Enzymatic transhalogenation of dendritic RGD peptide constructs with the fluorinase

Article abstract of DOI:10.1039/c6ob00239k

The substrate scope of fluorinase enzyme mediated transhalogenation reactions is extended. Substrate tolerance allows a peptide cargo to be tethered to a 5′-chloro-5′-deoxynucleoside substrate for transhalogenation by the enzyme to a 5′-fluoro-5′-deoxynuc

Full text of DOI:10.1039/c6ob00239k

Organic &
Biomolecular Chemistry
View Article Online
View Journal
PAPER
Enzymatic transhalogenation of dendritic RGD
peptide constructs with the fluorinase†
Cite this: DOI: 10.1039/c6ob00239k
Stephen Thompson,a Ian N. Flemingb and David O’Hagan*a
The substrate scope of fluorinase enzyme mediated transhalogenation reactions is extended. Substrate
tolerance allows a peptide cargo to be tethered to a 5’-chloro-5’-deoxynucleoside substrate for trans-
halogenation by the enzyme to a 5’-fluoro-5’-deoxynucleoside. The reaction is successfully extended from
that previously reported for a monomeric cyclic peptide (cRGD) to cargoes of dendritic scaffolds carrying
two and four cyclic peptide motifs. The RGD peptide sequence is known to bind upregulated αVβ3 integrin
motifs on the surface of cancer cells and it is demonstrated that the fluorinated products have a higher
affinity to αVβ3 integrin than their monomeric counterparts. Extending the strategy to radiolabelling of the
peptide cargoes by tagging the peptides with [18F]fluoride was only moderately successful due to the
poor water solubility of these higher order peptide scaffolds although the strategy holds promise for
peptide constructs with improved solubility.
Received 28th January 2016,
Accepted 15th February 2016
DOI: 10.1039/c6ob00239k
www.rsc.org/obc
While two step strategies involving prosthetic groups
remain widely used for radiolabelling of peptides and proteins
1. Introduction
The fluorinase enzyme (E.C. 2.5.1.63) was originally isolated1 with fluorine-18,8 it is attractive to consider the development
from the soil bacterium Streptomyces cattleya and has recently of straightforward, “last step” radiolabelling protocols. Such
been identified in a number of other bacterial species.2 The protocols need to be relatively rapid due to the short half-life
enzyme shows promise as a biocatalyst for the incorporation of of fluorine-18 (t1/2 = 109.7 min), and be conducted under con-
fluorine-18 into [18F]-labelled PET radiotracers, and has the ditions that preserve the structural and functional integrity of
singular attribute that the C–F bond forming reaction is the biomolecule.9 The poor nucleophilicity of aqueous [18F]flu-
carried out under neutral aqueous conditions.3–6 The fluori- oride ion, the form most readily prepared on a cyclotron with
nase catalyses the combination of fluoride ion and S-adenosyl- an [18O]H2O target,10 often precludes direct labelling of bio-
L-methionine 1 (SAM) to generate 5′-fluoro-5′-deoxyadenosine 2 molecules using a “last step” approach. To circumvent the
(FDA) and L-(S)-methionine 3 (L-Met) (Scheme 1A).1,7 We have limited reactivity of aqueous [18F]fluoride, fluoride sequester-
recently demonstrated that the enzyme has a specificity toler- ing strategies have been explored. For example boron,11 alu-
ance at the 2-position of the adenine base.5 Replacement of minium12 and silicon13 based methods have been developed
hydrogen with an acetylene group at this position, as shown in in an effort to achieve “last step” radiolabelling of peptides by
Scheme 1B, gave the modified substrate ClDEA 4 which was an bioconjugation approaches which secure these elements to the
efficient substrate for transhalogenation in the presence of peptide in advance. The strategies have proven successful and
Se-adenosyl-L-selenomethionine (SeSAM).7 This transformation these technologies are being further developed,14,15 but there
was utilised to prepare [18F]FDEA [18F]-6 under ambient are challenges as the conditions required to achieve fluoride
aqueous conditions. The pendant acetylene moiety was then ion sequestration require various combinations of low pH,
efficiently reacted with an azido-cRGD (cyclic arginyl-glycinyl- organic solvents and heating. The fluorinase has the capacity
aspartyl) peptide using the CuAAC (Cu-catalysed alkyne–azide to catalyse C–F bond formation at ambient temperature and
cycloaddition) click reaction, to generate a [18F]-radiolabelled near neutral pH. Within this context, we have previously
cRGD construct.6
demonstrated5 a fluorinase enzyme-based system for “last
step” radiolabelling of a small cRGD peptide 7. The peptide
was tethered to
a 5′-chloro-5′-deoxynucleoside substrate
aSchool of Chemistry and Biomedical Sciences Research Centre, University of
through a PEG linker prior to enzymatic radiolabelling with
[18F]fluoride, as shown in Scheme 1B. This proved successful
and the radiolabelled construct [18F]-8 was shown to be stable
to defluorination in vivo (in a rat), and the cRGD moiety
retained high affinity for αVβ3 integrins. In order to explore
St Andrews, North Haugh, St Andrews KY16 9ST, UK. E-mail: do1@st-andrews.ac.uk
bAberdeen Biomedical Imaging Centre, School of Medicine and Dentistry,
University of Aberdeen, Foresterhill, Aberdeen, AB25 2ZD, UK
†Electronic supplementary information (ESI) available. See DOI: 10.1039/
c6ob00239k
This journal is © The Royal Society of Chemistry 2016
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
Scheme 1 A. The native reaction catalysed by the fluorinase enzyme. B. Fluorinase catalysed transhalogenation reactions with C-2 modified
substrates.
