Single-step radiofluorination of peptides using continuous flow microreactor

Article abstract of DOI:10.1039/c2ob00015f

¹⁸F radiolabelling of peptides bearing two different prosthetic groups was successfully conducted in a continuous flow microfluidic device for the first time. Radiochemical yields were dependent on precursor concentration, reaction temperature and flow rate. The choice of leaving group had a dramatic influence on the reaction outcome. Rapid reaction optimization was possible.

Full text of DOI:10.1039/c2ob00015f

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Organic &
Biomolecular
Chemistry
Cite this: Org. Biomol. Chem., 2012, 10, 3871
www.rsc.org/obc
PAPER
Single-step radiofluorination of peptides using continuous flow microreactor†
Svetlana V. Selivanova,a Linjing Mu,a Johanna Ungersboeck,b Timo Stellfeld,c Simon M. Ametamey,*a
Roger Schiblia and Wolfgang Wadsak*b
Received 3rd January 2012, Accepted 23rd March 2012
DOI: 10.1039/c2ob00015f
18F radiolabelling of peptides bearing two different prosthetic groups was successfully conducted in a
continuous flow microfluidic device for the first time. Radiochemical yields were dependent on precursor
concentration, reaction temperature and flow rate. The choice of leaving group had a dramatic influence
on the reaction outcome. Rapid reaction optimization was possible.
Introduction
Experimental†
18F-labelled peptides are used as radioactive imaging agents for
positron emission tomography (PET), an in vivo imaging tech-
nique that has gained broad application in nuclear and molecular
medicine and in drug development during the last decade.1–4
The conventional synthesis of 18F-peptides by direct radiofluori-
nation requires heating in basic conditions at elevated tempera-
tures resulting often in decomposition of the starting material
(precursor) and/or the product.5 Another disadvantage is that a
large excess of the peptidic precursor (milligrams) is required
to obtain the radiolabelled product (micrograms) in high radio-
chemical yields (RCY). Therefore, purification of the final
product is necessary by time-consuming high performance liquid
chromatography (HPLC) in order to remove the excess precursor.
The short half-life of 18F radioisotope (109.7 minutes) poses a
constraint on the acceptable synthesis time. It has been
suggested, that microfluidic devices could bring multiple advan-
tages to radiopharmaceutical tracer development.6 Several
groups reported the radiolabelling of a variety of small molecules
with short-lived radioisotopes, such as 18F and 11C, in a
microreactor.7–11 Herein we investigated if a microfluidic
device could be used for the radiolabelling of peptides with 18F
using small amounts of precursor and preferably under mild
conditions.
Peptidic precursors 1–3 (Fig. 1) and their corresponding non-
radioactive fluorinated compounds (reference compounds) were
generously provided by Bayer HealthCare (Berlin, Germany).
No-carrier-added 18F-fluoride was produced via the
18O(p,n)18F nuclear reaction by irradiation of isotopically
enriched 18O-water in a fixed-energy Cyclone 18/9 cyclotron
(IBA).12 Dried 18F-fluoride–cryptate complex was prepared
using standard separation and azeotropic drying procedure in the
presence of Kryptofix 2.2.2 and potassium carbonate.10
The radiolabelling reactions were performed using the
NanoTek continuous flow system (Advion BioSciences). The
reactions were conducted in DMSO and were optimized by
varying reaction temperature, precursor concentration, reagent
ratio, and flow rate (residence time). Each reaction condition was
tested at least three times. On exit from the microreactor the reac-
tion mixtures were quenched with aqueous trifluoroacetic acid
(0.1%) and analysed using ultra performance liquid chromato-
graphy (UPLC) and thin layer chromatography (TLC) to deter-
mine radiolabelling efficiency.‡ Product identity was confirmed
by co-elution with the nonradioactive reference compound.
Results and discussion
Three peptides, bombesin derivatives of 7–8 amino acids, conju-
gated to a prosthetic group containing trimethylammonium
aCenter for Radiopharmaceutical Sciences, ETH Zurich, IPW HCI
H427, Wolfgang-Pauli-Strasse 10, 8093 Zurich, Switzerland.
E-mail: simon.ametamey@pharma.ethz.ch; Fax: +41 44 633 1367;
Tel: +41 44 633 7463
‡All commercially available reagents and solvents were used as
received. For liquid chromatography Waters Acquity UPLC system con-
nected in line with a FlowStar LB 513 radiodetector (Berthold Technol-
ogies) was used. The analytical column was a UPLC dedicated reversed-
phase Acquity BEH C18, particle size 1.7 μm, 100 × 2.1 mm (Waters).
The mobile phase consisted of a gradient of water or aqueous buffer and
acetonitrile. The gradient was developed for each peptide. Reversed-
phase TLC plates were Alugram RP-18W (Macherey-Nagel). The
mobile phase was a mixture of 50 mM phosphate buffer (pH 7.4) with
acetonitrile 3 : 7. Developed TLCs were visualized with InstantImager
(Canberra Packard).
bRadiochemistry and Biomarker Development Unit, Department of
Nuclear Medicine, Medical University of Vienna, Waehringer Guertel
18-20, A-1090 Vienna, Austria. E-mail: wolfgang.wadsak@meduniwien.
ac.at; Fax: +43 1 40400 1557; Tel: +43 1 40400 5255
cBayer HealthCare, Global Drug Discovery, Berlin, Germany
†Electronic supplementary information (ESI) available: Representative
UPLC and TLC chromatograms of a reaction mixture, comparison chart
of selected analytical data obtained using UPLC and TLC, and an
example of product purification procedure using a cartridge. See DOI:
10.1039/c2ob00015f
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 3871–3874 | 3871

