Enzymatic Fluorination of Biotin and Tetrazine Conjugates for Pretargeting Approaches to Positron Emission Tomography Imaging

Article abstract of DOI:10.1002/cbic.201800234

The use of radiolabelled antibodies and antibody-derived recombinant constructs has shown promise for both imaging and therapeutic use. In this context, the biotin–avidin/streptavidin pairing, along with the inverse-electron-demand Diels–Alder (iEDDA) rea

Full text of DOI:10.1002/cbic.201800234

Accepted Article
Title: Enzymatic fluorination of biotin and tetrazine conjugates for pre-
targeting approaches to PET imaging
Authors: Phillip T Lowe, Sergio Dall'Angelo, Andrew Devine, Matteo
Zanda, and David O'Hagan
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemBioChem 10.1002/cbic.201800234
Link to VoR: http://dx.doi.org/10.1002/cbic.201800234
A Journal of
www.chembiochem.org

ChemBioChem
10.1002/cbic.201800234
Enzymatic fluorination of biotin and tetrazine conjugates for
pre-targeting approaches to PET imaging.
†
a
† b
a
b
a
Phillip T. Lowe , Sergio Dall'Angelo , Andrew Devine, Matteo Zanda, and David O'Hagan*
[
a] Dr. P. T. Lowe, A. Devine, Prof. D. O’Hagan
School of Chemistry and Centre for Biomolecular Sciences
University of St. Andrews, North Haugh, St. Andrews, Fife, KY16 9ST (UK)
E-mail: do1@st-andrews.ac.uk
[
b] Dr. S. Dall’Angelo, Prof. M. Zanda
John Mallard Scottish PET Centre
School of Medicine, Medical Sciences and Nutrition
University of Aberdeen, Foresterhill, Aberdeen, AB25 2ZD (UK)
E-mail: m.zanda@abdn.ac.uk
[
†] These authors contributed equally to this work.
The use of radiolabelled antibodies and antibody-derived recombinant constructs have shown
promise for both imaging and therapeutic use. In this context, the biotin-avidin/streptavidin pairing
along with the inverse electron demand Diels–Alder reaction (IEDDA) have found application in pre-
targeting approaches for positron emission tomography (PET). This study reports the fluorinase
enzyme mediated transhalogenation (ClDA substrates to FDA products) of two antibody pre-targeting
18
tools, a FDA-PEG-tetrazine and a [ F]FDA-PEG-biotin and each is assessed for their compatibility
towards IEDDA ligation to trans-cyclooctene or affinity to avidin. A protocol was developed to avoid
18
radiolytically promoted oxidation of biotin during the synthesis of [ F]FDA-PEG-biotin. The study adds
to the repertoire of conjugates for use in fluorinase catalysed radio-synthesis for PET and shows that
the fluorinase will accept a wide range of ClDA substrates tethered at C-2 of the adenine ring with a
18
pegylated cargo. The method is exceptional as the nucleophilic reaction with [ F]fluoride takes place
in water at neutral pH and at ambient temperature.
This article is protected by copyright. All rights reserved.

ChemBioChem
10.1002/cbic.201800234
Introduction
Positron emission tomography (PET) has emerged as a valuable tool in clinical imaging due to its high
sensitivity and complementarity to other in vivo clinical imaging modalities, such as computerised
[
1] 18
tomography (CT) and magnetic resonance imaging (MRI). [ F]Fluoride is the most commonly used
radionuclide for PET imaging as it possesses a relatively long half-life (t1/2 = 109.8 min) for a clinical
positron emitter and image resolution is enhanced as it has a short positron range (2.3 mm in water).
A recent drive to improve imaging performance by PET has led to a shift from the direct labelling of
traditional low molecular-weight and rapidly circulating (hours) radiotracers, which often suffer non-
specific binding and poor localisation in vivo, to the use of large but very high affinity proteins such as
[
2]
monoclonal antibodies or their fragments, which circulate much more slowly (days). Such proteins
cannot be usefully labelled by fluorine-18 at point of injection as the isotope half-life is inconsistent
with the pre-incubation time required for their location at the target. Instead, a pre-targeted approach
is required where the antibody is administered in its unlabelled form, but with a reactive motif
18
attached, and then several hours to days later, a low molecular weight F-labelled reactive partner is
administered which circulates rapidly and covalently modifies the in situ antibody. Such strategies
[
3]
have begun to exploit bioorthogonal click reactions and make use, for example, of strain promoted
alkyne–azide cycloaddition (SPAAC) or inverse electron demand Diels–Alder reactions (IEDDAs),
[
4]
without the need for catalysis. IEDDA based methods in particular involve complimentary tetrazine
and trans-cyclooctene (TCO) motifs which combine remarkably rapidly (reaction rates up to k 2000
2
-
1
-1
[4d]
M s in 9:1 methanol/water), allowing for protein modification at low radiotracer concentrations.
Scheme 1. Schematic diagrams of pre-targeting approaches to PET. A) The inverse electron demand Diels–Alder
(IEDDA) bioorthogonal click reaction. B) The (+)-biotin/(strep)avidin affinity pairing.
In this context, the intitial approach has been to fluorinate the TCO motif as direct nucleophilic
fluorination of the reactive tetrazine motif is challenging due to decomposition.[ Thus efficient
5]
This article is protected by copyright. All rights reserved.

