18F-labeling of peptides by means of an organosilicon-based fluoride acceptor

Article abstract of DOI:10.1002/anie.200600795

(Chemical Equation Presented) F on: Fluorotriorganosilyl-derivatized Tyr³-octreotate was labeled with ¹⁸F^- providing the first practical formulation in ¹⁸F-radiochemistry for the labeling of a peptide (see schem

Full text of DOI:10.1002/anie.200600795

Angewandte
Chemie
Radioactive Peptides
undesired by-products.[1] The research over the last decade
clearly indicates that the success of PET in nuclear medicine
justifies the intense search for versatile labeling formulations
for the syntheses of 18F-radiopharmaceuticals. Especially the
development of rational 18F-labelling strategies for peptides,
until now characterized by multistep procedures, is consid-
ered to be one of the most important tasks.[4] As an alternative
to conventional 18F-labelling chemistry, the use of
[18F]fluorosilanes as labeling synthons was first proposed by
Rosenthal et al. who treated chlorotrimethylsilane with n.c.a.
(no carrier added) 18FÀ in aqueous acetonitrile isolating the
corresponding [18F]fluorosilane in 65% yield.[5] A preliminary
in vivo evaluation revealed fast hydrolysis of the compound
accompanied by high radioactivity (18F) uptake by the bone
making it unsuitable as a labeling synthon.
DOI: 10.1002/anie.200600795
18F-Labeling of Peptides by means of an
Organosilicon-Based Fluoride Acceptor**
Ralf Schirrmacher,* Gerrit Bradtmöller,
Esther Schirrmacher, Oliver Thews, Julia Tillmanns,
Thomas Siessmeier, Hans G. Buchholz,
Peter Bartenstein, Björn Wängler, Christof M. Niemeyer,
and Klaus Jurkschat*
Fluorine-18 is among the most commonly used radionuclides
for positron emission tomography (PET).[1] This non-invasive
imaging technique is capable of providing in vivo information
about the distribution of radiolabeled biomolecules by 1808
coincidence detection of two simultaneously emitted photons
from positron–electron annihilation. Although a number of
different radiotracers have been successfully employed in
PET, only a few, such as 2-[18F]fluoro-2-deoxy-d-glucose
(FDG)[2] and [18F]fluorodopa,[3] have gained widespread
application in nuclear medicine. The reason for this is that
the regioselective introduction of 18F into tracer molecules is
often non-specific and radiochemical yields (RCY) of the 18F-
labelled product are low. The introduction of 18FÀ into tracer
molecules requires high temperatures and often leads to
Another approach is based on the work by Pilcher et al.[6]
who fluorinated organosilanoles with nonradioactive HF in
high yields. An analogous labeling strategy was proposed in a
symposium abstract.[7] However, so far no labeling experi-
ments using aqueous 18FÀ/[18F]HF solutions have been
reported. Most recently Ting et al. used organotriethoxysi-
lanes as labeling precursors for the synthesis of
[18F]fluorosilanes but no practical application for the synthesis
of potential radiopharmaceuticals has been demonstrated.[8]
Herein we report the syntheses of substituted
[18F]organofluorosilanes using organochlorosilanes as label-
ing precursors and their in vitro and in vivo stability. As an
alternative labeling approach we also describe the 18F–19F
isotopic exchange using [19F]di-tert-butylphenyl fluorosilane
as a highly efficient silicon-based fluoride acceptor (SiFA
compound). As proof of applicability we transferred the SiFA
approach to the development of a simple and practical
formulation for the synthesis of a 18F-labelled SiFA derivat-
ized Tyr3-octreotate, a peptide used in oncology for the
visualization of neuro-endocrine tumors.[9]
[*] Dr. R. Schirrmacher,[+] Dr. E. Schirrmacher,[+] J. Tillmanns,
Dr. T. Siessmeier, Dipl.-Ing. H. G. Buchholz, Prof. Dr. P. Bartenstein
Klinik und Poliklinik für Nuklearmedizin
Johannes Gutenberg Universität Mainz
55131 Mainz (Germany)
Fax: (+49)6131-172386
We synthesized three [18F]organofluorosilanes, namely
[18F]fluorotriphenylsilane (1), [18F]fluoro-tert-butyldiphenyl-
silane (2), and [18F]fluorodi-tert-butylphenylsilane (3), and
evaluated their in vitro stability in human serum as well as
their in vivo stability in rats, studied by animal-PET. These
data are essential for finding the most suitable compound and
for evaluating the labeling concept.
