16.8–17.2 min) was
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem., 2013, 11, 4551–4558 | 4555

View Article Online
Paper
Organic & Biomolecular Chemistry
concentrated under reduced pressure and lyophilized to give Preparation of c(RGDfK)-Aoa
1
the product as an amorphous solid (80–85%). H NMR (D2O,
The synthesis of c(RGDfK)-Aoa was accomplished following
400 MHz) δ 7.36–7.24 (m, 5H), 4.69 (t, 1H, J = 7.2 Hz), 4.65 (dd,
0.5 H, J = 6.3, 9.9 Hz), 4.56 (dd, 0.5 H, J = 6.0, 10.0 Hz, 9.9 Hz),
4.45 (dd, 0.5 H, J = 7.6, 4.7 Hz), 4.42 (dd, 0.5 H, J 8.7, 6.0 Hz),
4.36 (dd, 0.5 H, J = 8.7, 6.0 Hz), 4.30 (dd, 0.5 H, J = 9.5, 4.7 Hz),
4.22 (dd, 1H, J = 14.8, 2.8 Hz), 3.94 (t, 1H, J = 6.6 Hz,), 3.92 (t,
1H, J = 6.6 Hz), 3.80 (dd, 0.5 H, J = 8.9, 3.6 Hz), 3.57 (dd, 0.5 H,
J = 8.9, 3.9 Hz), 3.52–3.49 (m, 2H), 3.49 (dd, 1H, J = 14.8,
6.8 Hz), 3.26–2.90 (m, 4.5 H), 2.91–2.86 (m, 1H), 2.70 (dd, 1H),
2.57 (dd, 1H), 2.51 (dd, 0.5 H, J = 19.0, 4.2 Hz), 2.44 (dd, 0.5 H,
J = 19.0, 4.2 Hz), 1.93–1.85 (m, 1H), 1.74–1.67 (m, 1H),
1.66–1.60 (quintet, 2H, J = 7.0 Hz), 1.59–1.50 (m, 4H), 1.38
(quintet, 2H, J = 7.0 Hz), 1.30 (m, 2H). 13C NMR (D2O,
125 MHz, the number of signals are not the same as the
number of carbons due to the overlaps and splits between the
two diastereomers) δ 179.2, 178.1, 177.0, 173.2, 172.6, 172.5,
172.3, 171.8, 171.2, 156.4, 136.4, 129.3, 129.2, 128.8, 128.7,
127.1, 75.6, 55.7, 55.2, 54.7, 54.3, 52.7, 52.6, 50.6, 43.6, 40.7,
40.5, 39.5, 39.8, 37.2, 36.4, 36.2, 35.6, 26.9, 26.5, 25.6, 24.4; m/z
(ESI+) 791.3512 [M + H]+, C34H51N10O10S requires 791.3510.
the procedure reported by McCusker et al.23 starting from
Fmoc-Asp(NovaSyn® TGA)-OAll resin (1 g, 0.2 mmol g−1
loading) (Scheme 2). After removal of the (4,4-dimethyl-2,6-
dioxocyclohexylidene)ethyl (Dde) protecting group by hydra-
zine monohydrate–DMF (2 : 98) for 3 min, the lysine amino
function was coupled with (Boc-aminooxy)acetic acid, using
HATU (2 eq.) and sym-collidine (4 eq.), rt for 2 h. A solution of
TFA–CH2Cl2 (1 : 1) was added to allow the cleavage of the
peptide from the resin and the removal of pentamethyl-di-
hydrobenzofuran-5-sulfonyl (Pbf) and tert-butyloxycarbonyl
(Boc) protecting groups, at rt for 2 h. The solution was concen-
trated and precipitated with Et2O. The precipitate was purified
by a Phenomenex Luna C18 (250 × 10.00 mm, 5 μ); mobile
phase: A (H2O, TFA 0.1%) B (MeCN); gradient: from 15% B to
30% B in 15 min; flow: 5 mL min−1; tR: 6.7 min) and lyophi-
lised to give 30 mg of a white, fluffy solid (22% yield based on
initial loading resin). m/z (ESI)+: 677.2 [M + H]+, C29H45N10O9
requires 677.3.