this approach further, we wished to examine whether the sub- tors can be enhanced by attaching multiple ligands to a single
strate scope at the C-2 position would extend to larger, more scaffold.19,20 Incorporating the RGD (arginyl–glycinyl–aspartyl)
complex peptides.
tripeptide motif into cyclic pentapeptides gives cRGDs, a class
The search for high affinity ligands for biological receptors of high affinity ligands for αVβ3 integrin, a cell surface protein
has led to the development of imaging agents based on multi- identified as a biomarker of angiogenesis and strongly associ-
valent antibody fragments16 or multivalent small peptides.17 ated with malignant tumours.21,22 Multimeric cRGD peptides
Multimeric interactions play a key role in recognition events are validated constructs employed for PET imaging using
between biological ligands and receptors.18 Weak non-covalent either fluorine-18 or heavy metal PET isotopes.19,23,24 Usually
interactions between monovalent ligands and individual recep- the cRGD peptide is conjugated to a metal chelating group for
Scheme 2 Synthesis of protected aldehyde 11 and amine 14 for reductive amination to generate the desired alkynyl frameworks 15 and 16.
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Organic & Biomolecular Chemistry
Paper
using a “last step” radiofluorination protocol. These are the
aluminium chelates, [18F]alfatide I14,29 and [18F]alfatide II15,30
which were prepared using a kit-like31 methodology and they
have been evaluated in humans.14,15 In addition, the CuAAC
reaction between an alkyne and an azide-modified cRGD
peptide has also proved useful for the assembly of multimeric
cRGD peptides, which were subsequently labelled with
radiometals.32–35
Herein, we report our initial results on “last step” fluorina-
tion of structurally complex multimeric cRGD peptides with
the fluorinase enzyme. A primary objective at this stage was to
establish if the fluorinase could tolerate higher molecular
weight cargoes attached to the C-2 position of the adenine
base. Although successful biotransformations could be
achieved with fluoride, it became apparent that reduced solu-
bility of higher molecular weight substrates proved challenging
for efficient radiolabelled experiments, where the fluoride-18
concentrations are necessarily low.
Scheme 3 Synthesis of halogenated 2-iodo coupling partners 18 and
20 for Sonogashira coupling.
2. Results and discussion
2.1. Synthesis of di- and tetra-alkynes for a CuAAC reaction
co-ordination to radiometals, often at the final step in the syn-
thesis of such a tracer. Fewer fluorine-18-based multivalent
radiotracers have been developed, as incorporation of fluorine-
18 into such scaffolds often requires a multi-step, prosthetic
group-based strategy.25–28 However, two multimeric cRGD-
based radiotracers have been reported recently, synthesised
A multimeric peptide substrate for the fluorinase requires an
enzyme recognition/reaction site attached to a linker to extend
the peptide cargo away from the surface of the enzyme.
In order to assemble such a substrate, we envisaged a strategy
Scheme 4 Synthesis of pendant di- and tetra-acetylene cores attached to 5’-halo-5’deoxyadenosine motifs.
This journal is © The Royal Society of Chemistry 2016
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
similar to that used for the synthesis of 5 and 7 (Scheme 1B).5 gave amine 14. With both aldehyde 11 and amine 14 in
A Sonogashira coupling was identified for coupling a 2-iodo- hand, attention turned to coupling the linker to the multi-
nucleoside to an alkynyl-PEG linker, which had been pre- meric core.
functionalised with either a protected dialkyne or a protected
The reductive amination after condensation of amine 14
tetraalkyne. Following deprotection to the free terminal with aldehyde 1138 generated both the mono- and di-alkylated
alkynes, a multi-CuAAC reaction with an azide bearing cRGD products 15 and 16. Separation of the two products by column
peptide, similar to that reported by Liskamp et al.,34 should chromatography gave the di-alkyne linker-core assembly 15 in
furnish multimeric cRGD substrates. The synthesis of the pro- 35% yield (based on aldehyde 11), while the terta-alkyne
tected alkynes and halogenated 2-iodo-nucleosides is summar- linker-core assembly 16 was formed in 21% yield (based on
ised in Schemes 2 and 3 respectively.
aldehyde 11).