E-mail: schirrmacher@klinik.nuklearmedizin.uni-mainz.de
Dipl.-Chem. G. Bradtmöller,[+] Prof. Dr. K. Jurkschat
Lehrstuhlfür Anorganische Chemie
Universität Dortmund
44221 Dortmund (Germany)
Fax: (+49)231-755-5048
E-mail: klaus.jurkschat@uni-dortmund.de
Dr. B. Wängler
The reaction in acetonitrile of the triorganochlorosilanes
(5–11.8 mmolmLÀ1) Ph3SiCl, tBuPh2SiCl, and tBu2PhSiCl,
with the azeotropically dried complex 18FÀ/Kryptofix2.2.2./
K+ at room temperature provided almost quantitatively the
corresponding [18F]triorganofluorosilanes 1–3 (Figure 1), as
demonstrated by means of radio-HPLC. Their identities were
confirmed by coelution of the radioactive probes spiked with
the related nonradioactive 19F-analogues. The specific activity
of 1, 2, and 3 was determined using UV-calibration curves and
was in the range 1500–1700 GBqmmolÀ1.
Abteilung für Radiopharmazeutische Chemie
Deutsches Krebsforschungszentrum Heidelberg (Germany)
Priv.-Doz. Dr. O. Thews
Institut für Physiology und Pathophysiologie
Universität Mainz (Germany)
Prof. Dr. C. M. Niemeyer
Lehrstuhlfür Biool gisch-Chemische Mikrostrukturtechnik
Universität Dortmund (Germany)
[+] These authors contributed equally.
[**] This work was supported by the InternationalIsotope Society
CentralEuropean Division. The authors thank Phiilps for support
and would like to thank Prof. Dr. H. J. Wester (TU Munich) for
providing a protocol for oxime ligation. C.M.N. and G.B. acknowl-
edge financialsupport from the Zentrum für Angewandte Chemi-
sche Genomik, a joint research initiative of the European Union and
the Ministry of Innovation and Research of North Rhine-Westphalia.
To investigate the applicability of the 18F-labelled com-
pounds for the development of Si–18F containing radio-
pharmaceuticals, their in vitro stability in human serum was
investigated (Figure 1). In agreement with previously pub-
lished data,[10,11] the [18F]triphenylfluorosilane 1 was found to
be stable for 4 h in neutral water (data not shown) but
unstable at pH 7.4–7.6 in human serum. In contrast, the tert-
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
Angew. Chem. Int. Ed. 2006, 45, 6047 –6050
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
6047

Communications
Figure 1. Top : 18F-fluorosilanes 1–3 for in vitro evaluation in human
serum and in vivo evaluation in rats by animal PET. Bottom: Only
compounds 2 and 3 are stable in human serum (37.48C, pH 7.4–7.6;
Figure 2. Top: Coronalmaximum intensity projections (MIP) of the
second frame (5–10 min after injection, a and c) and the last frame
(45–50 min after injection, b and d). The data reveala much higher
bone uptake for compound 2 (a, b) than for 3 (c, d). Bottom: Plot of
time activity in regions of interest in bone and liver for compound 2
&
!
*
*
=1, =2, =3). RCP=radiochemicalpurity.
butyl-substituted derivatives 2 and 3 displayed a high stability
in this biological matrix.
*
(
*
=lumbar vertebra, =liver) and 3 (& =lumbar vertebra, =liver).
cps=normalized counts per second.