Cold bioconjugation of c(RGDfK)-Aoa with FDR
Cold preparative scale bioconjugation of c(RGDfC)-Ahm with
FDR
To a solution of c(RGDfK)-Aoa (3.5 mg, 5 μmol) in 100 μL of
water was added FDR (1 mg, 6.5 μmol) in sodium acetate
buffer (110 μL, 0.5 M, pH 4.6) (Fig. 2). The mixture was allowed
to react at 25 °C for 10 min, then was diluted with water
(1 mL) and purified by a Phenomenex Luna C18 (250 ×
10.00 mm, 5 μ); mobile phase: A (H2O + 0.1% TFA), B (MeCN);
FDR (1.8 mg, 11.8 μmol) in NH4OAc buffer (295 μL, 250 mM,
pH 4.6) was added to a screw cap glass vial containing
c(RGDfC)-Ahm (5 mg, 5.9 μmol). The mixture was incubated at
25 °C for 30 min and then purified by reverse phase semi-prep
HPLC using a Phenomenex Kingsorb C18 column (250 ×
10 mm, 5 μ) with an isocratic condition of 25% of acetonitrile
in water with 0.1% formic acid at 3.5 mL min−1. The fraction
(tR 14.5–15.5 min) was concentrated under reduced pressure
and lyophilized to give the product as an amorphous solid
(4.9 mg, 86%, E : Z = 84 : 16). Minor resonances are denoted by
an asterisk. 1H NMR (D2O, 500 MHz) δ 7.55 (d, 0.8 H, J =
6.7 Hz), 7.37–7.25 (m, 5H, PhH), 6.89* (d, 0.2 H, J = 6.0 Hz),
5.05* (dd, 0.2 H, J = 6.0, 3.2 Hz), 4.70–4.63 (m, 2.5 H),
4.58–4.53 (m 1.5 H), 4.48–4.42 (m, 1.8 H), 4.39 (dd, 0.5 H, J =
7.8, 6.0 Hz), 4.30 (dd, 0.5 H, J = 10.0, 4.8 Hz), 4.22 (dd, 1H, J =
14.3, 3.2 Hz), 4.09 (t, 1H, J = 6.3 Hz), 3.92–3.77 (m, 3 H), 3.52
(t, 2H, J = 6.6 Hz), 3.48 (dd, 1H, J = 14.7, 2.8 Hz), 3.26–3.08 (m,
5 H), 3.03 and 3.01 (2 × d, 1 H, J = 12.0 Hz), 2.90–2.84 (m, 1 H),
2.53 (dd, 1H, J = 16.0, 6.9 Hz), 2.57 (dd, 1H, J = 16.0, 6.9 Hz),
2.51–2.41 (m, 1 H), 1.91–1.86 (m, 1H), 1.75–1.68 (m, 1H),
1.68–1.62 (m, 2H), 1.62–1.50 (m, 4H), 1.42–1.35 (m, 4H),
1.35–1.27 (m, 2H). 13C NMR (D2O, 125 MHz, number of
signals are not the same as the number of carbons due to the
overlaps and splits between the two diastereoisomers) 179.7,
179.0, 178.1, 177.9, 178.0, 173.3, 172.8, 172.6, 172.4, 171.7,
gradient: from 15% B to 30% B in 15 min; flow: 5 mL min−1
;
tR: 9.2 min). The purified peptide was lyophilized affording
4 mg of a white solid (99% yield) containing a mixture of E/Z
isomers (4 : 1). Minor resonances are denoted by an asterisk.
1H NMR (400 MHz, D2O) δ 7.63 (d, 0.8H, J = 6.5 Hz) 7.32–7.17
(m, 5H), 6.96* (d, 0.2H, J = 6.2 Hz), 5.06* (dd, 0.2H, J = 6.1,
3.5 Hz), 4.63–4.60 (m, 1H), 4.57–4.49 (m, 1H), 4.43 (dd, 0.8H,
J = 6.5, 3.8 Hz), 4.28 (dd, 1H, J = 8.9, 5.6 Hz), 4.13 (d, 1H, J =
15.0 Hz), 3.89–3.73 (m, 3H), 3.42 (d, 1H, J = 15.0 Hz), 3.17–3.05
(m, 4H), 2.99 (dd, 1H, J = 13.2, 6.2 Hz), 2.92–2.79 (m, 2H), 2.64
(dd, 1H, J = 16.7, 6.5 Hz), 1.83–1.74 (m, 1H), 1.62–1.52 (m, 2H),
1.48–1.38 (m, 3H), 1.34–1.27 (m, 2H), 0.92–0.84 (m, 2H), 13C
NMR (101 MHz, D2O) δ 174.6, 174.6, 173.1, 172.7, 171.8, 171.4,
171.3, 156.7, 153.7*, 152.8, 136.0, 129.1, 128.8, 127.3, 84.6 (d,
J = 165.4 Hz), 72.1 (d, J = 6.8 Hz), 71.9, 70.0, 69.8*, 69.3, 65.9*,
55.4, 55.1, 52.4, 49.7, 43.5, 40.5, 38.6, 37.0, 34.5, 29.9, 27.5,
27.3, 24.4, 22.5, 19F{1H} NMR (376 MHz, D2O) δ −233.99,
−234.18*. m/z (ESI)+: 811.4 [M + H]+, C34H52FN10O12 requires
811.3.