ChemBioChem
10.1002/cbic.201800234
[
6]
18
methods for radiolabelling tetrazines with fluorine-18 are desirable. In this regard an Al[ F]NOTA-
labelled tetrazine tracer was successfully utilised in a pre-targeting PET study, expressing BxPC3
pancreatic cancer xenografts in mice.[ There remains room to explore other F-labelled tetrazine
strategies and in this study we have investigated enzymatic fluorination in water/buffer under
ambient conditions.
4c]
18
Similarly, pre-targeting strategies have been designed around the (+)-biotin/(strep)avidin affinity
[
7]
-15
pairing. The interaction is among the strongest non-covalent interactions (K = 1 x 10 M) known
d
and is unaffected by extremes of pH and temperature. Initial approaches to multistep imaging
involved the use of unlabelled antibodies conjugated to (+)-biotin, followed by the administration of
radiolabelled (strept)avidin.[ Alternatively, a more efficient pre-targeting approach involves the
administration of large (strept)avidin-antibody conjugates, followed by a rapidly circulating, low
8]
18
[9]
molecular weight [ F]biotin derivative. There has been concern when employing this strategy
regarding the role of endogenous biotin competing with the radiolabelled chaser in vivo, though this
[
10]
has been successfully addressed in mice models through a biotin-deficient diet.
Radiolabelling of (+)-biotin has generally been achieved through the pre-synthesis and then coupling
1
8
18
[7b]
18
[10b]
of F-labelled prosthetic groups, typically [ F]4-fluorobenzylamine , p-[ F]fluorobenzoic acid
1
8
[11]
and [ F]FDG . The approach involves several steps and requires efficient fluorination strategies.
Direct labelling approaches, such as the nucleophilic fluorination of a (+)-biotin-mesylate precursor
are more appealing.[ We have been exploring an even more direct approach involving enzymatic
12]
18
fluorination, which can be accomplished without drying the [ F]fluoride which is generated in a
18
cyclotron in [ O]water, prior to nucleophilic chemistry. Such an approach requires minimal
manipulation of the radioisotope. A similarly direct approach has been explored with boroaryl
functionalized (+)-biotin,[ however in order to avoid deboronylation during the radiolabelling
13]
protocol,
a
low
pH
is
required.
This article is protected by copyright. All rights reserved.

ChemBioChem
10.1002/cbic.201800234
Scheme 2. A) The native reaction catalyzed by the fluorinase enzyme. B) Fluorinase mediated transhalogenation
reaction of C-2 modified substrates, 6 and 7 exemplify previously achieved alkyne/azide "click"[ and barbas
linker/thiol-ligation[15] bioconjugation strategies.
14]
The fluorinase enzyme catalyses the conversion of S-adenosyl-L-methionine (AdoMet) and fluoride
[
16]
ion to generate 5'-fluoro-5'-deoxyadenosine (5'-FDA, Scheme 2). It also catalyses the nucleophilic
displacement of chloride by L-methonine in the reverse direction to generate AdoMet,[17] and if
fluoride is present in this reaction 5'-FDA is formed in a sequential step, such that an overall
transhalogenation reaction occurs. Moreover, the replacement of AdoMet with AdoSeMet increases
the overall transhalogenation efficiency, as selenium is a better nucleophile and leaving group than
sulfur (Scheme 2). The fluorinase displays a high degree of substrate specificity, however it can
tolerate an acetylene moiety at the C2 position on the adenine base of 5'-ClDA, additionally a
pegylated cargo can also be tethered to the terminus of this acetylene. In this way the fluorinase will
18
[14]
catalyse ‘last-step’ aqueous [ F]fluorination of 5'-ClDA-PEG-peptides. To date, methodologies to
This article is protected by copyright. All rights reserved.

ChemBioChem
10.1002/cbic.201800234
permit bioconjugation of RGD peptides to the fluorinase binding 5'-ClDA moiety either through ‘click’
[
14, 18]
based chemistry (via azide functionalized peptides)
or utilising a modified Barbas bioconjugation
strategy (via cysteine containing peptides)[ have been developed (Scheme 2). Recently, the
15]
fluorinase enzyme has also become the target of engineering via directed evolution, in an effort to
[
19]
achieve higher radiochemical conversions (RCC) and broaden the structural diversity of substrates.
Scheme 3. Design of fluorinase binding ClDA-PEG-acid 8 and ClDA-PEG-amine 9 constructs for amide bond
ligation to biologically relevant molecules.
In order to demonstrate a wider array of conjugation methods (employing amide linkage via ClDA-
PEG-acid 8 and ClDA-PEG-amine 9, Scheme 3) compatible with fluorinase labelling of higher molecular-
weight proteins, we now report the fluorinase mediated transhalogenation of ClDA-PEG-tetrazine 16
to FDA-PEG-tetrazine 17 demonstrating compatibility with the tetrazine ring under ambient, aqueous
conditions. Similarly ClDA-PEG-biotin 26 was used in a proof of principle study to deliver a biotin
18
radiotracer, [ F]27, as a tool for radiolabelling pre-targeted (strept)avidin conjugated antibodies.
Results and Discussion
Synthesis of ClDA-PEG-acid 8 and ClDA-PEG-tetrazine 16
A three step approach was employed for the synthesis of both ClDA-PEG-tetrazine 16 and ClDA-PEG-
biotin 26.[
14, 18b]
This involved the synthesis of a fluorinase compatible 5'-chloro-5'-deoxyadenosine
attached to an amine (9) or carboxylic acid (8) functionality through a PEG spacer. Amide coupling to
either benzylamino tetrazine 15 or the carboxylate of (+)-biotin 25, generated tetrazine 16 (Scheme
4
) or ClDA-PEG-biotin 26 (Scheme 5) constructs respectively. Briefly, 5'-chlorodeoxy-2-iodo-adenosine
1
3, a key intermediate in these syntheses, was prepared from guanosine using a previously established
protocol.[ The synthesis of 16 required that the pegylated linker was functionalized with an alkyne
terminus. This was accomplished by first treating tetraethylene glycol 10 with propargyl bromide to
provide the monoalkylated PEG derivative 11, which, after purification, was treated with tert-butyl
bromoacetate to afford 12. This PEG linker serves two purposes; firstly, it aids in the aqueous solubility
of the conjugate, an essential requirement for fluorinase catalysis. Secondly, the linker serves to
project any functionalized moiety from the surface of the fluorinase and into the bulk solvent.
18b]
This article is protected by copyright. All rights reserved.