Encouraged by these results, we performed the first
in vivo animal PET scans with compounds 2 and 3 in healthy
Sprague-Dawley rats to investigate the in vivo stability of the
organochlorosilanes as well as organoalkoxysilanes are sensi-
tive to hydrolysis, which limits complex syntheses.
Si 18F bond. To this end, 20–40 MBqof HPLC-purified 2 and
À
3, dissolved in physiological saline/10% ethanol, were
injected into the rats and the thoraco-abdominal area was
scanned using an animal PET scanner (Philips, Mosaic animal
imaging) for 50 min with time frames of 5 min. Because
unbound 18FÀ readily accumulates in bone (osteotropic), the
stability of the Si–18F bond was determined by measuring the
time curve of radioactivity uptake in bone (lumbar vertebra)
as a dynamic measurement of Si–18F hydrolysis. In contrast to
their rather similar in vitro stability (Figure 1), the in vivo
stability of 2 and 3 was found to be markedly different. As
shown on the PET scans, for the mono-tert-butyl-substituted
[18F]triorganofluorosilane 2, fast bone uptake of radioactivity
was detected, whereas in case of di-tert-butyl-substituted
compound 3, only a little uptake in late time frames was
observed (Figure 2).
Moreover, PET studies indicated that, probably owing to
their high lipophilicity, both compounds are metabolized
mainly by the liver. Because 2 is hydrolyzed in vivo to give
unbound 18FÀ, the main excretion of radioactivity occurred
through the bladder, in contrast to compound 3, which
displayed intestinal excretion (see Supporting Information).
To use organosilanes for 18F-labelling of radiopharmaceut-
icals, such as peptides for tumor imaging,[12] a suitable labeling
strategy must include fast coupling steps to cope with the half-
life of 110 min of 18F. Ideally, as in 99mTc-labelling chemistry,
no final HPLC purification step should be required for the
formulation of the final radiopharmaceutical. Unfortunately,
when applying triorganochlorosilanes as precursors for 18F-
labelling reactions, the resulting [18F]triorganofluorosilane
has to be separated from the triorganochlorosilane by means
of HPLC. This step is time consuming, leads to a decreased
RCY, and requires a trained radiochemist. Furthermore,
Another major problem in conventional 18F-radiochemis-
try is that 19F-compounds to be labeled with 18F by isotopic
exchange reactions are used in high concentrations (usually
10–100 mm) resulting in overall products of low specific
activity. Therefore, the application of the concept of isotopic
exchange to [18F]triorganofluorosilanes would only be mean-
ingful if 1) the concentration of the organofluorosilane is low
enough to guarantee a high specific activity, 2) the 18F–19F
exchange rate is sufficiently high, and 3) the isotopic
exchange reaction is not negatively influenced by various
functional groups (such as, NH2, COOH) usually present in
biological molecules, such as peptides. To circumvent these
difficulties and to reduce the labeling to only one reaction
step, preferably excluding HPLC purification, we investigated
the reaction of [19F]-tBu2PhSiF (4) with n.c.a. 18FÀ/Krypto-
fix2.2.2./K+ complex in acetonitrile (Figure 3).