Hot bioconjugation of c(RGDfC)-Ahm with [18F]FDR
171.3, 156.6, *151.5 (E), 150.4 (Z), 135.8, 128.8, 128.7, 127.0, [18F]FDR was produced at the John Mallard Scottish PET
1
2
84.4 (d, JCF = 161 Hz), 73.9, 71.9, 69.9 (d, JCF 14 Hz), 69.3, centre according to the procedure previously reported,9 in a
55.7, 55.2, 54.5, 54.0, 52.6, 50.7, 43.8, 43.6, 40.4, 38.9, 37.9, final concentration of 300 MBq mL−1 solution and radiochemi-
36.8, 35.8, 35.5, 33.1, 33.9, 28.0, 27.0, 25.7, 24.5, 24.3. 19F{1H} cal purity greater than 98%. To 2 mg of c(RGDfC)-Ahm peptide
NMR (470 MHz, D2O), −234.6, *−235.3; m/z (ESI+) 925.3872 dissolved in 20 μL of sodium acetate buffer (1 M, pH 4.6) in an
[M + H]+, C39H58N10O13FS requires 925.3890.
eppendorf vial were added 200 μL of [18F]FDR (40–75 MBq).
4556 | Org. Biomol. Chem., 2013, 11, 4551–4558
This journal is © The Royal Society of Chemistry 2013

View Article Online
Organic & Biomolecular Chemistry
Paper
The mixture was allowed to react at 25 °C for 15 min. Then the
buffered solution was injected in the semi-prep HPLC system
and purified using an isocratic method (15% ACN in water with
0.1% TFA, 4.6 mL min−1). Decay corrected radiochemical yield,
calculated considering the activity added to the peptide, was 80%
(55% non-decay corrected). The collected fraction was diluted
with 10 mL of water and loaded into a Waters Oasis® HLB Car-
tridge (conditioning 2 mL EtOH, 5 mL water). The cartridge was
washed with 20 mL of water. The desired product was collected
eluting with 1.5 mL of a solution 1 : 1 ethanol–water. This non-
optimised procedure allowed the recovery of 75% of the loaded
activity. Radio HPLC analysis of the formulated product showed
>98% radiochemical and chemical purity. Identity of the product
was confirmed by comparison with the cold reference. It was not
possible to determine the specific activity since no UV signal of
the purified product was detectable from the HPLC system.
K. Jurkschat, C. Wangler and R. Schirrmacher, Nat. Proto-
cols, 2012, 7, 1946–1955, and references therein.
2 (a) L. Cai, S. Lu and V. W. Pike, Eur. J. Org. Chem., 2008,
2853–2873; (b) S. M. Ametamey, M. Honer and
P. A. Schubiger, Chem. Rev., 2008, 108, 1501–1516.
3 (a) S. M. Okarvi, Eur. J. Nucl. Med., 2001, 28, 929–938;
(b) B. Kuhnast and F. Dollé, Curr. Radiopharm., 2010, 3,
174–201; (c) D. E. Olberg and O. K. Hjelstuen, Curr. Top.
Med. Chem., 2010, 10, 1669–1679.
4 (a) J. W. Fletcher, B. Djulbegovic, H. P. Soares, B. A. Siegel,
V. J. Lowe, G. H. Lyman, R. E. Coleman, R. Wahl,
J. C. Paschold, N. Avril, L. H. Einhorn, W. W. Suh,
D. Samson, D. Delbeke, M. Gorman and A. F. Shields,
J. Nucl. Med., 2008, 49, 480–508. For a recent review on
carbohydrates and derivatives in molecular imaging:
(b) G. R. Morais, R. A. Falconer and I. Santos, Eur. J. Org.
Chem., 2013, 1401–1414.
5 X.-G. Li, M. Haaparanta and O. Solin, J. Fluorine Chem.,
2012, 143, 49–56.
6 O. Jacobson, L. Zhu, Y. Ma, I. D. Weiss, X. Sun, G. Niu,
D. O. Kiesewetter and X. Chen, Bioconjugate Chem., 2011,
22, 422–428.
7 T. Poethko, M. Schottelius, G. Thumshirn, U. Hersel,
M. Herz, G. Henriksen, H. Kessler, M. Schwaiger and
H.-J. Wester, J. Nucl. Med., 2004, 45, 892–902.
