Last-Step Enzymatic [18F]-Fluorination of Cysteine-Tethered RGD Peptides Using Modified Barbas Linkers

Article abstract of DOI:10.1002/chem.201601361

We report a last-step fluorinase-catalyzed [¹⁸F]-fluorination of a cysteine-containing RGD peptide. The peptide was attached through sulfur to a modified and more hydrophilic variant of the recently disclosed Barbas linker which was itself link

Full text of DOI:10.1002/chem.201601361

DOI: 10.1002/chem.201601361
Full Paper
&
Chemical Biology
1
8
Last-Step Enzymatic [ F]-Fluorination of Cysteine-Tethered RGD
Peptides Using Modified Barbas Linkers
+
[a]
+ [b]
[c]
[b]
Qingzhi Zhang , Sergio Dall’Angelo , Ian N. Fleming, Lutz F. Schweiger,
[
b]
[a]
Matteo Zanda,* and David O’Hagan*
1
8
Abstract: We report a last-step fluorinase-catalyzed [ F]-flu-
orination of a cysteine-containing RGD peptide. The peptide
was attached through sulfur to a modified and more hydro-
philic variant of the recently disclosed Barbas linker which
was itself linked to a chloroadenosine moiety via a PEGylated
chain. The fluorinase was able to use this construct as a sub-
18
strate for a transhalogenation reaction to generate [ F]-ra-
diolabeled RGD peptides, which retained high affinity to
cancer-cell relevant a b integrins.
v
3
Introduction
gated to aminoxy or hydrazine-decorated peptides through ox-
18
imation or hydrazone formation, [ F]-fluorobenzoate
1
8
[5]
18
18
[6]
During the past decade, bioactive peptides, in particular cyclic
RGD peptides, have been increasingly identified as targets for
radiolabeling as positron emission tomography (PET) tracers
due to their high target-selectivity, non-immunogenicity, good
([ F]FBA) or N-succinimidyl 4-[ F]-fluorobenzoate ([ F]SFB)
a
e
for reaction with the N -amino group of the backbone or N -
amino group of lysine residues by acylation or amidation, N-(2-
18
18
[7]
[4-[ F]fluorobenzamido]ethyl)maleimide ([ F]FBEM)
which
[1,2]
18
tissue penetration and rapid metabolic clearance. Fluoride
covalently modify at cysteine RSH residues, and [ F]azide or
18
18
[8,9]
[
F] is the isotope of choice for PET as it has a moderate to
[ F]alkyne substrates
which combine with alkyne or azide
long half-life (109.8 min) and is amenable to relatively extend-
ed synthesis and preparative manipulations. It also has a high
modified peptides by “click” reactions. The aldehyde form of
18
18
[10]
the sugars 2-deoxy-2-[ F]fluoro-d-glucose ([ F]FDG) and 5-
+
+
18
18
[11,12]
percentage of b decay (97%) and a low b energy (635 Kev)
and therefore a short positron range (2.3 mm in water). This
leads to increased image resolution. Also the ease of produc-
deoxy-5-[ F]fluoro-d-ribose ([ F]FDR)
have also been used
for aminooxy group conjugation to peptides and have some
advantage in that they display increased hydrophilicity and
a favorable radiopharmacokinetic profile. Also a new prosthetic
18
tion of [ F] at nm levels and in high specific activity by proton
+
18
18
18
[13]
(
H ) bombardment of an [ O]H O target on a cyclotron, pro-
group, 4-[ F]fluorophenylboronic acid ([ F]FPB) was devel-
oped for Suzuki coupling with aryliodide derivatives of biomol-
ecules.
2
vides a ready source of the isotope in radiochemistry laborato-
1
8
ries, particularly within hospital environments. [ F]-Fluorination
of peptides and other biomolecules is usually achieved by con-
18
While the best coupling reactions between the [ F] pros-
thetic group and peptides are mild, biorthogonal and site-spe-
1
8
jugation with small [ F]-labeled prosthetic synthons through
complimentary functionalization. The most frequently em-
18
cific, the syntheses of the [ F]-building blocks requires pre-
1
8
18
18
18
ployed small [ F] molecules are prepared by nucleophilic [ F]
drying of the [ F]-fluoride, which is generated from [ O]H O.