The dendrimeric core was prepared by dialkylation of
With these linker-core assemblies in hand, attention turned
methyl 3,5-dihydroxybenzoate
9 with propargyl bromide to the synthesis of the iodinated coupling partners. 2-Iodo-
(Scheme 2).34 The acetylenes of the newly installed propargylic adenosine 17,5 was treated with thionyl chloride in pyridine,
ethers were then silylated to furnish ester 10. Ester 10 was followed by ammonia in aqueous methanol to furnish the
then reduced to the corresponding benzylic alcohol and peri- 5′-chloro-5′-deoxy adenosine 18, as shown in Scheme 3.39 For
odinane oxidation36 generated aldehyde 11 which was used the synthesis of the fluorinated nucleoside, 2-iodoadenosine
directly in future reductive amination reactions without 17 was protected as its 2′,3′-acetonide 19 before being fluori-
further purification. The linker was prepared by monoalkyl- nated by the action of TsF and TBAF in refluxing THF.
ation of tetraethylene glycol 12 with propargyl chloride. The Deprotection to generate 20 was achieved in good yield with
free hydroxyl was converted to the corresponding tosyl ester catalytic anhydrous TsOH in refluxing MeOH.40
and then nucleophilic displacement by azide gave the corres-
Sonogashira couplings41–43 of the halogenated 2-iodo-
ponding azido-PEG. Reduction under Staudinger conditions37 nucleosides 18 and 20 with acetylenes 15 and 16 were now
Scheme 5 A. Synthesis of ClDEA-PEG-(RGD)2 30 and FDEA-PEG-(RGD)2 31. B. Synthesis of ClDEA-PEG-(RGD)4 32 and FDEA-PEG-(RGD)4 33.
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Organic & Biomolecular Chemistry
Paper
explored. 5′-Chloro-5′-deoxy-2-iodoadenosine 18 was added to
It became a focus to ‘click’ the multimeric alkynes to the
an excess of either alkyne 15 or 16 in dry, degassed DMF, in cRGD peptides in aqueous CuAAC reactions with c(RGDfK[N3])
the presence of Pd(PPh3)2Cl2 (10 mol%), CuI, and triethyl- 29 as illustrated in Scheme 5A. The reactions were conducted
amine. TLC and LC-MS was used to monitoring the formation with CuSO4·TBTA (TBTA = tris(benzyltriazolylmethyl)amine44)
of coupled products 21 and 25 (Scheme 4).
and sodium ascorbate as the catalyst, and were monitored by
Coupled di-alkyne 21 and tetra-25 were subject to column HPLC. The triazole products were trapped on a C18 cartridge,
chromatography before fluoride-based deprotection of the silyl and eluted with MeCN/water mixture and purified by semi-
groups using 3HF·Et3N in methanol. The products 22 and 26 preparative HPLC. This gave samples of ClDEA-PEG-(RGD)2 30
were each passed through a plug of silica gel, then through a (1.8 mg, 78% yield) and FDEA-PEG-(RGD)2 31 (0.67 mg, 27%
C18 reverse phase cartridge, before being purified by semi- yield), the integrity of which was supported by MALDI-TOF MS
preparative HPLC to give analytically pure samples of ClDEA- analyses (see ESI†). The fluorinated analogue, FDEA-PEG-
PEG-(CuCH)2 22 and ClDEA-PEG-(CuCH)4 26 for CuAAC reac- (RGD)2 31, was required as a standard for assessing affinity to
tions. Following a similar protocol, FDEA-PEG-(CuCH)2 24 targets.
and FDEA-PEG-(CuCH)4 28 were prepared and purified by
semi-preparative HPLC.
ClDEA-PEG-(CuCH)4 26 was found to be poorly soluble
in water and the CuAAC reaction with c(RGDfK[N3]) 29
Fig. 1 A. Fluorinase catalysed transhalogenation of ClDEA-PEG-(RGD)2 30 to FDEA-PEG-(RGD)2 31. B. HPLC time course (220 nm) of incubation of
ClDEA-PEG-(RGD)2 30, blue, (tR = 12.2 min) with the fluorinase, showing samples taken every 0.5 h over 4 h. A new peak was evident at tR = 8.2 min,
red, identified as FDEA-PEG-(RGD)2 31. Summed mass spectrum (m/z = 400–1000) of the LC-MS peaks corresponding to the tetrameric species
from a sample taken at C. t = 0 h, showing peaks corresponding to multiply charged species of ClDEA-PEG-(RGD)2 30, and at D. t = 4 h, showing
peaks corresponding to multiply charged species of FDEA-PEG-(RGD)2 31.