View Article Online
Fig. 1 Chemical structures of precursors 1–3.
Scheme 1 Radiolabelling of peptides with nucleophilic 18F.
leaving group or triarylsulfonium moiety (Fig. 1) were radio-
labelled with 18F in the microreactor (Scheme 1). No side pro-
ducts were detected in the reaction mixture for all described
peptides and reaction conditions.
As reported earlier, the trimethylammonium leaving group can
be replaced by 18F under conventional heating in a vial in
DMSO at 70–95 °C for 10–15 minutes and results in ∼5–20%
isolated decay corrected (d.c.) RCY after HPLC purification.13
The optimal temperature in the microreactor was 70–80 °C yield-
ing labelling efficiency up to 90% (estimated by UPLC§)
Fig. 2 Influence of the reaction temperature on the labelling efficiency,
estimated by UPLC. All reactions were performed at a flow rate of 20 μl
min−1. Precursor concentrations were 0.5 mg ml−1 for precursor 1,
1.0 mg ml−1 for precursor 2, and 4 mg ml−1 for precursor 3. The stan-
dard deviation range is inside the symbol area if it is not visible.
(Fig. 2). More reproducible labelling results were obtained in
this temperature range as well. The reaction progressed even at
35 °C resulting in 30–40% labelling yield. This is an important
finding since it opens an opportunity to label thermally labile
peptide molecules with 18F radioisotope under very mild
conditions.
Moreover, at such low temperatures it is very unlikely for the
radiolabelled peptide to lose its biological activity due to racemi-
zation when heated in a basic environment. The reaction
efficiency dropped significantly when the reaction temperature
was raised over 90 °C, probably due to peptide degradation.
The reaction efficiency increased with higher concentration
of the starting material and reached a plateau. Accounting for
the fact that such precursors are often in limited availability,
§UPLC was used for the purpose of estimation of the reaction efficiency.
For this, an aliquot of each quenched reaction mixture solution was
injected into UPLC. The obtained chromatograms were processed using
dedicated UPLC Empower software. For each chromatogram all radio-
active peaks were integrated and the percentage of the corresponding
product peak was plotted on graphs as a function of reaction conditions
(Fig. 2–4). Normally, these estimated values are higher than the radio-
chemical yield of the isolated product. Trends in radiolabelling reaction
efficiency estimated by TLC were in accordance with the UPLC data,
although the values (reaction efficiency, %) were somewhat lower (see
ESI;† Comparison of analytical data obtained using UPLC and TLC for
peptide 2). The percentage yields obtained (radiolabelling reaction
efficiency) are relative and were used to compare and select the best
reaction conditions. These are not absolute isolated RCYs.
3872 | Org. Biomol. Chem., 2012, 10, 3871–3874
This journal is © The Royal Society of Chemistry 2012