2
1
8
fluoride chemistry to generate FÀC bonds. These include
Good nucleophilic reactions require anhydrous solvents to
ensure fluoride nucleophilicity. This can often require harsh
conditions incompatible with peptides, such as elevated tem-
18
18
[3,4]
[
F]fluorobenzaldehyde ([ F]FB)
which is generally conju-
18
+
[
a] Q. Zhang, Dr.Prof. D. O’Hagan
peratures. Therefore, a last-step [ F]-fluorination of a substrate
under ambient aqueous conditions is attractive for radiosyn-
thesis and is a current research focus within the radiochemistry
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
18
community. To this end [ F]-fluoride has been reacted in last-
step protocols with organosilicon decorated peptides to give
+
[
b] Dr. S. Dall’Angelo, Dr. L. F. Schweiger, Prof. M. Zanda
John Mallard Scottish PET Centre
1
8
18
19
[14,15]
School of Medicine, Medical Sciences and Nutrition
University of Aberdeen, Foresterhill, Aberdeen, AB25 2ZD (UK)
E-mail: m.zanda@abdn.ac.uk
[ F-SiFA]-conjugate by FÀ F isotopic exchange.
Similarly,
arylfluoroboronate-modified peptides have been induced to
18
capture carrier-added [ F]-fluoride under aqueous acidic con-
[
c] Dr. I. N. Fleming
18
[16,17]
dition in a single step to generate [ F]-ArBF -peptides.
Al-
3
Aberdeen Biomedical Imaging Centre, Institute of Medical Sciences
Foresterhill, Aberdeen, AB25 2ZD (UK)
18
ternatively, [ F]-fluoride can be locked into chelator (NODA or
NOTA) functionalized peptides by coordination chemistry with
+
[
] These authors contributed equally to this work.
3
+
18
[18,19]
Al as a last step treatment, to generate Al–[ F] tracers.
Supporting information for this article can be found under http://
dx.doi.org/10.1002/chem.201601361.
Chem. Eur. J. 2016, 22, 10998 – 11004
10998
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Scheme 1. Fluorinase-catalyzed fluorination and chlorination of AdoMet as well as the conversion of 5’-ClDA to 5’-FDA.
1
8
Our focus has been on enzymatic incorporation of [ F] into
candidate radiotracers under ambient conditions. The fluori-
nase enzyme, originally isolated from Streptomyces cattleya,
catalyzes the conversion of S-adenosyl-l-methionine (AdoMet)
and fluoride ion to 5’-fluoro-5’-deoxyadenosine (5’-FDA) and l-
Conventional site-specific conjugates of cysteine with malei-
mide suffer from degradation via retro-Michael reaction, and
competition from thio-containing peptides or proteins such as
[
23]
glutathione or serum albumin present in vivo.
Barbas and co-workers recently introduced methylsulfonyl
oxadiazole for bioconjugation through nucleophilic aromatic
substitution by cysteine residues to generate a relatively stable
[
20]
methionine (l-Met). Fluorinase also functions as a chlorinase,
particularly in the reverse direction. The reversibility allows
’-chloro-5’-deoxyadenosine (5’-ClDA) to be used as a substrate
[21]
[
24,25]
5
oxadiazole-protein thiolate.
We saw an opportunity for
for fluorination in a one step reaction via AdoMet under ambi-
combining this Barbas linker with our fluorinase technology in
order to generate a thiol ligated peptides for last step enzy-
ent (pH 7.8 phosphate buffer, 378C) aqueous conditions
18
(
Scheme 1).
matic [ F]-fluorination under ambient aqueous condition
(Scheme 2).