ChemBioChem
10.1002/cbic.201800234
Scheme 4. Synthesis of ClDA-PEG-acid 8, ClDA-PEG-tetrazine 16 and FDA-PEG-tetrazine 17. Reagents and
conditions: a) NaH, propargyl bromide, THF, 0 °C to r.t., overnight; b) NaH, tert-butyl bromoacetate, THF, 0 °C
to r.t., overnight; c) Pd
2
(dba)
3
, Et N, CuI, DMF, 80 °C, 24 h; d) TFA, DCM, 0 °C to r.t., 45 min; e) PyBOP, DIPEA,
3
DMF, r.t., 24 h; f) Fluorinase, L-Se-Met, KF, phosphate buffer (pH 7.8), 37 °C, 72 h.
With both 12 and 13 in hand, our attention turned to further assembly via Sonogashira cross-coupling.
This was accomplished,[ using an excess of alkyne 12 over 5'-chlorodeoxy-2-iodo-adenosine 13 in
20]
the Pd
chromatography and C18 cartridge purification. ClDA-PEG-ester 14 was treated with TFA to generate
as a TFA salt, and this could be used directly in the subsequent coupling reactions, without the need
2
(dba)
3
catalysed reaction. The ClDA-PEG-ester 14 was recovered in high purity after
8
for further purification. Finally, an amide coupling reaction (PyBOP, DIPEA) was employed to attach
carboxylic acid 8 to benzylamino tetrazine 15 (Scheme 4). This afforded the ClDA-PEG-tetrazine 16
conjugate in high purity after preparative HPLC.
Evaluation of ClDA-PEG-tetrazine as a substrate for fluorinase mediated transhalogenation
-
1
Substrate 16 was incubated with the fluorinase (1 mg mL , these relatively high amounts are due to
−
1 [17]
the fluorinase having a slow kcat of 0.03 min )
under previously established conditions in
phosphate buffer at pH 7.8 (see Supporting Information). A time course profile (Figure 1) showed that
1
1
6 was an excellent substrate for enzyme catalysed transhalogenation generating FDA-PEG-tetrazine
7, in ~60% conversion after 4 h, and this offers a new approach to fluorinating tetrazine derivatives.
Importantly, no obvious degradation was observed within the initial hours of these experiments (the
timeframe required for radiolabeling experiments), indicating that ClDA-PEG-tetrazine and its
fluorinated analogue are stable under these conditions. The additional peak observed at 12.5 min is
indicative of the intermediate L-Se-Met adduct (which consistently possesses a greater retention time
to that of its corresponding ClDA and FDA derivatives).
This article is protected by copyright. All rights reserved.

ChemBioChem
10.1002/cbic.201800234
Figure 1. HPLC time course (UV, 254 nm) of the incubation of ClDA-PEG-tetrazine 16, blue (t
R
= 11.9 min), with
the fluorinase, L-Se-Met, KF, phosphate buffer (pH 7.8) at 37 °C. Traces show the formation of FDA-PEG-tetrazine
1
7, purple (t = 10.8 min), and the consumption of ClDA-PEG-tetrazine 16. For full conditions see the
R
Experimental Section and Supporting Information.
ClDA-PEG-tetrazine/dienophile IEDDA ligation
With FDA-PEG-tetrazine 17 in hand, it was explored for its ability to participate in IEDDA-initiated
ligation. For this study (E)-cyclooct-4-enol 18 was chosen as a coupling partner. Thus, a solution (40
μl, 50 mM) of (E)-cyclooct-4-enol in DMSO was added to an aqueous solution of 17, and the mixture
was incubated at room temperature for 2 min. An aliquot was then subjected to HPLC and HRMS
analysis of the new product isolated was consistent with the formation of the fully aromatised ligated
product 19 (Scheme 5).
Scheme 5. IEDDA initiated conjugation between 17 and (E)-cyclooct-4-enol 18 (identity of fully aromatised
ligation product 19 was confirmed by HRMS after isolation by HPLC.
This article is protected by copyright. All rights reserved.