The reaction in acetonitrile of compound 4 (1 mg, 4.1 nm)
with “dried” n.c.a. 18FÀ/Kryptofix2.2.2./K+ complex (1 GBq)
at room temperature in 100 mL acetonitrile without stirring
gave [18F]triorganofluorosilane 3 in yields between 80–95%
after only 15 min. The specific activity of 3 was in the range of
194–230 GBqmmolÀ1. This exchange reaction is remarkably
fast and exceeds the RCY of 50–70% usually obtained by
nucleophilic 18F–19F substitution of activated aromatic com-
pounds, which normally requires high reaction temperatures
and HPLC purification.[13,14] The specific activity obtained is
competitive with clinically used radiotracers, which show
specific activities between 100–1000 GBqmmolÀ1. As a result
of the very mild reaction conditions, a noteworthy advantage
of the SiFA method is the exclusive formation of product 3
(Figure 3). Moreover, the specific activity, which is of crucial
importance when imaging saturable neuroreceptors,[15] might
6048 www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 6047 –6050

Angewandte
Chemie
fix2.2.2./K+ complex (280–360 MBq) in 800 mL acetonitrile
for 10–15 min at room temperature yielding [18F]6 in yields
between 95–97% (Scheme 1). This reaction is the fastest,
Figure 3. Top: synthesis of 3 using 4 and 18FÀ/Kryptofix2.2.2/K+;
Bottom: radio-HPLC chromatogram of the crude reaction mix of 4
(1 mg (4.1 nm)) with18FÀ/Kryptofix2.2.2/K+ complex (1 GBq) in CH3CN
(100 mL) after 15 min reaction time at room temperature.
be further enhanced by starting with higher amounts of 18F
radioactivity.
Current conventional methods for the syntheses of 18F-
labelled peptides are based on multistep procedures involving
prosthetic groups suitable for bioconjugation, for example,
acylation, amidation, alkylation, photochemical conjuga-
tion,[4,12] and the recently introduced method of chemo-
selective oxime formation.[16,17] All these methods are com-
plex, which is a major drawback owing to the short half-life of
18F. Hence, in contrast to the radiolabeling of peptides with
radiometals,[4,12] no suitable robust practical formulation has
been described for the routine 18F-labeling of peptides. Ideally
the radiolabeling should proceed directly in the commercially
available 18FÀ/[18O]H2O solution or in dried 18FÀ/Krypto-
fix2.2.2./K+ complex in dipolar aprotic solvents. To date,
nucleophilic 18F-fluorination from aqueous solution has been
considered impossible because the fluoride ion is highly
solvated and thus has low nucleophilicity.
Scheme 1. Synthesis of [18F]6 by two different procedures. Method A: 6
(100 mg, 74 nmol), 18FÀ/Kryptofix2.2.2/K+ (280–360 MBq), CH3CN
(800 mL), room temperature, 10–15 min. Method B: 6 (100 mg,
74 nmol) in CH3CN (40 mL), 18FÀ/[18O]H2O (200–300 mL, 175–
250 MBq), 958C, 30 min.
most selective and mildest 18F-labelling chemistry docu-
mented to date. To remove the toxic Kryptofix2.2.2, potas-
sium ions and residual 18FÀ, the acetonitrile was diluted with
sodium dihydrogenphospate (0.25n, pH 4.5, 10 mL) and
passed through a C-18-solid-phase-extraction cartridge. The
radiolabeled peptide [18F]6 was retained on the cartridge
(trapping efficiency ꢀ 80%) and washed with water to reduce
the amount of residual 18FÀ to < 1%. To obtain [18F]6 as an
injectable solution, the cartridge was eluted with ethanol
(1 mL, elution efficiency ꢀ 80%) and diluted with isotonic
saline (9 mL). Notably, the entire experimental procedure
took only 25 min and the overall radiochemical yield was
between 55–65% giving [18F]6 in a purity of > 98%.