The reaction benefits from the fact that the equilibrium
favors 5’-FDA in fluorination but AdoMet in chlorination. In ad-
dition, the use of l-selenomethione (l-SeMet) in place of l-Met
increases the reaction rate and improves the efficiency of the Results and Discussion
transhalogenation of (5’-ClDA) to FDA. Fluorinase displays
RGD conjugation to a chlorodeoxyadenosine through modi-
fied Barbas linkers
a high substrate specificity, however, we have recently demon-
strated that the introduction of an acetylene group at C-2 of
[
22]
the adenine base of 5’-ClDA is tolerated by the enzyme. In
addition this acetylene can be extended at its terminus to
carry a pegylated cargo. Decoration of 5’-ClDA in this way has
allowed bioconjugation through “click” chemistry with azido
functionalized peptides. In this context we have reported
To obtain the conjugate, we attempted to click ClDA-TEG (2)
and 2-methylsufonyl 5-(4-azidotetraethyleneglycoxyphenyl)oxa-
[
23]
diazole (1a)
in the first step, but the free methylsulfonyl
moiety was gradually hydrolyzed during work up and this was
not progressed. However when the sequence was reversed
(Scheme 3) by first reacting the methylsulfonate 1a with the
RGD, then the azido phenyloxadiazole thiolate (APOT-RGD) 3a
was obtained in excellent yield.
1
8
a “last-step” enzymatic [ F]-fluorination of a 5’-ClDA-RGD pep-
[22]
tide conjugate with the fluorinase. This approach was good
for lysine modified azido containing RGD peptides. It became
an objective to develop a protocol compatible with cysteine li-
gation due to the general importance of cysteine in bioconju-
gation, often through an engineered cysteine thiol. A success-
ful strategy would significantly broaden the utility of the enzy-
matic methodology for radiolabeling peptides and proteins.
Conjugate 3a is very insoluble in common solvents such as
MeOH, CH CN and H O and was only partially soluble in
3
2
DMSO. A dilute solution of conjugate 3a in DMSO/H O with
2
[22]
ClDA-TEG-acetylene (2) under Cu-catalyzed “click” conditions
generated ClDA-APOT-RGD (4a), which was purified by semi-
18
Scheme 2. Design of three components for last-step [ F] labeling of cysteine-containing peptide: fluorinase substrate, Barbas linker and the cysteine-contain-
ing peptide.
Chem. Eur. J. 2016, 22, 10998 – 11004
www.chemeurj.org
10999
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Scheme 3. Successive conjugation of methylsulfonyl azido phenyloxadiazole, RGD and ClDA-TEG or FDA-TEG and fluorinase-catalyzed transhalogenation. Re-
agents and conditions: a) c(RGDfC), PBS buffer (pH 7.4), 2 h; b) compound 2, CuTBTA in 55% DMSO, sodium ascorbate, H O, overnight; c) KF, l-SeMet, fluori-
2
nase, phosphate buffer (pH 7.8).
preparative HPLC. Again, 4a is insoluble in common solvents
with only partial solubility in DMSO.
The increased polarity and hydrophilicity of 4c is also reflect-
ed in reverse phase HPLC profiles. While 4a and 4b have
almost the same retention time, 4c elutes over 1 min earlier
than 4a and 4b under the conditions (Figure 1).
Simultaneously, a variant of the Barbas linker was prepared
without an aryl ring to explore solubility. Thus methylsulfonyl
oxadiazole (1d) became a synthesis target (Supporting Infor-
mation, Scheme S3). A key reaction involved nucleophilic sub-
stitution of the methylsulfonate 1d with the thiol of (RGDfC)
and this led to azido oxadiazole thiolate (AOT-RGD) 3d. “Click”
To improve the solubility, a second PEGylated chain was in-
troduced at the 3-position on the phenyl ring of the Barbas
linker. This involved the selective deacylation and Williamson
etherification at C-4 of the precursor benzoate in a one pot
[
26,27]
protocol to give 7b (Supporting Information, Scheme S1).
Further etherification of 3-OH with tosylate 13 afforded 8b.
Product 8b was condensed with hydrazine and then cyclized
with carbon disulfide in the presence of KOH to generate oxa-
diazole 10b. Finally thiol methylation and then oxidation with
m-CPBA gave bisPEGlated azido phenyloxadiazole 1b
[
24]
(
Scheme S1).
Conjugation of 1b with c(RGDfC) as shown in Scheme 3 af-
forded 3b (MAPOT-RGD), and this construct was “clicked” with
affording (ClDA-MAPOT-RGD) 4b. The solubility of 4b was
2
not significantly improved relative to 4a particularly in acetoni-
trile, methanol, ethanol and water.