8 P. Laverman, W. J. McBride, R. M. Sharkey, A. Eek,
L. Joosten, W. J. G. Oyen, D. M. Goldenberg and
O. C. Boerman, J. Nucl. Med., 2010, 51, 454–461.
Hot bioconjugation of c(RGDfK)-Aoa with [18F]FDR
To 2 mg of c(RGDfK)-Aoa peptide dissolved in 20 μL of sodium
acetate buffer (1 M, pH 4.6) in an Eppendorf vial were added
200 μL of [18F]FDR (15–47 MBq). The mixture was allowed to
react at 25 °C for 15 min. Then the buffered solution was
injected in the semi-prep HPLC system and purified using a
linear gradient from 15% ACN in water + 0.1% TFA to 20% ACN
in water + 0.1% TFA in 15 min. Decay corrected radiochemical
yield, calculated considering the activity added to the peptide,
was 60% (46% non-decay corrected). The collected fraction was
diluted with 10 mL of water and loaded into a Waters Oasis®
HLB Cartridge (conditioning 2 mL EtOH, 5 mL water). The car-
tridge was washed with 20 mL of water. The desired product was
9 L. Xiang-Guo, S. Dall’Angelo, L. F. Schweiger, M. Zanda and
D. O’Hagan, Chem. Commun., 2012, 48, 5247–5249.
collected eluting with 1.5 mL of a solution 1 : 1 ethanol–water. 10 C. J. Avraamides, B. Garmy-Susini and J. A. Varner, Nat.
This non-optimised procedure allowed the recovery of 80% of Rev. Cancer, 2008, 8, 604–617.
the loaded activity. Radio HPLC analysis of the formulated 11 (a) Z. Liu, F. Wang and X. Chen, Drug Dev. Res., 2008, 69,
product showed >98% radiochemical and chemical purity. It was
not possible to determine the specific activity since no UV signal
of the purified product was detectable from the HPLC system.
329–339; (b) K. Chen and X. Chen, Theranostics, 2011, 1,
189–200; (c) P. Schaffner and M. M. Dard, Cell. Mol. Life
Sci., 2003, 60, 119–132.
12 R. Haubner, R. Gratias, B. Diefenbach, S. L. Goodman,
A. Jonczyk and H. Kessler, J. Am. Chem. Soc., 1996, 118,
7461–7472.
13 J. M. Jeong, M. K. Hong, Y. S. Chang, Y.-S. Lee, Y. J. Kim,
G. J. Cheon, D. S. Lee, J.-K. Chung and M. C. Lee, J. Nucl.
Med., 2008, 49, 830–836.
Note added in proof
During production of this paper radiolabelling of the RGD
peptide Siglec-9 with [18F]FDR was reported; see ref 26.
14 T. J. Wadas, H. Deng, J. E. Sprague, A. Zheleznyak,
K. N. Weilbaecher and C. J. Anderson, J. Nucl. Med., 2009,
50, 1873–1880.
Acknowledgements
15 (a) M. Schottelius, B. Laufer, H. Kessler and H. Wester,
Acc. Chem. Res., 2009, 42, 969–980; (b) P. Haubner,
H.-J. Wester, U. Reuning, R. Senekowitsch-Schmidtke and
B. Diefenbach, et al., J. Nucl. Med., 1999, 40, 1061–1071;
(c) M. Schottelius and H. J. Wester, Methods, 2009, 48, 161–
177; (d) H. Cai and P. S. Conti, J. Labelled Compd. Radio-
pharm., 2013, 56, 264–279; (e) R. Haubner, B. Kuhnast,
C. Mang, W. A. Weber, H. Kessler, H.-J. Wester and
M. Schwaiger, Bioconjugate Chem., 2004, 15, 61–69.
We thank the EPSRC (GR/EP/I034734/1) for supporting this
work. SINAPSE (http://www.sinapse.ac.uk/) is gratefully
acknowledged for a fellowship to S.D. We would like to thank
the Aberdeen University flow cytometry facility for helping to
design and analyse the αvβ3 cell characterisation experiment.
References
1 B. Wängler, A. P. Kostikov, S. Niedermoser, J. Chin, 16 (a) A. J. Beer, R. Haubner, M. Goebel, S. Lunderschmidt,
K. Orchowski, E. Schirrmacher, L. Lovkova-Berends,
M. E. Spilker, H.-J. Wester, W. A. Weber and M. Schwaiger,
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem., 2013, 11, 4551–4558 | 4557

View Article Online
Paper
J. Nucl. Med., 2005, 46, 1333–1341; (b) A. J. Beer,
Organic & Biomolecular Chemistry
M. Yang, X. Sun, S. Liu and X. Chen, Amino Acids, 2011, 41,
439–447; (d) N. Guo, L. Lang, W. Li, D. O. Kiesewetter,
H. Gao, G. Niu, Q. Xie and X. Chen, PLoS One, 2012, 7,
e37506.