ChemBioChem
10.1002/cbic.201800234
1
8
18
Last step enzymatic F labelling of 16 to [ F]FDA-PEG-tetrazine
18
An enzymatic transhalogenation in the presence of [ F]fluoride was investigated with 16 and its
1
8
18
conversion into [ F]FDA-PEG-tetrazine ([ F]17) was clearly demonstrated against a product
reference. Using the previously established protocol,[ [ F]17 was generated with an approximate
15] 18
1
0% radiochemical conversion (decay uncorrected) after a 30 min incubation with the fluorinase
18
(Figure S46). Isolation of [ F]17 by semi-prep HPLC afforded the radiolabelled product with two minor
radiolabelled impurities (Figure S47). This process requires optimisation for pre-clinical dosing,
however it is clear that the tetrazine moiety is compatible with the fluorinase radiolabelling protocol.
Synthesis of ClDA-PEG-amine 9 and ClDA-PEG-biotin 26
The synthesis of ClDA-PEG-biotin 26 followed a similar strategy to that used for ClDA-PEG-tetrazine
1
6. Firstly, a pegylated linker was required with alkyne and amine termini. For this alcohol 11 was
treated with tosyl chloride and then 20 was subjected to nucleophilic displacement with sodium azide
to generate azide 21. A Staudinger reduction of this azide afforded free amine 22, which was then Boc
protected to generate 23.
Scheme 6. Synthesis of ClDA-PEG-amine 9, ClDA-PEG-biotin 26 and FDA-PEG-biotin 27. Reagents and conditions:
a) NaH, propargyl bromide, THF, 0 °C to r.t., overnight; b) Tosyl chloride, Et
C, 18 h; d) PPh , H O, THF, r.t., 48 h; e) (Boc) O, Et N, MeOH; f) Pd (dba) , Et
DCM, 0 °C to r.t., 45 min; h) PyBOP, DIPEA, DMF, r.t., 24h; i) Fluorinase, L-Se-Met, KF, phosphate buffer (pH 7.8),
7 °C, 48 h.
3
N, THF, r.t., 20 h; c) NaN
3
, DMF, 50
°
3
2
2
3
2
3
3
N, CuI, DMF, 80 °C, 24 h; g) TFA,
3
This article is protected by copyright. All rights reserved.

ChemBioChem
10.1002/cbic.201800234
A Sonogashira cross-coupling was carried out with an excess of alkyne 23 over 5'-chlorodeoxy-2-iodo-
adenosine 13 using the previously described protocol to generate ClDA-PEG-protected amine 24, after
HPLC purification, in high purity. Deprotection of 24 with TFA led to free amine 9 (as a TFA salt) which
was used directly in subsequent coupling reactions, without the need for further purification. As with
18
carboxylic acid 8, amine 9 offers a complementary tool for the fluorinase [ F]-labelling, for
conjugation to the C-termini of bioactive peptides. Amide coupling (PyBOP, DIPEA) allowed
preparation of the ClDA-PEG-biotin conjugate 26 in good yield and high purity after HPLC purification.
Evaluation of ClDA-PEG-biotin as a substrate for fluorinase mediated transhalogenation
18
As a prelude to [ F]-radiolabelling a ‘cold’ transhalogenation reaction with ClDA-PEG-biotin 26 was
-
1
explored with the fluorinase. Accordingly, substrate 26 was incubated with the fluorinase (1 mg mL )
L-Se-Met (75 μM) and KF (50 mM) in phosphate buffer, at pH 7.8 (see Supporting Information). An
analytical time course profile revealed chloride 26 to be an excellent substrate for fluorinase mediated
transhalogenation generating FDA-PEG-biotin 27, with an almost 80% conversion after 4 h. No
reaction was observed in the absence of the fluorinase and the water solubility of 26 circumvented
any requirement to add DMSO. A significant advantage in such reactions, as adding DMSO can
complicate work up by promoting gelling on heat denaturation.
Figure 2. HPLC time course (UV, 254 nm) of the incubation of ClDA-PEG-biotin 26, green (t
fluorinase, L-Se-Met, KF, phosphate buffer (pH 7.8) at 37 °C. Traces show the formation of FDA-PEG-biotin 27,
yellow (t = 8.9 min), and the consumption of ClDA-PEG-biotin 26. For full conditions see the Experimental
Section and Supporting Information.
R
= 9.5 min), with the
R
This article is protected by copyright. All rights reserved.

ChemBioChem
10.1002/cbic.201800234
The transhalogenation of 26 was carried out on a scale sufficient for isolation of an analytical sample
of FDA-PEG-biotin 27 to confirm its identity by NMR. A suitable sample was also required to evaluate
its binding affinity to avidin (see Supporting Information). The reaction was again monitored by HPLC
and was terminated by heat precipitation and centrifugation to remove the enzyme. Purification was
accomplished via semi-prep HPLC to obtain 27 (3mg) in high purity.
FDA-PEG-biotin/avidin Binding Assay
The avidin binding affinity of the newly synthesised FDA-PEG-biotin 27 was determined using a
colorimetric HABA assay. The assay exploits the UV spectral changes that occur upon the binding of
4
'-hydroxyazobenzene-2-carboxylic acid (HABA) to avidin. The HABA/avidin interaction is a relatively
-
6
-15
weak one (K
d
= 5.8 x 10 M versus that of (+)-biotin/Avidin; K = 1 x 10 M), and as such free (+)-biotin,
d
or indeed 27, will readily displace the HABA dye and result in a decrease in absorbance at 500nm
wavelength. FDA-PEG-biotin 27 was found to inhibit the HABA/avidin complex in a dose dependant
manner (Figure 3) and with a similar affinity to that between avidin and (+)-biotin. The PEG spacer and
18
the FDA moiety of the construct do not impede the binding interaction of 27 with avidin, thus [ F]27
has good potential as a candidate tracer for streptavidin-conjugated antibody PET studies.
100
90
80
70
60
50
40
30
20
10
0
0
50
100
150
200
Competitor
Figure 3. Inhibition of the HABA/avidin complex by (+)-biotin (maroon squares) and FDA-PEG-biotin 27 (blue
circles). Data are representative of three different experiments.
1
8
Last step enzymatic F labelling of biotin
18
An enzymatic transhalogenation was performed with 26 in the presence of [ F]fluoride, and its
18
18
18
conversion into [ F]FDA-PEG-biotin ([ F]27) was monitored. For hot labelling an [ O]aqueous,
This article is protected by copyright. All rights reserved.