There was the option of introducing a second PEGylated
chain at C-5 of the phenyl ring, however the molecular weight
of the construct was getting large, and therefore it seemed
more appropriate to explore the free diphenol 4c as a poten-
tially more soluble construct. The settled route to the analo-
gous precursor 1c is shown in Supporting Information
(
Scheme S2). Thiol conjugation of 1c with peptide c(RGDfC) in
THF/phosphate buffer afforded 3c, which was then “clicked”
with 2 under standard Cu-catalyzed conditions to give 4c
(
Scheme 3). It is immediately apparent that 3c and 4c have in-
creased solubility in MeOH and CH CN relative to 3a/b and
3
4
4
a/b. They have moderate water solubility (up to 0.1 mm for
c). The increased solubility of 3c/4c makes them easier to
Figure 1. Comparative HPLC traces to illustrate the relative polarities of 4a
(least polar), 4b, 4c and 4d (most polar) by retention time [min]. For HPLC
handle and they do not require [D ]DMSO for characterization.
conditions see the Experimental Section.
6
Chem. Eur. J. 2016, 22, 10998 – 11004
www.chemeurj.org
11000
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Scheme 4. Preparation of water soluble fluorinase substrate ClDA-AOT-RGD. Reagents and conditions: a) c(RGDfC), PBS buffer (pH 7.4), 2 h; b) compound 2,
CuTBTA in 55% DMSO, sodium ascorbate, H O, 12 h; c) KF, l-SeMet, fluorinase, phosphate buffer (pH 7.8).
2
reaction of 3d with 2 gave conjugate ClDA-AOT-RGD (4d;
Scheme 4).
tion, Figures S1 and S2). Substrate 4d was processed most effi-
ciently and at about twice the rate of substrates 4a–c. An ad-
ditional HPLC peak (at 16.1 min in Figure 2A and 14.9 min in
Figure 2B) was attributed to a decomposition product associat-
ed with the AdoSeMet intermediate of the transhalogenation
reactions based on MS analysis (exampled by Supporting Infor-
mation Figure S3).
The solubility of 4d in water is significantly increased
(
2.3 mm) relative to 4c (0.1 mm). Compounds 4a and 4b, are
essentially insoluble. Reverse phase HPLC retention times were
used as a proxy measure of solubility/polarity and 4d and then
4
c eluted most rapidly, as illustrated Figure 1.
The corresponding fluorinated conjugates 5a–d were pre-
[
22]
pared from reactions of 3a–d and FDA-TEG 15 (Supporting
Information, Schemes S4 and S5).
1
8
Enzymatic hot [ F]-fluorination of 4a–c to 5a–c
18
For hot labeling, the [ F]-fluoride is generated in GBq and
18
used in MBq aliquots for reactions. The actual [ F]-fluoride
concentration is very low in the picomolar range but the
enzyme is in excess in the micromolar range and therefore the
reactions are no longer catalytic.
“
Cold” enzymatic fluorination of 4a–d to 5a–d
Non-radiolabeled fluorinase reactions (last step in Scheme 3
and 4) were generally performed in phosphate buffer (pH 7.8,
2
0 to 50 mm) at 378C with the appropriate 5’-chloro-5’-deoxya-
This stoichiometry is dramatically reversed relative to the
cold reactions which use a large molar excess of fluoride over
denosine substrate (0.1 mm), l-SeMet (0.1 mm), fluorinase
À1
18
(
1 mgmL ) and potassium fluoride (75 mm). For 4a, DMSO
enzyme. The very low [ F] concentration requires a high sub-
5
% v/v was added for solubility, although DMSO is detrimental
strate concentration, and the low solubility of 4a and 4b in
to enzyme activity and promotes gelling on heat denaturation
of the enzyme on workup. The enzymatic conversion of 4a
proved sluggish (Supporting Information, Figure S1) and did
not improve for 4b with the added PEG chain (Figure S2).
Constructs 4c (up to 0.1 mm) and 4d (up to 2.3 mm) had im-
proved water solubility, and reactions were carried out in phos-
phate buffer (pH 7.8, 50 mm) without added DMSO. Workup of
the neat aqueous reaction mixture was more straightforward
than that for 4a and 4b which contained DMSO. The fluori-
nase enzyme was cleanly precipitated on heating at 958C for
5% DMSO limited their reaction concentrations to 0.15 and
18
À
0.48 mm, respectively. The conversion of
F into 5a and 5b
was only modest (Supporting Information, Figures S4 and S5,
based on the integration of the peaks of product and unreact-
18
À
[28]
ed
F ion by HPLC ). Substrate 4d, has good solubility but
proved to be hydrolytically unstable (Figure 1, bottom HPLC
trace shows a degraded product at 13.2 min) relative to 4a–c,
and 5d visibly decomposed on work up, therefore radiochemi-
cal trials were not explored further with this substrate.