R. Haubner, I. Wolf, M. Goebel, S. Luderschmidt,
M. Niemeyer, A.-L. Grosu, M.-J. Martinez, H.-J. Wester,
W. A. Weber and M. Schwaiger, J. Nucl. Med., 2006, 47,
763–769; (c) A. J. Beer, S. Lorenzen, S. Metz, K. Herrmann, 20 (a) M. Doss, H. J. Kolb, J. J. Zhang, M.-J. Belanger,
P. Watzlowik, H.-J. Wester, C. Peschel, F. Lordick and
M. Schwaiger, J. Nucl. Med., 2008, 49, 22–29; (d) T. Higuchi,
F. M. Bengel, S. Seidl, P. Watzlowik, H. Kessler,
R. Hegenloh, S. Reder, S. G. Nekolla, H.-J. Wester and
M. Schwaiger, Cardiovasc. Res., 2008, 78, 395–403.
J. B. Stubbs, M. G. Stabin, E. D. Hostetler, R. K. Alpaugh,
M. von Mehren, J. C. Walsh, M. Haka, V. P. Mocharla and
J. Q. Yu, J. Nucl. Med., 2012, 53, 787–795; (b) A. C. Valdivia,
M. Estrada, T. Hadizad, D. J. Stewart, R. S. Beanlands and
J. N. DaSilva, J. Labelled Compd. Radiopharm., 2012, 55, 57–
60.
17 (a) L. M. Kenny, R. C. Coombes, I. Oulie, K. B. Contractor,
M. Miller, T. J. Spinks, B. McParland, P. S. Cohen, 21 M. Glaser, M. Solbakken, D. R. Turton, R. Pettit, J. Barnett,
A.-M. Hui, C. Palmieri, S. Osman, M. Glaser, D. Turton,
J. Arukwe, H. Karlsen, A. Cutherbertson, S. K. Luthra and
A. At-Nahhas and E. O. Aboagye, J. Nucl. Med., 2008, 49,
E. Arstad, Amino acids, 2009, 37, 717–724.
879–886; (b) M. R. Battle, J. L. Goggi, L. Allen, J. Barnett 22 B. M. Hutchins, S. A. Kazane, K. Staflin, J. S. Forsyth,
and M. S. Morrison, J. Nucl. Med., 2011, 52, 424–430.
B. Felding-Habermann, V. V. Smider and P. G. Schultz,
18 (a) C. Hultsch, M. Schottelius, J. Auernheimer, A. Alke and
Chem. Biol., 2011, 18, 299–303.
H.-J. Wester, Eur. J. Nucl. Med. Mol. Imaging, 2009, 36, 23 C. F. McCusker, P. J. Kocienski, F. T. Boyle and
1469–1474; (b) M. Namavari, Z. Cheng, R. Zhang, A. De, A. G. Schätzlein, Bioorg. Med. Chem. Lett., 2002, 12, 547.
J. Levi, J. K. Hoerner, S. S. Yaghoubi, F. A. Syud and 24 Y. Sun, M. Fang, J. Wang, C. R. Cooper, K. J. Pienta and
S. S. Gambhir, Bioconjugate Chem., 2009, 20, 432–436. R. S. Taichman, Prostate, 2007, 67, 61–73.
19 (a) X. Zhang, Z. Xiong, Y. Wu, W. Cai, J. R. Tseng, 25 M. Piras, I. N. Fleming, W. T. A. Harrison and M. Zanda,
S. S. Gambhir and X. Chen, J. Nucl. Med., 2006, 47, 113– Synlett, 2012, 23, 2899–2902.
121; (b) L. Lang, W. Li, N. Guo, Y. Ma, L. Zhu, 26 X.-G. Li, A. Autio, H. Ahtinen, K. Helariutta, H. Liljenbàck,
D. O. Kiesewetter, B. Shen, G. Niu and X. Chen, Bioconju-
gate Chem., 2011, 22, 2415–2422; (c) Y. Yan, K. Chen,
A. Roivainen and A. J. Airaksinen, Chem. Commun., 2013,
49, 3682–3684.
4558 | Org. Biomol. Chem., 2013, 11, 4551–4558
This journal is © The Royal Society of Chemistry 2013