To circumvent these problems and to
utilize our labeling chemistry described
on potential radiopharmaceuticals, we syn-
thesized p-(di-tert-butylfluorosilyl) benzal-
dehyde, p-(tBu2FSi)C6H4C(O)H (5, see
Supporting Information) which is amena-
ble for chemoselective oxime ligation to aminooxy-function-
alized peptides.[16]
The organosilicon-functionalized aldehyde 5 was coupled
to an aminooxy-derivatized Tyr3-octreotate,[18] a peptide
which is used in nuclear medicine,[19] which was synthesized
by Fmoc solid-phase-peptide-synthesis (see Supporting Infor-
mation; Fmoc = (9H-fluoren-9-ylmethoxy)carbonyl) and
coupled to compound 5 by oxime formation in acetonitrile/
water (90:10) at pH 4, yielding peptide 6 in > 98% purity
after HPLC purification. In a first isotopic exchange experi-
ment we treated 6 (100 mg, 74 nmol) of with 18FÀ/Krypto-
To evaluate whether an isotopic exchange also occurs in
commercially available 18FÀ/[18O]H2O, we treated a solution
of
6 (100 mg, 74 mmol) in acetonitrile (40 mL) with
18FÀ/[18O]H2O (200 mL, 175–250 MBq) at room temperature
for 15 min. Only a 5% yield of [18F]6 was detected by radio-
HPLC. However, when conducting this reaction at 958C, the
desired compound was obtained in yields between 70–90%
(HPLC) within 30 min (Scheme 1). No hydrolysis of the
Si 18F bond and no disintegration of peptide 6 was observed
À
Angew. Chem. Int. Ed. 2006, 45, 6047 –6050
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 6049

Communications
[5] M. S. Rosenthal, A. L. Bosch, R. J. Nickles, S. J. Gatley, Int. J.
Appl. Radiat. Isot. 1985, 36, 318.
[6] A. S. Pilcher, H. L. Ammon, P. DeSong, J. Am. Chem. Soc. 1995,
117, 5166.
[7] J. C. Walsh, K. M. Akhoon, N. Satyamurthy, J. R. Barrio, M. E.
Phelps, S. S. Gambhir, T. Toyokuni, J. Labelled Compd. Radio-
pharm. 1999, 42, S1.
[8] R. Ting, M. J. Adam, T. J. Ruth, D. M. Perrin, J. Am. Chem. Soc.
2005, 127, 13094.
during HPLC quality control. Purification was performed in a
similar manner as described above by solid-phase extraction
using water (5 mL) for dilution. The calculated specific
activity of [18F]6 for both methods was 3–5 GBqmmolÀ1
which is sufficient for in vivo PET studies. We anticipate,
however, that even higher specific activities should be
obtainable by using larger amounts of 18FÀ for the isotopic
exchange.
[9] M. de Jong, W. A. P. Breeman, W. H. Bakker, P. P. M. Kooij, B. F.
Bernard, L. J. Hofland, T. J. Visser, A. Srinivasan, M. A.
Schmidt, J. L. Erion, J. E. Bugaj, H. R. Mäcke, E. P. Krenning,
Cancer Res. 1998, 58, 437.
[10] C. G. Swain, R. M. Esteve, R. H. Jones, J. Am. Chem. Soc. 1949,
71, 965.
[11] H. Szmant, G. A. Brost, J. Am. Chem. Soc. 1950, 72, 5763.
[12] “Radiolabeled Peptides for Tumor Imaging”: L. C. Knight in
Handbook of Radiopharmaceuticals, Radiochemistry and Appli-
cations (Eds.: M. Welch, C. S. Redvanly), Wiley, Chichester,
2003.
[13] R. Bolton, J. Labelled Compd. Radiopharm. 2002, 45, 485.
[14] M. S. Haka, M. R. Kilbourn, G. L. Watkins, S. A. Toorongian, J.
Labelled Compd. Radiopharm. 1989, 27, 823.
[15] “Neurological Applications”: N. I. Bohnen in Principles and
Practise of Positron Emission Tomography (Eds.: R. L. Wahl,
J. W. Buchanan), Lippincott Williams & Wilkins, Philadelphia,
2002, and references therein.
In conclusion, we have described a rapid and versatile
approach for the synthesis of 18F-labeled peptides as illus-
trated by the example of a SiFA-derivatized Tyr3-octreotate.
This compound was labeled with 18FÀ/Kryptofix2.2.2./K+
complex in acetonitrile at room temperature as well as in
aqueous 18FÀ/[18O]H2O solution at 958C. We anticipate that
the very mild reaction conditions and the fast and efficient
labeling make the SiFA strategy a valuable tool for the
development of 18F-radiopharmaceuticals. The in vivo evalu-
ation of compound [18F]6 as a tumor imaging agent is
currently under investigation.