Substrate 4c has good solubility in water/DMSO and an effi-
cient transhalogenation of 4c (reaction conc. ꢀ0.66 mm) to
5
min and could be readily centrifuged (13000 rpm, which cor-
18
responds to 16060 g, 5 min).
[ F]5c (final conc. of DMSO 2%, Supporting Information, Fig-
[
28]
The biotransformations of 4c to 5c and 4d to 5d (Figure 2)
are more efficient than that of 4a and 4b (Supporting Informa-
ure S6) was accomplished (87% by HPLC). At 2% DMSO, the
enzyme precipitated cleanly after heat denature without gel
Chem. Eur. J. 2016, 22, 10998 – 11004
www.chemeurj.org
11001
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 2. HPLC traces monitoring fluorinase-catalyzed conversions [%] of: A) 4c (t
B) 4d (t =14.4 min) to 5d (t =13.4 min) via AdoSeMet intermediate (t =14.9 min). For full conditions see the Experimental Section and Supporting Informa-
tion, Figures S1 and S2.
R R R
=15.7 min) to 5c (t =14.9 min) via AdoSeMet intermediate (t =16.1 min).
R
R
R
1
8
formation. The semi-prep HPLC purified [ F]5c was diluted
with water and then secured on a C18 reverse phase cartridge.
The product was finally eluted with EtOH (Figure 3
and Supporting Information, Figure S7) for bioaffinity
50.2) nm), 5b (IC50 80.1(Æ14.2) nm) and 5c (IC 177(Æ23) nm)
50
proved to be similar, indicating that modification of RGD pep-
1
8
experiment. The total procedure from [ F]-fluoride
1
8
(
422 MBq) to EtOH elution of [ F]5c (22.5 MBq) took
.5 h, and with a radiochemical yield of 5% (decay
1
uncorrected).
Bioactivity of FDA-AOT-RGD conjugate 5a–c
Bioaffinity assays were conducted on cell lines ex-
pressing a b integrins. The inhibitory concentrations
v
3
(
IC ’s) for 5a–c, competing with a biotinylated pep-
50
tide binding to immobilized a b integrins were ex-
v
3
plored. This is a proxy measure of binding affinity to
a b integrins on the surface of cancer cells. The IC
50
v
3
values were compared with that of the cyclic pep-
18
Figure 3. Analytical HPLC radio trace of purified [ F]5c. For full conditions see the Exper-
tide, cRGDfC (85.0 nm). The values, 5a (IC50 85.9(Æ
imental Section and Supporting Information, Figure S7.
Chem. Eur. J. 2016, 22, 10998 – 11004
www.chemeurj.org
11002
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
tide with the fluoroadenosine/Barbas linker construct does not
significantly affect its binding affinity.
100 mm or solid, 2–5 equiv) was added to the mixture. The mixture
was degassed with argon before CuTBTA (10 mm in 55% DMSO,
0.2–0.5 equiv) was added. The mixture was heated at 608C for
18
Binding of [ F]5c to a b integrins in cancer cells was evalu-
v
3
1
0 min and stirred at RT, overnight. After dilution with suitable
ated in U87MG and PC3 cells, which are known to express
amount of DMSO/H O, the mixture was centrifuged at 13000 rpm
[29]
18
2
high and medium levels of receptor. Binding of [ F]5c to
I
(
which corresponds to 16060 g) for 10 min to remove the Cu pre-
U87MG cells was higher than to PC3 cells. Cold c(RGDfK)
cipitate. The supernatant was isolated by semi-preparative HPLC to
give the product 4a–d or 5a–d after concentration and freeze
drying.