This journal is © The Royal Society of Chemistry 2016
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
(Scheme 5B) was sluggish, but the use of 50 : 50 DMSO : water water, with c(RGDfK[N3]) 29. HPLC purification gave the
as a reaction solvent led to complete conversion within desired reference compound FDEA-PEG-(RGD)4 33 (0.66 mg,
70 minutes. The reaction product was trapped on a C18 car- 27% yield) with a m/z = 3412.6 ([M + H]+) as determined by
tridge, and was washed with water to remove residual DMSO, MALDI-TOF MS (see ESI†).
and then the product was eluted with MeCN : water mixtures.
Purification by semi-preparative HPLC gave ClDEA-PEG-(RGD)4
32 (1.4 mg, 70% yield), the integrity of which was confirmed
2.2. Evaluation of multimeric RGDs as substrates for the
fluorinase
by MALDI–TOF MS (see ESI†). The tetra-acetylene, FDEA-PEG-
(CuCH)4 26 was similarly reacted in a mixture of DMSO :
With the multimeric cRGD fluorinase substrates in hand,
enzyme catalysed transhalogenation reactions were explored.
Fig. 2 A. Fluorinase catalysed transhalogenation of ClDEA-PEG-(RGD)4 32 to FDEA-PEG-(RGD)4 33. B. HPLC time course (220 nm) of incubation of
ClDEA-PEG-(RGD)4 32, green, (tR = 12.4 min) with the fluorinase, showing samples taken every 0.5 h over 4 h. A new peak was evident at tR
=
9.2 min, yellow, identified as FDEA-PEG-(RGD)4 33. Summed mass spectrum (m/z = 600–1000) of the LC-MS peak corresponding to the tetrameric
species of samples taken at B. t = 0 h, showing peaks corresponding to multiply charged species of ClDEA-PEG-(RGD)4 32 and at C. t = 16 h,
showing peaks corresponding to multiply charged species of FDEA-PEG-(RGD)4 22. Identical fragmentation peaks (denoted with an asterisk (*))
appear in both spectra, thought to arise from fragmentation of the adenine (C-2)–(CuCR) bond.
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Organic & Biomolecular Chemistry
Paper
ClDEA-PEG-(RGD)2 30 (40 µM) was added to a mixture of 2.3. Binding affinities of RGD multimers to immobilised
L-SeMet (75 µM) and potassium fluoride (50 mM) in a 1 mL αVβ3 integrin
volume to explore the transformation illustrated in Fig. 1A.
The binding affinities of the fluorinated multimers 31 and 33
to αVβ3 integrin were determined by their ability to displace a
biotin-labelled cRGD peptide from immobilised αVβ3 integrin
The reaction was initiated by addition of the fluorinase
(1 mg mL−1).
Aliquots of the reaction were removed for HPLC assay over a
in an ELISA assay.45 The data are shown in Table 1 including
4 h period in order to assess conversion. The HPLC traces in
the IC50 values of some reference peptides.5,6 The ELISA data
Fig. 1B show the time-dependent appearance of the product at
indicate that for this series the multimers have a progressively
tR = 8.2 min, a peak which was identified as FDEA-PEG-(RGD)2
higher integrin affinity (lower IC50) as the number of cRGD
31 by LC-MS comparison with the synthetic reference pre-
ligands increases e.g., monomeric 8 (IC50 = 74 nM), dimeric 31
viously prepared. The mass spectra of the HPLC peaks corres-
(IC50 = 60 nM) and tetrameric 33 (IC50 = 24 nM). The IC50
values measured for the multimeric constructs are also lower
than the parent azido-peptide 29 (IC50 = 90 nM) and the multi-
meric cRGD compounds have significantly lower IC50 values
compared with reference compounds containing linear RGD-
containing peptides.
Going from one to two cRGD motifs resulted in a marginal
increase in binding affinity however progression to a tetramer
ponding to the multimeric 5′-chloro substrates and 5′-fluoro
products are shown in Fig. 1C and D respectively. The peaks
for the product contain multiply charged ions of 31
([M + 2H]2+, [M + 3H]3+, and [M + 4H]4+ respectively), confirm-
ing its identity as FDEA-PEG-(RGD)2 31.