ChemBioChem
10.1002/cbic.201800234
1
8
18
solution of [ F]fluoride at GBq levels is generated and then an aliquot (MBq of [ F]fluoride) is added
to the biotransformation solution in buffer at pH 7.8 (see Experimental Section). The concentrations
are such that these fluorinase-catalyzed radiochemical reactions are no longer catalytic, as the enzyme
18
is in the micromolar range with [ F]fluoride ion in the pico-molar range. In this manner the
stoichiometry is dramatically reversed relative to the cold transhalogenation reactions. During these
experiments [ F]27 was produced with a 30-40% radiochemical conversion after only 30 min
incubation with the fluorinase. Figure 4 shows the radiochemical HPLC trace of the supernatant of a
reaction mixture at this time point.
18
18
Figure 4. (A) Reaction scheme of the fluorinase catalysed transhalogenation of 26 to [ F]FDA-PEG-biotin
1
8
18
(
[ F]27). (B) HPLC radioactivity trace of a fluorinase catalysed conversion of ClDA-PEG-biotin 26 to [ F]27 in the
presence of 18F , L-Se-Met in phosphate buffer (pH 7.8), at 37 °C after 30 min incubation. [ F]27 was observed
-
18
at t = 9.7 min, showing 40% radiochemical conversion during this experiment.
R
After a 30 min incubation,[15] the enzyme was cleanly precipitated by heat denaturation and the
solution removed after centrifugation, prior to purification by HPLC. During semi-prep HPLC the eluted
1
8
peak corresponding to [ F]27 was collected (Figure 4, t
R
= 9.7 min), diluted with water (50 mL) and
18
loaded onto a C18 reverse phase cartridge, which was then washed with water (20 mL) and [ F]27
18
eluted with ethanol. The purity of [ F]27 was determined by analytical radio-HPLC. This revealed the
existence of an additional radioactive peak at t = 4.3 min (Figure 5). A typical procedure from
F]fluoride (457 MBq) to the elution of [ F]26 (59.7 MBq) took 1 h 19 min. After semi-prep HPLC and
R
18
18
[
18
C18 cartridge purification, [ F]27 was recovered with a radiochemical yield of 13% (decay
uncorrected) and a radiochemical purity of 90%.
This article is protected by copyright. All rights reserved.

ChemBioChem
10.1002/cbic.201800234
1
8
18
Figure 5. Analytical HPLC radio trace of [ F]FDA-PEG-biotin ([ F]27) acquired using standard purification
1
8
18
protocols (red) and optimized purification protocols (black). Structures of [ F]FDA-PEG-biotin ([ F]27, t
min) and [ F]FDA-PEG-biotin sulfoxide ([ F]28, t
R
= 5.7
18
18
R
= 4.3 min) are shown. For full conditions see the Experimental
Section and Supporting Information.
1
8
Optimisation of [ F]FDA-PEG-biotin purification protocol
The chemical identity of the impurity (10%, t
R
= 4.3 min, Figure 5) in the analytical radio-HPLC trace
was elucidated by mass spectrometry analysis. An equivalent peak could be observed in
nonradioactive storage of a solution of 26 in MeCN/H
2
O at room temperature where the conversion
+
to product 28 is slow (3% at 48 h). Isolation by semi-prep HPLC and analysis by LC-MS found a [M+H]
mass of 741 amu, consistent with FDA-PEG-biotin sulfoxide (28) (see Figure 5). Indeed, previous
studies concerning the radiolabelling of biotin observed a similar outcome.[21] Furthermore, biotin
[
22]
sulfoxide is a known product in mammalian catabolism, and is observed in both urine and plasma.
The sulfoxide is also known to bind avidin,[ and thus ([ F]28) low levels are unlikely to negatively
affect PET studies, however we sought to exclude it from the biotransformation. Studies focused on
the radiolytic decomposition of concentrated radiopharmaceuticals[24] indicate that reactive oxygen
species (hydroxyl radicals and/or hydrogen peroxide) can be detrimental, and such species could
23]
18
18
18
clearly account for the sulfur oxidation of [ F]28. In an effort to minimize the formation of [ F]28, the
fluorinase mediated radiolabelling was repeated however in this instance making use of degassed
eluting solvents and established anti-oxidant stabilisers. Approximately the same levels of
radioactivity (~70 MBq) was collected by semi-prep HPLC into a vial containing 10 mg/mL ascorbate (a
known radical scavenger which is amenable to human use) in water solution (45 mL, degassed). This
solution was loaded directly onto the C18 cartridge which had been pre-equilibrated with the same
18
18
degassed solution and [ F]27 was then eluted with ethanol (0.4 mL). HPLC analysis of the [ F]27
18
revealed it to be free of [ F]28, and with a radiochemical purity of >99% (see Figure 5). Furthermore,
18
the stability of [ F]27 in the ethanol fraction (left open to the atmosphere) was monitored over time
by using formulations of both water and aqueous ascorbate (10 mg/mL) solution. This did not reveal
18
any appreciable oxidation of [ F]27. Under these optimised conditions there was no observable
18
degradation of either [ F]27 or its chlorinated precursor. Such stability affords the prospect of the
This article is protected by copyright. All rights reserved.