Experimental Section
Syntheses of the organosilanes tBu2PhSiX (X = H,[20] F,[21] Cl,[20] I[21]
)
and tBuPh2SiF[22] have been described elsewhere, however, different
procedures were applied herein and all details are given in the
Supporting Information, together with the synthetic procedures for
p-(di-tert-butylfluorosilyl) benzaldehyde (5) and peptide 6. Note-
worthy, in contrast to earlier reports tBu2PhSiF (4) was obtained as
crystalline material and its molecular structure was determined by
single-crystal X-ray diffraction.[23]
[16] G. Thumshirn, U. Hersel, S. L. Goodman, H. Kessler, Chem. Eur.
J. 2003, 9, 2717.
[17] T. Poethko, M. Schottelius, G. Thumshirn, U. Hersel, M. Herz, G.
Henriksen, H. Kessler, M. Schwaiger, H. J. Wester, J. Nucl. Med.
2004, 45, 892.
[18] Tyr3-octreotate was synthesized by Fmoc solid-phase peptide
synthesis according to a published procedure (E. Schirrmacher,
R. Schirrmacher, C. Beck, W. Mier, N. Trautman, F. Rꢀsch,
Tetrahedron Lett. 2003, 44, 9143).
[18F]6: Method A: Compound 6 (100 mg, 74 nmol) was dissolved
in acetonitrile and 18FÀ/Kryptofix2.2.2./K+ complex (280–360 MBq) in
acetonitrile was added to obtain a total volume of 800 mL. The
solution was kept at room temperature without stirring for 10–15 min,
diluted with sodium dihydrogenphospate (0.25n, pH 4.5, 10 mL) and
passed through a C-18-Sepak cartridge (Merck). The cartridge was
washed with water (2 mL) and eluted with ethanol (1 mL) to obtain
[18F]6 (160–240 MBq) as an injectable solution.
[19] C. R. Reubi, M. Gugger, B. Waser, Eur. J. Nucl. Med. 2002, 29,
855.
[20] T. Kusukawa, W. Ando, J. Organomet. Chem. 1998, 559, 11.
[21] H.-W. Lerner, S. Scholz, M. Bolte, Z. Anorg. Allg. Chem. 2001,
627, 1638.
[22] R. Damrauer, R. A. Simon, B. Kanner, Organometallics 1988, 7,
1161.
[23] G. Bradtmꢀller, K. Jurkschat, M. Schꢁrmann, Acta Crystallogr.
Sect. E 2006, 62, 1303.
[18F]6: Method B: Compound 6 (100 mg, 74 nmol) dissolved in
acetonitrile (40 mL) was added to 18FÀ/[18O]H2O (200–300 mL; 175–
250 MBq; purchased from PetNet Erlangen) and heated at 95 8C for
30 min in a sealed reaction vial. Water (5 mL) was added and the
solution was passed through a C-18-SepPac cartridge (Merck). The
cartridge was washed with water (2 mL) and eluted with ethanol
(1 mL) to obtain [18F]6 (95–150 MBq) as an injectable solution.
Received: March 1, 2006
Revised: June 13, 2006
Published online: August 3, 2006
Keywords: fluorination · isotopic labeling · organofluorosilanes ·
.
peptides
[1] “Chemistry of Fluorine-18 Radiopharmaceuticals”: S. E. Snyder,
M. Kilbourn in Handbook of Radiopharmaceuticals, Radio-
chemistry and Applications (Eds.: M. Welch, C. S. Redvanly),
Wiley, Chichester, 2003.
[2] T. Ido, C.-N. Wan, J. S. Fowler, A. P. Wolf, J. Org. Chem. 1977, 42,
2341.
[3] G. Firnau, C. Nahmias, E. S. Garnett, Int. J. Appl. Radiat. Isot.
1973, 24, 182.
[4] P. Blower, Dalton Trans. 2006, 1705.
6050 www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 6047 –6050