A series of control experiments, where the reaction was con-
ducted in the absence of either enzyme or L-SeMet, did not
result in any conversion of 30 to 31, confirming an enzyme
catalysed process. In the absence of fluoride, the consumption
of 30 was observed, however, no fluorinated product was pro-
showed a significant decrease in IC50 and gave a compound
with the highest affinity of those tested. The linker between
the two or four cRGD units is not so extended and steric con-
duced. This result is consistent with the generation of a
straints will likely prevent simultaneous multivalent binding
SeSAM intermediate, which cannot be converted to product 31
of cRGDs to more than one surface site. Instead the increased
affinity is most likely due to an increased effective molarity at
in the absence of fluoride.
A similar analysis was conducted for the enzymatic reaction
the surface.46 Some multimeric constructs reported in the lit-
with ClDEA-PEG-(RGD)4 32, as illustrated in Fig. 2A. ClDEA-
erature have IC50 values in the 1–100 nM range,24 similar to
PEG-(RGD)4 32 (20 µM) was added to a mixture of L-SeMet
this study although a direct comparison of different assays
(75 µM) and potassium fluoride (50 mM), in a total volume of
and dissimilar peptide motifs make accurate comparison
1 mL. The reaction was initiated by the addition of the fluori-
difficult.
nase (1 mg mL−1) and monitored over 4 h. The resultant HPLC
profile (Fig. 2B) indicated that 32 was consumed, while a new
2.4. Enzymatic radiolabelling of multimers with fluorine-18
peak, at tR = 9.2 min, steadily increased throughout the assay.
The retention time of this new peak was identical to that of
synthetic FDEA-PEG-(RGD)4 33, and its identity was further
confirmed by LC-MS.
The ability of the fluorinase to employ these large peptide
assemblies of 10 or 20 amino acids as substrates opened up
prospects of labelling these peptides with [18F]fluoride for
PET. Radiolabelling trials of ClDEA-PEG-(RGD)2 30 (0.04 mM)
was initially investigated using similar reaction conditions to
the assays described with [19F]fluoride, but with an aqueous
[18F]fluoride solution generated on the cyclotron. The enzyme
Mass spectra the tetrameric substrate and product are
shown in Fig. 2C and D respectively. In the sample analysed at
t = 0 h (Fig. 2C), peaks corresponding to multiply charged
species of ClDEA-PEG-(RGD)4 32 are observed at m/z = 858.9
([M + 4H]4+) and 687.2 ([M + 5H]5+). In the sample analysed
after an extended 16 h reaction (Fig. 2D), peaks at m/z = 854.8
([M + 4H]4+) and 683.8 ([M + 5H]5+) are clearly observed. The
presence of these ions is consistent with the formation of
FDEA-PEG-(RGD)4 33 as the product of enzymatic
transhalogenation.
Tolerance of the fluorinase to these large multimeric
peptides suggests that the C-2 position of a chlorinated
nucleoside represents a site for the attachment of a diverse
range of peptide cargos for use in enzymatic fluorination.
The constructs explored here consist of 10 and 20 amino
acids, and they were efficiently fluorinated under neutral
aqueous conditions. With the knowledge that the multimeric
constructs were suitable substrates for fluorinase catalysed
transhalogenation, it was of interest to evaluate the
effect the multiple RGD motifs on binding affinity to αVβ3
integrin.
Table 1 IC50 values of selected RGD-containing peptides measured as
the ability to compete with c(RGDfK[PEG-PEG-biotin]) for binding to
immobilised αVβ3 integrin. Results are the average standard error (s.e.)
from three independent experiments, each performed in triplicate. Q is
the normalised affinity of the peptides, referenced to GRGDSPK
Compound
IC50 s.e./nM
Q
RGDa
8560 2240
2130 410
90 10
330 30
74 16
60 13
24 18
4.019
1.000
0.042
0.155
0.033
0.027
0.011
GRGDSPKa
c(RGDfK[N3]) 29a
FDEA-cRGDa
FDEA-TEG-cRGD 8a
FDEA-PEG-(cRGD)2 31
FDEA-PEG-(cRGD)4 33
a Affinities for these constructs were previously reported, and were
measured using an identical assay.
This journal is © The Royal Society of Chemistry 2016
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
concentration was increased from 1 mg mL−1 (29 µM) to [18F]-31 were observed by radio-HPLC. Optimisation trials were
10 mg mL−1 (290 µM), as it had previously been shown that carried out with ClDA 4 to [18F]FDA [18F]-2 as a substrate, and
high enzyme concentrations improves conversions in radio- the data suggest that an important factor for improving radio-
labelling experiments.47 Notably [18F]fluoride ion concentrations chemical incorporations is substrate concentration. A RCI of
are necessarily very low in the pico-molar range. Radiochemi- greater than 90% was observed when ClDA 4 was incubated at
cal incorporations (RCI) were found to be low under these con- a concentration of 0.6 mM, compared with a 10% RCI at
ditions, and only traces of the product, [18F]FDEA-PEG-(RGD)2 0.04 mM (see ESI†).