ChemBioChem
10.1002/cbic.201800234
development of an automated recycling protocol, potentially incorporating immobilised enzyme (to
18
circumvent the need for centrifugation), to be used in the synthesis of [ F]27 to allow for the
18
formulation of higher doses. This fluorinase mediated F-labelling methodology provides a novel
1
8
18
route into [ F]biotin in the form of [ F]27, and demonstrates a direct ‘last step’ labelling approach
18
through the formation of a chemically stable C- F bond from a chlorinated precursor. Furthermore,
after optimization of the purification protocol with the use of non-toxic radical scavengers, a
radiochemically pure sample of [ F]27 was achieved.
18
[18
F]FDA-PEG-biotin streptavidin binding experiments
18
To confirm the utility of [ F]27 as a candidate PET tracer for clinical use an assessment of its the
18
streptavidin binding affinity was performed. Accordingly [ F]27 was incubated in PBS buffer (pH 7.8)
in the presence and absence of streptavidin (see Supporting Information). Thin layer chromatography
was then conducted on both samples (+/- streptavidin) which were analysed using a radio-TLC scanner
18
to determine the percentage of streptavidin bound [ F]27. The radio-traces obtained from samples
1
8
[
F]27 incubated without streptavidin revealed a broad peak with a R/F value of 0.39. Upon
18
incubation with streptavidin [ F]27 is almost completely retained (above 96%) at the origin of the TLC
plate (R/F value of 0.02, see Supporting Information) indicating its almost complete complexation with
streptavidin. This finding is in accordance with the high affinity binding of non radioactive FDA-PEG-
Biotin 27 and avidin.
Conclusions
18
A FDA-PEG-tetrazine construct 17 and a radiolabelled labelled (+)-biotin derivative [ F]27 were
generated from their ClDA-chlorinated precursors 16 and 26 respectively. In both cases each was
shown to efficiently couple to models of their complementary partners anticipated for in vivo studies.
18
As part of the study a protocol to avoid radiolytically triggered oxidation of [ F]27 was developed.
This adds to the repertoire of conjugates for use in fluorinase catalysed radio-synthesis for PET and
shows that the fluorinase will accept a wide range of ClDA substrates tethered at C-2 of the adenine
ring with a pegylated cargo.
Experimental Section
Cold Transhalogenation assay
A typical cold transhalogenation assay is exampled in 26. In a total reaction volume of 1000 μL (in 50
−
1
mM phosphate buffer, at pH 7.8), recombinant fluorinase (0.8 mg/mL ) was incubated with ClDA-
PEG-biotin 26 (0.1 mM), L-Se-Met (0.075 mM) and KF (50 mM) at 37 °C. Samples (50 μL) were
periodically removed, the protein precipitated by heating at 95 °C for 5 min, before being clarified by
centrifugation (13 000 rpm, 10 min). Samples of the supernatant (40 μL) were removed for analysis by
HPLC. HPLC analysis was performed on a Shimadzu Prominence system using a Kinetix 5 μm XB-C18
This article is protected by copyright. All rights reserved.

ChemBioChem
10.1002/cbic.201800234
1
00A (150 mm × 4.6 mm) column and a guard cartridge. Mobile phase: 0.05% TFA in water (solvent A)
and 0.05% TFA in MeCN (solvent B); Linear Gradient: 15% solvent B to 95% solvent B over 25 min, 95%
−
1
for 5 min, and back to 15% B for 10 min to re-equilibrate the column. Flow rate: 1 mL/min ; Detection:
54 nm; Injection volume: 40 μL.
2
HABA Assay
Preliminary experiments were performed varying the concentrations of HABA and avidin to determine
the optimal conditions for the competitive binding studies. In the final experiments a range of
concentrations (0-200 μM) of FDA-PEG-Biotin 27 and (+)-biotin were added to a HABA/avidin complex,
obtained by mixing avidin and HABA to final concentrations of (2.5 mg/ml) and 0.25 μM respectively.
Changes in absorbance were measure at 500 nm and the % inhibition was calculated using the
following formula: {[(absorbance with competitor)
−
(absorbance without HABA)/(Av
complex)]/[(absorbance without competitor) − (absorbance without HABA)/(Av complex)]} × 100.
[18
F] Labelling of 26 to [ F]27
18
18 -
A typical F labelling experiment of 26 was performed as follows: L-selenomethionine (40 μL of a
mM solution in water) and compound 26 (0.4 mg in 50 μL of water) were added successively to an
Eppendorf tube containing a solution of fluorinase (5 mg in 50 mm phosphate buffer, 110 μL). The
2
18
-
18
contents were mixed well with a pipette and to this mixture was added [ F ]fluoride in [ O]water
457 MBq, 50 μL), making a total volume of 250 μL. The contents were again well mixed and incubated
at 37 °C for 30 min. After this time the reaction was stopped and the mixture denatured by heating at
5 °C for 5 min and water (400 μL) added before being clarified by centrifugation (13 000 rpm,
(
9
corresponding to 16060 g, 5 min). The supernatant was injected into a Shimadzu Prominence HPLC
system equipped with a quaternary pump, a degasser, a flow cell detector and a diode array detector
using a Phenomenex Kingsorb C18 (250 × 10.00 mm, 5 μm) column and a guard cartridge. Mobile
phase: 0.05% TFA in water (solvent A) and 0.05% TFA in MeCN (solvent B); Linear Gradient: 15%
solvent B to 38% solvent B over 16 min, 95% for 5 min, and back to 15% B for 10 min to re-equilibrate
−
1
18
the column. Flow rate: 5 mL/min . The radioactive fraction corresponding to the reference of [ F]27
was collected, diluted with 10 mg/mL ascorbate in water (50 mL) and loaded onto a preactivated
Waters Oasis HLB® Cartridge (conditioned with 2 mL EtOH and 5 mL 10 mg/mL ascorbate/water). The
cartridge was washed with 20 mL of 10 mg/mL ascorbate/water and the desired product was collected
by eluting with 0.6 mL of ethanol, to give 59.7 MBq (13 %, decay uncorrected) of >99% pure product
1
8
of [ F]27.
Streptavidin binding Radio TLC assay
18
To PBS buffer (300 μL, pH 7.8) or streptavidin (0.6 mg) in PBS buffer (300 μL, pH 7.8) was added [ F]27
(
0.8 MBq, in a 50 L solution of 1:5 EtOH:water), the reaction mixture was mixed well and incubated
◦
at 37 C for 30 min. Thin layer chromatography was then conducted on both samples (+/- streptavidin)
using aluminium plates coated with silica gel (60F245 Merck), eluted with a solution 75:25:5
This article is protected by copyright. All rights reserved.