(
10 mm) was included in the binding assay to permit identifica-
tion of specific radiotracer binding to a b integrin (decreased
v
3
1
8
by c(RGDfK). Inclusion of cold peptide decreased [ F]5c bind-
ing to both U87MG and PC3 cells by approximately 75 and
Fluorinase assay of 4a–c to 5a–c
5
0%, respectively (Figure 4). These data are consistent with the
[
29]
literature and flow cytometry analysis which demonstrated
In a total reaction volume of 200 mL, recombinant fluorinase
À1
that our U87MG cells contained 9-times more a b than PC3
(1 mgmL , in phosphate buffer) was incubated with l-SeMet
v
3
(
0.1 mm), KF (75 mm) and 4a (0.1 mm) or 4b (0.1 mm) in 5%
(
[
data not shown). This data provide strong evidence that
18
DMSO/phosphate buffer (pH 7.8, 50 mm) or 4c or 4d (0.1 mm) in
phosphate buffer (pH 7.8, 50 mm) at 378C. Samples (20 mL) were
periodically removed, denatured by heating at 958C for 5 min, di-
luted with phosphate buffer (80 mL, 20 mm) before being clarified
by centrifugation (13000 rpm, which corresponds to 16060 g,
F]5c is binding selectively to a b integrin in cancer cells.
v
3
1
0 min). Samples of the supernatant (80 mL) were removed for
analysis by HPLC. HPLC analysis (Figure 2, and Supporting Informa-
tion, Figures S2 and S3) was performed on a Shimadzu Prominence
HPLC system fitted with a SIL-20A HT autosampler, LC-20AT solvent
delivery system, SPD-20 UV/Vis detector using a Phenomenex Luna
5
mm, C-18 100A (250”4.6 mm) column and a guard cartridge.
Mobile phase: 0.05% TFA in water (solvent A) and 0.05% TFA in
acetonitrile (solvent B); Linear gradient: 20–50% B in 20 min, 95%
B in 25 min followed by equilibration of the column with initial
18
À1
Figure 4. Bound [ F]5c [cpmmg ] protein per MBq radiotracer.
À1
condition; flow rate of 1 mLmin ; detection: 254 nm; injection
volume: 50 mL.
Conclusions
We have modified a cysteine-containing RGD peptide with
a fluorinase-recognized chloroadenosine for fluorinase mediat-
ed radiolabeling with the fluorine-18 isotope. The recently in-
troduced Barbas bioconjugation strategy was variously modi-
fied to improve solubility. The diphenol 4c, which derived
from the sulfonyl oxadiazole 1c, proved to be the most soluble
construct and performed best for enzymatic mediated transha-
1
8
18
[
F] Labelling of 4a–c to [ F]5a–c
18
À
The typical
F labeling of 4 is exampled with 4c: l-selenomethio-
nine (40 mL), phosphate (50 mm, 30 mL) and fluorinase (5 mg in
50 mm phosphate buffer, 125 mL) were added successively to an
Eppendorf tube containing substrate of 4c (5 mL, 33 mm in DMSO).
The contents were mixed well with a pipette. To this mixture was
added 50 mL of [ F]-fluoride in [ O]water (380 MBq). The contents
were well mixed and incubated at 378C for 30 min. After this time,
an aliquot of the sample (20 mL) was taken for HPLC analysis. Once
18
18
1
8
logenation to generate radiolabeled [ F]5c. Sulfonyl oxadi-
azole 1c may find more general utility due to its relatively
high aqueous solubility.
18
the conversion of 4c to [ F]5c was confirmed by the analysis, the
remaining mixture was denatured by heating at 958C for 5 min
and the precipitated protein was removed by centrifugation
Experimental Section
(
13000 rpm, which corresponds to 16060 g, 5 min). The superna-
tant 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 Luna 5 mm, C-18
100 A (250”10 mm) column and a guard cartridge; Mobile phase:
0.05% TFA in water (solvent A) and 0.05% TFA in acetonitrile (sol-
vent B); linear gradient: 0–4 min, 0% B, 4–10 min, 30%, 10–25 min
33%, 28 min 90% B, 35 min 0% B, 40 min stop. Flow rate:
Bioconjugation of Barbas linker with c(RGDfC)
Compound 1a–d (1.5–4 equiv) in THF (400 mL) was added to a sus-
pension of c(RGDfC) (1 equiv) in in PBS buffer (200 mL, 100 mm,
pH 7.4). The mixture was stirred at RT for 0.5–2 h and then concen-
trated under reduced pressure to remove the organic solvent, leav-
ing an aqueous suspension containing product and excess of start-
ing material. The excess 1a–d was removed effectively by extrac-
tion with diethyl ether; the aqueous suspension was dissolved/di-
À1
4 mLmin . The radioactive fraction corresponding to the reference
18
of [ F]5c was collected, diluted with water (10 mL) and loaded
ꢁ
luted with H O/DMSO and purified by semi-preparative HPLC to
onto a preactivated Waters Oasis HLB Cartridge (conditioning with
2
give 3a–d.