View Article Online
Fig. 3 Influence of the precursor concentration on the labelling
efficiency, estimated by UPLC. Reactions were performed at 80 °C for
precursor 1, at 50 °C for precursor 2, and at 160 °C for precursor 3.
Flow rate was 20 μl min−1
.
reactions could be performed even at a concentration of 0.5 mg
ml−1 with reasonable yields (Fig. 3). For example, we radio-
labelled precursor 1 at a concentration of 0.5 mg ml−1, 80 °C,
and 50 μl min−1 flow rate with 61% radiolabelling efficiency as
estimated by UPLC. Purification of the reaction mixture on a
SepPak C18 Light cartridge (Waters) resulted in 43% isolated
d.c. RCY. Flow rates between 10 and 50 μl min−1 had no influence
on reaction efficiency for precursor 1 (Fig. 4). For precursor 2,
an increase in radiolabelling yield with slower flow rates was
more noticeable at lower reaction temperatures (Fig. 4). Slower
flow rate results in longer reaction time. It should be noted that
potassium carbonate was used as a base during reactions in the
microreactor. The use of potassium carbonate gave negligible
yields in conventional radiosyntheses; therefore caesium carbon-
ate was employed instead. A difference in solubility may play
a role: caesium carbonate is more soluble in organic solvents
than potassium carbonate. Usually dry water-free solvents are
required for nucleophilic radiofluorination reactions. It may be
that potassium carbonate remains mostly undissolved during
conventional heating while microreactor conditions, usually
characterized by high surface area to volume ratio of reactants,
may improve its chance for interaction, thus giving better yields
than in a vial-based radiosynthesis.
Fig. 4 Influence of the flow rate on the labelling efficiency, estimated
by UPLC. Precursor concentrations were 0.5 mg ml−1 for precursor 1,
1.0 mg ml−1 for precursor 2, and 4 mg ml−1 for precursor 3.
It was found that harsh conditions were required for the direct
radiolabelling of precursor 3, containing the diarylsulfonium
leaving group, if potassium carbonate is used as a base. This is
also the case in conventional radiosynthesis (unpublished data).
The radiolabelling was negligible at temperatures below 130 °C
(Fig. 2). The optimal temperature was established at
170–180 °C, giving about 70% labelling efficiency at 20 μl
min−1 flow rate and 4 mg ml−1 precursor concentration. The pre-
cursor concentration has a much more pronounced influence on
the reaction outcome for the diarylsulfonium than for the tri-
methylammonium leaving group (Fig. 3). The reaction efficiency
increased more than 7-fold when the concentration of the precur-
sor was doubled from 1 to 2 mg ml−1. At 4 mg ml−1 the radiola-
belling yield still rose significantly but further increase of the
precursor concentration resulted in a reduction in labelling
efficiency. So far, we have no explanation for this phenomenon.
The influence of the flow rate was similar to the other two pre-
cursors and was more noticeable at 160 °C and below (Fig. 4).
Radiolabelled compound 18F-3 could be reliably produced in
22–28% (n = 5) isolated d.c. RCY at optimized conditions
(170 °C, 20 μl min−1 flow rate, 4 mg ml−1 precursor
concentration).
Conclusions
Direct radiolabelling of peptides with 18F-fluoride can be suc-
cessfully achieved in a reproducible manner using the Advion
NanoTek continuous flow microreactor. Fast synthesis time and
general module setup allows for rapid optimisation of the reac-
tion conditions and multiple on-demand productions of a given
radiolabelled peptide. The amount of precursor required for the
radiolabelling can be substantially decreased, in particular for
peptides having trimethylammonium leaving group, which facili-
tates purification and helps to save precious starting material.
Incorporation of a cartridge purification system and analytical
HPLC into the module design would be beneficial.
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 3871–3874 | 3873

View Article Online
6 A. M. Elizarov, Lab Chip, 2009, 9, 1326.
7 S.-Y. Lu, P. Watts, F. T. Chin, J. Hong, J. L. Musachio, E. Briarda and
V. W. Pike, Lab Chip, 2004, 4, 523.
8 J.-H. Chun, S. Lu, Y.-S. Lee and V. W. Pike, J. Org. Chem., 2010, 75,
3332.
9 V. R. Bouvet, M. Wuest, L. I. Wiebe and F. Wuest, Nucl. Med. Biol.,
2011, 38, 235.
Acknowledgements
We thank Dr Lee Collier from Advion Biosciences Inc. (Ithaca,
NY, USA) for continuous support during set-up and testing of
the NanoTek synthesizer.
10 J. Ungersboeck, C. Philippe, L.-K. Mien, D. Haeusler, K. Shanab,
R. Lanzenberger, H. Spreitzer, B. K. Keppler, R. Dudczak, K. Kletter,
M. Mitterhauser and W. Wadsak, Nucl. Med. Biol., 2011, 38, 427.
11 S. Kealey, C. Plisson, T. L. Collier, N. J. Long, S. M. Husband,
L. Martarello and A. D. Gee, Org. Biomol. Chem., 2011, 9, 3313.
12 IBA movie: http://www.youtube.com/watch?v=6BxyqFK2KRI
13 L. Mu, M. Honer, J. Becaud, M. Martic, P. A. Schubiger,
S. M. Ametamey, K. Graham, S. Borkovski, L. Lehmann,
L. M. Dinkelborg and A. Srinivasan, Bioconjugate Chem., 2010, 21,
1864.
Notes and references
1 M. E. Phelps, J. Nucl. Med., 2000, 41, 661.
2 K. Serdons, A. Verbruggen and G. M. Bormans, Methods, 2009, 48, 104.
3 S. M. Ametamey, M. Honer and P. A. Schubiger, Chem. Rev., 2008, 108,
1501.
4 W. C. Eckelman, R. C. Reba and G. J. Kelloff, Drug Discovery Today,
2008, 13, 748.
5 S. M. Okarvi, Eur. J. Nucl. Med. Mol. Imaging, 2001, 28, 929.
3874 | Org. Biomol. Chem., 2012, 10, 3871–3874
This journal is © The Royal Society of Chemistry 2012