Fig. 3 A. Reaction scheme and resultant isocratic HPLC radioactivity trace of an experiment where ClDEA-PEG-(RGD)2 30 (0.3 mM) was freeze dried with
the fluorinase (20 mg mL−1) prior to reaction. [18F]FDEA-PEG-(RGD)2 18F]-31 was observed at tR = 13.7 min, showing 10% RCI. B. Reaction scheme and
[
resultant gradient HPLC radioactivity trace of an experiment where ClDEA-PEG-(RGD)4 32 (0.3 mM) was freeze dried with the fluorinase (20 mg mL−1
)
prior to reaction. [18F]FDEA-PEG-(RGD)4 18F]-33 was observed at tR = 10.1 min, showing 5% RCI. An additional peak was also observed at tR = 5.2 min,
[
identified as [18F]FDA [18F]-2, produced due to the presence of residual SAM 1 co-purified with the fluorinase.48 The peak is also present for ClDEA-PEG-
(RGD)2 30 labelling experiments, but is not observed in A due to the HPLC conditions utilised (see ESI† for further discussion).
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Organic & Biomolecular Chemistry
Paper
By contrast, ClDEA-PEG-(RGD)2 30 and ClDEA-PEG-(RGD)4
32 were found to be poorly soluble in water, and concen-
Acknowledgements
trations of 0.6 mM were not achievable in buffer. The upper We thank EPSRC and the Scottish Imaging Network (SINAPSE)
concentration limit for the peptides was found to be ∼0.3 mM. for grants. DO’H thanks the Royal Society for a Wolfson
ClDEA-PEG-(RGD)2 30 and ClDEA-PEG-(RGD)4 32 were incu- Research Merit Award and ST is grateful to the John and
bated in separate experiments with L-SeMet (0.08 mM) and Kathleen Watson Scholarship for financial support. We are
[18F]fluoride in water (138 µL–160 µL, 15.2–35.4 MBq). Some grateful to Dr Catherine Botting and Dr Sally Shirran of the
precipitation of the substrate was observed upon addition of St Andrews Mass Spectrometry Service for MALDI-MS acqui-
aqueous [18F]fluoride. The reaction mixtures were incubated at sitions. We also thank Dr Sally Pimlott of the University of
37 °C for 30 minutes, and then the enzyme was heat denatured Glasgow for the use of radiochemistry facilities.
and the mixture centrifuged. A sample of the clarified super-
natant from each reaction was analysed by HPLC, and the
resultant HPLC radio-chromatograms are shown in Fig. 3.
Radio-HPLC of the two reaction mixtures (Fig. 3A and B)
References
revealed the presence of the respective [18F]-products, along
with a broad peak (tR = 2.2 min), identified as [18F]fluoride.49
The peak at tR = 13.7 min in Fig. 3A was identified as [18F]-31
by spiking and comparison of the retention time to that of a
[19F]-reference sample. Similarly the peak at tR = 10.1 min in
Fig. 3B was identified as [18F]-33 relative to a [19F]-reference.
[18F]-FDA [18F]-2 was also observed in this experiment, which
must arise as a consequence of low levels of L-AdoMet (SAM)
bound to the enzyme and residual from the purification
process. This outcome only arose in the reactions of low con-
version. Reactions were also conducted with the addition of
DMSO 1% (v/v) as a co-solvent, however this did not improve
the radiochemical conversions and extending the reaction
time or increasing the concentration of L-SeMet had no signifi-
cant effect. Therefore despite the efficient conversions
observed in the “cold” experiments at high [19F]fluoride ion
concentrations, low [18F]fluoride concentrations require to be
compensated for by high substrate concentrations, and this
limited the efficiency of the radiolabelled experiments.
1 D. O’Hagan, C. Schaffrath, S. L. Cobb and
J. T. G. Hamilton, Nature, 2002, 416, 279.
2 H. Deng, L. Ma, N. Bandaranayaka, Z. Qin, G. Mann,
K. Kyeremeh, Y. Yu, T. Shepherd, J. H. Naismith and
D. O’Hagan, ChemBioChem, 2014, 15, 364–368.
3 M. Onega, J. Domarkas, H. Deng, L. F. Schweiger,
T. A. D. Smith, A. E. Welch, C. Plisson, A. D. Gee and
D. O’Hagan, Chem. Commun., 2010, 46, 139–141.
4 S. Dall’Angelo, N. Bandaranayaka, A. D. Windhorst,
D. J. Vugts, D. van der Born, M. Onega, L. F. Schweiger,
M. Zanda and D. O’Hagan, Nucl. Med. Biol., 2013, 40, 464–
470.