2 mL EtOH, 5 mL water). The cartridge was washed with 20 mL of
water. The desired product was collected by eluting with 1 mL of
ethanol, to give about 22.5 MBq (5%, decay uncorrected) of pure
product of [ F]5c as evidenced by analytical HPLC using a Phe-
nomenex Luna 5 mm, C-18 100A (250”4.6 mm, 5 mm) column and
“Click” reaction of 3a–d with 2 or 15 for 4a–d or 5a–d
18
ClDA-TEG (2) or FDA-TEG (15; 1.2–2.0 equiv) in a suitable amount
of water was added to 3a–d (1 equiv) in a suitable amount of
DMSO. Sodium ascorbate (fresh aqueous degassed solution
a guard cartridge; mobile phase: A (H O+0.05% TFA), B (MeCN+
2
0.05% TFA); linear gradient: 0–4 min 0% B, 4–10 min 0–30% B,
Chem. Eur. J. 2016, 22, 10998 – 11004
www.chemeurj.org
11003
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
1
4
0–25 min 30–33% B, 25–30 min 33–90% B, 30–35 min 0% B,
0 min stop. Flow rate: 1.0 mLmin .
[15] C. Wängler, S. Niedermoser, J. Chin, K. Orchowski, E. Schirrmacher, K.
Jurkschat, L. Iovkova-Berends, A. P. Kostikov, R. Schirrmacher, B. Wängler,
Nat. Protoc. 2012, 7, 1946–1955.
À1
[16] Y. Li, J. Guo, S. Tang, L. Lang, X. Chen, D. M. Perrin, Am. J. Nucl. Med.
Mol. Imaging 2013, 3, 44–56.
Acknowledgements
[17] S. Liu, R. Park, P. S. Conti, Z. Li, Am. J. Nucl. Med. Mol. Imaging 2013, 3,
97–101.
We thank the Engineering and Physical Sciences Research
Council, UK, for a research grant.
[18] W. J. McBride, C. A. D’Souza, H. Karacay, R. M. Sharkey, D. M. Goldenberg,
Bioconjugate Chem. 2012, 23, 538–547.
[19] W. Wan, N. Guo, D. Pan, C. Yu, Y. Weng, S. Luo, H. Ding, Y. Xu, L. Wang,
L. Lang, Q. Xie, M. Yang, X. Chen, J. Nucl. Med. 2013, 54, 691–698.
1
8
Keywords: F labeling · Barbas linker · bioconjugation ·
chemical biology · RGD peptide
[20] D. O’Hagan, C. Schaffrath, S. L. Cobb, J. T. G. Hamilton, Nature 2002, 416,
79–280.
2
[
[
[
21] 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. 2006, 45,
[
[
[
1] S. Liu, Bioconjugate Chem. 2015, 26, 1413–1438.
2] S. Richter, F. Wuest, Molecules 2014, 19, 20536–20556.
3] T. Poethko, M. Schottelius, G. Thumshirn, U. Hersel, M. Herz, G. Henrik-
sen, H. Kessler, M. Schwaiger, H. J. Wester, J. Nucl. Med. 2004, 45, 892–
759–762; Angew. Chem. 2006, 118, 773–776.
22] S. Thompson, Q. Zhang, M. Onega, S. McMahon, I. Fleming, J. H. Nai-
smith, J. Passchier, D. O’Hagan, Angew. Chem. Int. Ed. 2014, 53, 8913–
8918; Angew. Chem. 2014, 126, 9059–9064.
9
02.