ChemBioChem
10.1002/cbic.201800234
MeCN:MeOH:Water. TLC plates were examined using a radio-TLC scanner to determine the
18
percentage of streptavidin bound [ F]27. Thin layer chromatograms radio-traces of the samples
1
8
[
F]27 in PBS buffer revealed an R/F value of 0.39, upon incubation with streptavidin thin layer
18
chromatograms radio-traces indicated almost completely retention of [ F]27 (about 96%) in the origin
of the TLC plate (R/F value of 0.02, see Supporting Information for full experimental details and
Figures).
See Supporting Information for experimental detail on: Compound synthesis and characterisation,
assay conditions and fluorinase overexpression and purification.
Acknowledgements
We thank the Engineering and Physical Sciences Research Council, UK, for a research grant.
18
Keywords: fluorinase· amide bioconjugation · tetrazine · biotin · F labelling
References
[
[
1]
2]
O. Mawlawi, D. W. Townsend, Eur. J. Nucl. Med. Mol. Imaging, 2009, 36 Suppl 1, S15-29.
a) J. Hoppin, K. D. Orcutt, J. Y. Hesterman, M. D. Silva, D. Cheng, C. Lackas, M. Rusckowski, J.
Pharmacol. Exp. Ther., 2011, 337, 350-358; b) M. Rashidian, E. J. Keliher, M. Dougan, P. K.
Juras, M. Cavallari, G. R. Wojtkiewicz, J. T. Jacobsen, J. G. Edens, J. M. J. Tas, G. Victora, R.
Weissleder, H. Ploegh, ACS Cent. Sci., 2015, 1, 142-147; c) R. Tavare, M. N. McCracken, K. A.
Zettlitz, S. M. Knowles, F. B. Salazar, T. Olafsen, O. N. Witte, A. M. Wu, Proc. Natl. Acad. Sci.
U. S. A., 2014, 111, 1108-1113; d) J. Clark, D. O’Hagan, J. Fluorine Chem., 2017.
[
[
3]
4]
E. M. Sletten, C. R. Bertozzi, Angew. Chem., Int. Ed., 2009, 48, 6974-6998.
a) M. Patra, K. Zarschler, H.-J. Pietzsch, H. Stephan, G. Gasser, Chem. Soc. Rev., 2016, 45,
6
415-6431; b) R. Rossin, M. S. Robillard, Curr. Opin. Chem. Biol., 2014, 21, 161-169; c) J. P.
Meyer, J. L. Houghton, P. Kozlowski, D. Abdel-Atti, T. Reiner, N. V. Pillarsetty, W. W. Scholz,
B. M. Zeglis, J. S. Lewis, Bioconjugate Chem., 2016, 27, 298-301; d) M. L. Blackman, M.
Royzen, J. M. Fox, J. Am. Chem. Soc., 2008, 130, 13518-13519; e) N. K. Devaraj, R.
Weissleder, S. A. Hilderbrand, Bioconjugate Chem., 2008, 19, 2297-2299.
a) J. Seckute, N. K. Devaraj, Curr. Opin. Chem. Biol., 2013, 17, 761-767; b) Z. Li, H. Cai, M.
Hassink, M. L. Blackman, R. C. Brown, P. S. Conti, J. M. Fox, Chem. Commun., (Cambridge,
U.K) 2010, 46, 8043-8045; c) J. C. Knight, S. Richter, M. Wuest, J. D. Way, F. Wuest, Org.
Biomol. Chem., 2013, 11, 3817-3825.
C. Denk, D. Svatunek, T. Filip, T. Wanek, D. Lumpi, J. Frohlich, C. Kuntner, H. Mikula, Angew.
Chem., Int. Ed. Engl., 2014, 53, 9655-9659.
a) E. Blom, B. Langstrom, I. Velikyan, Bioconjugate Chem., 2009, 20, 1146-1151; b) R. W.
Sirianni, M. Q. Zheng, T. R. Patel, T. Shafbauer, J. Zhou, W. M. Saltzman, R. E. Carson, Y.
Huang, Bioconjugate Chem., 2014, 25, 2157-2165; c) M. Mamede, T. Saga, H. Kobayashi, T.
[
5]
[
[
6]
7]
This article is protected by copyright. All rights reserved.

ChemBioChem
10.1002/cbic.201800234
Ishimori, T. Higashi, N. Sato, M. W. Brechbiel, J. Konishi, Clin. Cancer Res., 2003, 9, 3756-
762.
D. J. Hnatowich, F. Virzi, M. Rusckowski, . Nucl. Med., Society of Nuclear Medicine, 1987, 28,
294-1302.
D. M. Goldenberg, R. M. Sharkey, G. Paganelli, J. Barbet, J.-F. Chatal, J. Clin. Oncol., 2006, 24,
23-834.
a) G. J. Förster, E. B. Santos, P. M. Smith-Jones, P. Zanzonico, S. M. Larson, J. Nucl. Med.,
006, 47, 140-149; bT. Kudo, M. Ueda, H. Konishi, H. Kawashima, Y. Kuge, T. Mukai, A.
Miyano, S. Tanaka, S. Kizaka-Kondoh, M. Hiraoka, H. Saji, Mol. Imaging Biol., 2011, 13, 1003-
010.
M. Simpson, L. Trembleau, R. W. Cheyne, T. A. D. Smith, Appl. Radiat. Isot., 2011, 69, 418-
22.
3
[
[
[
8]
1
9]
8
10]
2
1
[
[
11]
12]
4
a) E. Blom, O. Itsenko, B. Långström, J. Labelled Compd. Radiopharm., 2011, 54, 681-683; b)
M. Claesener, H.-J. Breyholz, S. Hermann, A. Faust, S. Wagner, O. Schober, M. Schäfers, K.
Kopka, Nucl. Med. Biol., 2012, 39, 1189-1194.
[
13]
a) T. A. D. Smith, M. Simpson, R. Cheyne, L. Trembleau, Appl. Radiat. Isot., 2011, 69, 1395-
1
400; b) L. Trembleau, M. Simpson, R. W. Cheyne, I. Escofet, M. V. C. A. L. Appleyard, K.
Murray, S. Sharp, A. M. Thompson, T. A. D. Smith, New J. Chem., 2011, 35, 2496-2502.
S. Thompson, Q. Zhang, M. Onega, S. McMahon, I. Fleming, S. Ashworth, J. H. Naismith, J.
Passchier, D. O'Hagan, Angew. Chem., Int. Ed. Engl., 2014, 53, 8913-8918.
[
[
[
[
[
14]
15]
16]
17]
18]
Q. Zhang, S. Dall'Angelo, I. N. Fleming, L. F. Schweiger, M. Zanda, D. O'Hagan, Chem. Eur. J.,
2
016, 22, 10998-11004.
D. O'Hagan, C. Schaffrath, S. L. Cobb, J. T. G. Hamilton, C. D. Murphy, Nature, 2002, 416, 279-
79.
2
H. Deng, S. L. Cobb, A. R. McEwan, R. P. McGlinchey, J. H. Naismith, D. O'Hagan, D. A.
Robinson, J. B. Spencer, Angew. Chem., Int. Ed. Engl., 2006, 45, 759-762.
a) S. Thompson, M. Onega, S. Ashworth, I. N. Fleming, J. Passchier, D. O'Hagan, Chem.
Commun., 2015, 51, 13542-13545; b) S. Thompson, I. N. Fleming, D. O'Hagan, Org. Biomol.
Chem., 2016, 14, 3120-3129.
[19]
a) W. L. Yeo, X. Chew, D. J. Smith, K. P. Chan, H. Sun, H. Zhao, Y. H. Lim, E. L. Ang, Ang, Chem.
Commun., 2017, 53, 2559-2562; b) H. Sun, W. L. Yeo, Y. H. Lim, X. Chew, D. J. Smith, B. Xue,
K. P. Chan, R. C. Robinson, E. G. Robins, H. Zhao, E. L. Ang, Angew. Chem., Int. Ed. Engl.,
2
016, 55, 14277-14280.
[
[
[
20]
21]
22]
P. T. Lowe, S. Dall'Angelo, T. Mulder-Krieger, A. P. Ijzerman, M. Zanda, D. O'Hagan,
Chembiochem., 2017, 18, 2156-2164.
M. Asti, M. Iori, P. A. Erba, G. Atti, D. Farioli, C. Guidotti, A. Versari, Nucl. Med. Commun.,
2
012, 33, 1179-1187.
a) A. Bogusiewicz, S. L. Stratton, D. A. Ellison, D. M. Mock, Am. J. Clin. Nutrit., 2008, 88, 1291-
296; b) K. S. Wang, A. Patel, D. M. Mock, J. Nutr., 1996, 126, 1852-1857; c) J. Zempleni, D. B.
1
McCormick, D. M. Mock, Am. J. Clin. Nutr., 1997, 65, 508-511.
D. M. Mock, G. L. Lankford, N. I. Mock, J. Nutr., 1995, 125, 941-946.
a) P. J. Scott, B. G. Hockley, H. F. Kung, R. Manchanda, W. Zhang, M. R. Kilbourn, Appl.
[
[
23]
24]
Radiat. Isot., 2009, 67, 88-94; bR. M. Fawdry, Appl. Radiat. Isot.,2007, 65, 1193-1201.
This article is protected by copyright. All rights reserved.

ChemBioChem
10.1002/cbic.201800234
Entry for the Table of Contents
FULL PAPER
Phillip T. Lowe,[a] Sergio Dall'Angelo,[b]
Andrew Devine,[ Matteo Zanda,[b] and
David O'Hagan*[
The synthesis and evaluation of new
a]
amine
and
acid
functionalized
a]
fluorinase substrates, allowing for the
attachment to an extensive range of
carboxylate or amine containing
bioactive molecules to a fluorinase
binding 5'-ClDA moiety, prior to their
Page No. – Page No.
Enzymatic fluorination of biotin and
tetrazine conjugates for pre-targeting
approaches to PET imaging.
enzymatic
radiolabelling
with
18
[
F]fluoride. Two new substrates of this
class have been designed specifically for
their application towards a pretargeting
approach for radiolabelling conjugated
antibodies and proteins for PET
imaging.
This article is protected by copyright. All rights reserved.