5 S. Thompson, Q. Zhang, M. Onega, S. McMahon,
I. Fleming, S. Ashworth, J. H. Naismith, J. Passchier and
D. O’Hagan, Angew. Chem., Int. Ed., 2014, 53, 8913–8918.
6 S. Thompson, M. Onega, S. Ashworth, I. N. Fleming,
J. Passchier and D. O’Hagan, Chem. Commun., 2015, 51,
13542–13545.
7 H. Deng, S. L. Cobb, A. R. McEwan, R. P. McGlinchey,
J. H. Naismith, D. O’Hagan, D. A. Robinson and
J. B. Spencer, Angew. Chem., Int. Ed., 2006, 45, 759–762.
8 P. W. Miller, N. J. Long, R. Vilar and A. D. Gee, Angew.
Chem., Int. Ed., 2008, 47, 8998–9033.
3. Conclusions
Enzymatic transhalogenations using the fluorinase was
9 S. Richter and F. Wuest, Molecules, 2014, 19, 20536–20556.
explored with relatively complex di- and tetra-cRGD constructs. 10 P. W. Miller, N. J. Long, R. Vilar and A. D. Gee, Angew.
This exploited the previously identified substrate tolerance at Chem., Int. Ed., 2008, 47, 8998–9033.
C-2 of the adenine base in the substrate motif. The multimeric 11 R. Ting, M. J. Adam, T. J. Ruth and D. M. Perrin, J. Am.
assemblies, ClDEA-PEG-(RGD)2 30 and ClDEA-PEG-(RGD)4 32, Chem. Soc., 2005, 127, 13094–13095.
were assessed as fluorinase substrates and both were found to 12 C. A. D’Souza, W. J. McBride, R. M. Sharkey, L. J. Todaro
undergo efficient enzyme-catalysed transhalogenations. The
and D. M. Goldenberg, Bioconjugate Chem., 2011, 22, 1793–
multimers also showed higher affinities to immobilised αVβ3
1803.
integrins, which opened up the prospect of using them as 13 R. Schirrmacher, G. Bradtmöller, E. Schirrmacher,
cancer imaging agents in [18F]-radiolabelled form. This proved
only partially successful due to the relatively poor solubility of
the multimeric constructs and low [18F]-fluoride concen-
O. Thews, J. Tillmanns, T. Siessmeier, H. G. Buchholz,
P. Bartenstein, B. Wängler, C. M. Niemeyer and
K. Jurkschat, Angew. Chem., Int. Ed., 2006, 45, 6047–6050.
trations. It follows that the development of this approach to 14 W. Wan, N. Guo, D. Pan, C. Yu, Y. Weng, S. Luo, H. Ding,
larger peptides or proteins will depend on using more highly
Y. Xu, L. Wang, L. Lang, Q. Xie, M. Yang and X. Chen,
soluble peptide substrates in buffer. The peptides used here
J. Nucl. Med., 2013, 54, 691–698.
are ‘click’ derived with low solubility, however the approach 15 C. Yu, D. Pan, B. Mi, Y. Xu, L. Lang, G. Niu, M. Yang,
should prove more useful with natural peptides and antibodies
which have evolved high water solubility.
W. Wan and X. Chen, Eur. J. Nucl. Med. Mol. Imaging, 2015,
42, 2021–2028.
This journal is © The Royal Society of Chemistry 2016
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
16 A. Goel, J. Baranowska-Kortylewicz, S. H. Hinrichs, 33 C. Wängler, S. Maschauer, O. Prante, M. Schäfer,
J. Wisecarver, G. Pavlinkova, S. Augustine, D. Colcher,
B. J. Booth and S. K. Batra, J. Nucl. Med., 2001, 42, 1519–
1527.
R. Schirrmacher, P. Bartenstein, M. Eisenhut and
B. Wängler, ChemBioChem, 2010, 11, 2168–2181.
34 I. Dijkgraaf, A. Y. Rijnders, A. Soede, A. C. Dechesne,
G. W. van Esse, A. J. Brouwer, F. H. M. Corstens,
O. C. Boerman, D. T. S. Rijkers and R. M. J. Liskamp, Org.
Biomol. Chem., 2007, 5, 935–944.
17 S. Liu, D. S. Edwards, M. C. Ziegler, A. R. Harris,
S. J. Hemingway and J. A. Barrett, Bioconjugate Chem., 2001,
12, 624–629.
18 M. Mammen, S.-K. Choi and G. M. Whitesides, Angew. 35 H. Li, H. Zhou, S. Krieger, J. J. Parry, J. J. Whittenberg,
Chem., Int. Ed., 1998, 37, 2754–2794.
A. V. Desai, B. E. Rogers, P. J. a. Kenis and D. E. Reichert,
19 S. Liu, Mol. Pharm., 2006, 3, 472–487.
Bioconjugate Chem., 2014, 25, 761–772.
20 C. Fasting, C. A. Schalley, M. Weber, O. Seitz, S. Hecht, 36 D. B. Dess and J. C. Martin, J. Org. Chem., 1983, 48, 4155–
B. Koksch, J. Dernedde, C. Graf, E. W. Knapp and R. Haag,
Angew. Chem., Int. Ed., 2012, 51, 10472–10498.
21 C. J. Avraamides, B. Garmy-Susini and J. A. Varner, Nat.
Rev. Cancer, 2008, 8, 604–617.
22 R. Haubner, R. Gratias, B. Diefenbach and S. Goodman,
J. Am. Chem. Soc., 1996, 118, 7461–7472.
4156.
37 H. Staudinger and J. Meyer, Helv. Chim. Acta, 1919, 2, 635–
646.
38 A. F. Abdel-Magid, K. G. Carson, B. D. Harris,
C. A. Maryanoff and R. D. Shah, J. Org. Chem., 1996, 61,
3849–3862.
23 S. Liu, Bioconjugate Chem., 2009, 20, 2199–2213.
24 S. Liu, Bioconjugate Chem., 2015, 26, 1413–1438.
39 M. J. Robins, F. Hansske, S. F. Wnuk and T. Kanai,
Can. J. Chem., 1991, 69, 1468–1474.
25 G. Thumshirn, U. Hersel, S. L. Goodman and H. Kessler, 40 V. Iaroshenko, D. Sevenard, A. Kotljarov, D. Volochnyuk,
Chem. – Eur. J., 2003, 9, 2717–2725. A. Tolmachev and V. Sosnovskikh, Synthesis, 2009, 731–740.
26 T. Poethko, M. Schottelius, G. Thumshirn, U. Hersel, 41 K. Sonogashira, Y. Tohda and N. Hagihara, Tetrahedron
M. Herz, G. Henriksen, H. Kessler, M. Schwaiger and
H.-J. Wester, J. Nucl. Med., 2004, 45, 892–902.
27 C. Hultsch, M. Berndt, R. Bergmann and F. Wuest, Appl.
Radiat. Isot., 2007, 65, 818–826.
Lett., 1975, 16, 4467–4470.
42 R. Chinchilla and C. Nájera, Chem. Soc. Rev., 2011, 40,
5084–5121.
43 M. Schilz and H. Plenio, J. Org. Chem., 2012, 77, 2798–2807.
28 S. Liu, Z. Liu, K. Chen, Y. Yan, P. Watzlowik, H. J. Wester, 44 T. R. Chan, R. Hilgraf, K. B. Sharpless and V. V. Fokin, Org.
F. T. Chin and X. Chen, Mol. Imaging Biol., 2010, 12, 530–
538.
29 W. Cheng, Z. Wu, S. Liang, H. Fu, S. Wu, Y. Tang, Z. Ye and
H. Wang, PLoS One, 2014, 9, e100521.
Lett., 2004, 6, 2853–2855.
45 M. Piras, I. Fleming, W. Harrison and M. Zanda, Synlett,
2012, 2899–2902.
46 A. J. Beer, H. Kessler, H.-J. Wester and M. Schwaiger,
Theranostics, 2011, 1, 48–57.
30 J. Guo, N. Guo, L. Lang, D. O. Kiesewetter, Q. Xie, Q. Li,
H. S. Eden, G. Niu and X. Chen, J. Nucl. Med., 2014, 55, 47 H. Deng, S. L. Cobb, A. D. Gee, A. Lockhart, L. Martarello,
154–160.
R. P. McGlinchey, D. O’Hagan and M. Onega, Chem.
Commun., 2006, 652–654.
31 W. J. McBride, C. A. D’Souza, H. Karacay, R. M. Sharkey
and D. M. Goldenberg, Bioconjugate Chem., 2012, 23, 538– 48 C. Dong, F. Huang, H. Deng, C. Schaffrath, J. B. Spencer,
547. D. O’Hagan and J. H. Naismith, Nature, 2004, 427, 561–565.
32 J. A. F. Joosten, N. T. H. Tholen, F. Ait El Maate, 49 D. Ory, J. Van den Brande, T. de Groot, K. Serdons, M. Bex,
A. J. Brouwer, G. W. van Esse, D. T. S. Rijkers,
R. M. J. Liskamp and R. J. Pieters, Eur. J. Org. Chem., 2005,
3182–3185.
L. Declercq, F. Cleeren, M. Ooms, K. Van Laere,
A. Verbruggen and G. Bormans, J. Pharm. Biomed. Anal.,
2015, 111, 209–214.
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2016