23] B. Q. Shen, K. Xu, L. Liu, H. Raab, S. Bhakta, M. Kenrick, K. L. Parsons-Re-
ponte, J. Tien, S. E. Yu, E. Mai, D. Li, J. Tibbitts, J. Baudys, O. M. Saad, S. J.
Scales, P. J. McDonald, P. E. Hass, C. Eigenbrot, T. Nguyen, W. A. Solis,
R. N. Fuji, K. M. Flagella, D. Patel, S. D. Spencer, L. A. Khawli, A. Ebens,
W. L. Wong, R. Vandlen, S. Kaur, M. X. Sliwkowski, R. H. Scheller, P. Pola-
kis, J. R. Junutula, Nat. Biotechnol. 2012, 30, 184–189.
[
[
[
4] K. Bruus-Jensen, T. Poethko, M. Schottelius, A. Hauser, M. Schwaige, H. J.
Wester, Nucl. Med. Biol. 2006, 33, 173–183.
5] J. L. Sutcliffe-Goulden, M. J. O’Doherty, P. K. Marsden, I. R. Hart, J. K. Mar-
shall, S. S. Bansal, Eur. J. Nucl. Med. Mol. Imaging 2002, 29, 754–759.
6] S. Richter, M. Wuest, S. S. Krieger, B. E. Rogers, M. Friebe, R. Bergmann, F.
Wuest, Nucl. Med. Biol. 2013, 40, 1025–1034.
[
[
[
[
[
24] N. Toda, S. Asano, C. F. Barbas III, Angew. Chem. Int. Ed. 2013, 52, 12592–
[
[
7] W. Cai, X. Zhang, Y. Wu, X. Chen, J. Nucl. Med. 2006, 47, 1172–1180.
8] K. Kettenbach, H. Schieferstein, T. L. Ross, Biomed Res. Int. 2014, 2014,
12596; Angew. Chem. 2013, 125, 12824–12828.
25] J. T. Patterson, S. Asano, X. Li, C. Rader, C. F. Barbas III, Bioconjugate
Chem. 2014, 25, 1402–1407.
26] Q. Zhang, K. L. Raheem, N. P. Botting, A. M. Z. Slawin, C. D. Kay, D.
O’Hagan, Tetrahedron 2012, 68, 4194–4201.
27] K. M. Müller, L. Brunsveld, Angew. Chem. Int. Ed. 2009, 48, 2921–2924;
Angew. Chem. 2009, 121, 2965–2968.
28] D. Ory, J. Van den Brande, T. de Groot, K. Serdons, M. Bex, L. Declercq, F.
Cleeren, M. Ooms, K. Van Laere, A. Verbruggen, G. Bormans, J. Pharm.
Biomed. Anal. 2015, 111, 209–214.
3
61329.
[
9] M. Pretze, D. Pietzsch, C. Mamat, Molecules 2013, 18, 8618–8665.
10] M. Namavari, Z. Cheng, R. Zhang, R. De, J. Levi, J K. Hoerner, S. S. Ya-
[
ghoubi, F. A. Syud, S. S. Gambhir, Bioconjugate Chem. 2009, 20, 432–
4
36.
[
[
[
[
11] X. G. Li, S. Dall’Angelo, L. F. Schweiger, M. Zanda, D. O’Hagan, Chem.
Commun. 2012, 48, 5247–5249.
12] S. Dall’Angelo, Q. Z. Zhang, I. Fleming, M. Piras, L. F. Schweiger, D.
O’Hagan, M. Zanda, Org. Biomol. Chem. 2013, 11, 4551–4558.
13] Z. Gao, V. Gouverneur, B. G. Davis, J. Am. Chem. Soc. 2013, 135, 13612–
[
29] X. Zhang, Z. Xiong, Y. Wu, W. Cai, J. R. Tseng, S. S. Gambhir, X. Chen, J.
Nucl. Med. 2006, 47, 113–121.
1
3615.
14] V. Bernard-Gauthier, C. Wängler, E. Schirrmacher, A. Kostikov, K. Ju-
rkschat, B. Wängler, R. Schirrmacher, Biomed Res. Int. 2014, 2014,
Received: March 22, 2016
4
54503.
Published online on July 4, 2016
Chem. Eur. J. 2016, 22, 10998 – 11004
www.chemeurj.org
11004
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim