Solid phase synthesis of [18F]labelled peptides for positron emission tomography

Article abstract of DOI:10.1016/S0960-894X(00)00179-7

A strategy for the solid phase synthesis of [¹⁸F]labelled peptides has been developed. The peptides were prepared on solid support and acylated with 4-[¹⁸F]fluorobenzoic acid using HATU within 3 min and the labelled peptide was relea

Full text of DOI:10.1016/S0960-894X(00)00179-7

Bioorganic & Medicinal Chemistry Letters 10 (2000) 1501±1503
Solid Phase Synthesis of [18F]Labelled Peptides for Positron
Emission Tomography
Julie L. Sutcli€e-Goulden,b Michael J. O'Dohertyb and Sukhvinder S. Bansala,*
aDepartment of Pharmacy, Kings College London, Franklin-Wilkins Building, 150 Stamford Street, London SE1 8WA, UK
bClinical PET Centre, Guy's, King's and St Thomas' Hospital, Lambeth Palace Road, London SE1 7EH, UK
Received 23 December 1999; accepted 17 March 2000
AbstractÐA strategy for the solid phase synthesis of [18F]labelled peptides has been developed. The peptides were prepared on solid
support and acylated with 4-[18F]¯uorobenzoic acid using HATU within 3 min and the labelled peptide was released from the solid
support within 7 min. The [18F]labelled peptides were produced in good purity with a speci®c activity of 20±25 GBq/mmol. # 2000
Elsevier Science Ltd. All rights reserved.
At present [18F] has been incorporated into many bio-
logical molecules including peptides,1 4 proteins5,6 and
nucleotides7 using a prosthetic group approach for
imaging studies. Peptides are usually labelled in solution
of arginine was protected with the 2,2,4,6,7-penta-
methyldihydrobenzofuran-5-sulphonyl (Pbf) group14,15
and the E-amino group of lysine protected with a 4-
methyltrityl (Mtt) group.16
by treatment with an activated [18F]¯uorobenzoate
derivative.8
With peptides containing more than one
The synthesis of the peptides was achieved by Fmoc-
based solid phase techniques from C!N; stepwise
synthesis was carried out on a polyethylene glycol-
polystyrene (PEG-PS) support derivatised with a xan-
then-2-oxovalerate (XAL) linker.17 Fmoc deprotection
was achieved by the treatment with 20% piperidine/
DMF. Fmoc amino acids were used in 4-fold excess and
activation was carried out using [(1H-benzotriazol-1-yl)
(dimethylamino)methylene] - N - methylmethanaminium
tetra¯uoroborate-N-oxide (TBTU)18 in the presence of
diisopropylethylamine (DIPEA). At the completion of the
synthesis, a small sample of the peptide was cleaved from
the support by incubation with 95% tri¯uoroacetic acid
(TFA) containing water, phenol and triisopropylsilane
(TIPS) as carbocation scavengers.19 The cleavage product
was assessed to be greater than 95% pure by analytical
RP-HPLC and covalent structure was con®rmed by FAB
and MALDI-TOF MS.
11
basic group, either these basic groups are replaced4 or
the di€erence in pKa between the a-amino and the e-
amino is utilised to produce the [18F]labelled peptide.12
In these procedures, the reaction results in mixtures of
labelled, unlabelled and multiply labelled compounds,
along with other impurities such as excess reagents and
by-products and requires extensive puri®cation.
Owing to these diculties with solution phase synthesis,
we decided to develop a new route to the synthesis of
[18F]labelled peptides using a solid phase approach.
Model peptides were designed to investigate the solid
phase strategy for the [18F]labelling of peptides. In the
design criterion cysteine and tryptophan were both
omitted since they are susceptible to oxidation. The
peptide sequence contained most other naturally occur-
ring amino acids and contained a RGD motif, H-
KPQVTRGDVFTM-NH2. At the onset, we decided to
use hyperacid labile protection where possible, since
these groups would be rapidly deprotected. The side
chains of the amino acids were protected with trityl based
protecting groups13 where possible, the guanidine group
Before proceeding with radiotracer synthesis, `cold' reac-
tions of 4-¯uorobenzoic acid with the H-K(Mtt)PQ (Trt)
VT(Trt)R(Pbf)GD(tBu)VFT(Trt)M-XAL-PEG-PS (pep-
tide-XAL-PEG-PS) were investigated in order to estab-
lish the optimal reaction conditions for the radiotracer
synthesis. Experiments to optimise product yields were
carried out on the reaction of 4-¯uorobenzoic acid and
the peptide-XAL-PEG-PS using a range of activating
agents (Table 1).
*Corresponding author. Tel.: +44-171-333-4785; fax: +44-020-7848-
4785; e-mail: sukhi.bansal@kcl.ac.uk
0960-894X/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(00)00179-7

1502
J. L. Sutcli€e-Goulden et al. / Bioorg. Med. Chem. Lett. 10 (2000) 1501±1503
Table 1. Percentage yields for the coupling of the 4-¯uoro-
benzoylpeptidea
Table 2. Percentage yields for the cleavage of the 4-¯uoro-
benzoylpeptidea
Coupling
Time (min)
Cleavage Conditions
Time (min)
2
3
10
44
>95
57
20
2
5
10
20
DIPCDI/HOBt
TBTU
DIPCDI/HOAt
HATU
16
82
33
87
24
87
37
>95
66
>95
74
TFA:Phenol:H2O: TIPS, RT
TFA:Phenol:H2O:TIPS, 37 ꢀC
Ð
24
Ð
59
33
>95
65
>95
>95
>95
aCalculated by integration of the peaks corresponding to the 4-¯uor-
obenzoyl peptide and the parent peptide in HPLC of the crude pep-
tides directly after extraction with ether and water of the cleavage
reagent TFA:H2O:Phenol:TIPS (90:4:5:1).
aCalculated by integration of the peaks corresponding to the 4-¯uoro-
benzoyl peptide and the parent peptide in HPLC of the crude peptides
directly after extraction with water and ether of the cleavage reagent
TFA: H2O: Phenol: TIPS (90:4:5:1).
4-[18F]Fluorobenzoic acid 3 was synthesised using a mod-
i®cation of published procedures25 in a relatively simple
three step procedure (Scheme 1) involving methylation of
ethyl 4-(dimethylamino) benzoate 1, nucleophilic dis-
placements of the ammonium group 2 at 90ꢀC within 3
min using the powerful nucleophilic radio¯uorinating
agent, [18F] F K+ 4,7,13,16,21,24-hexaoxa-1,10-diazabi-
cyclo[8.8.8] hexacosane (APE 2.2.2) and saponi®cation of
the ester moiety at 90ꢀC within 3 min followed by SPE
puri®cation. The [18F]¯uoride was produced using a RDS
112 negative ion cyclotron by the 18O(p, n)18F reaction
on 18O-enriched water held in a silver target with 11.2
MeV proton beam. The [18F]¯ouride was treated with
APE 2.2.2 and potassium carbonate in acetonitrile and
then azeotropically dried at 120ꢀC using three aliquots
of acetonitrile.
Time-course observations of the product after cleavage
from the resin showed that more than 65% of the reac-
tion occurred during the ®rst 20 min and the reaction had
reached greater than 95% using the aminium activating
species within 10 min. N-[(dimethylamino)-1H-1,2,3-
triazolo[4,5,b]pyridin-1-ylmethylene]-N-methylmethan-
aminium hexa¯uorophosphate-N-oxide (HATU)20 was
found to be particularly e€ective and the reaction had
reached 95% within 3 min. HATU and diisopropyl-
carbodiimide (DIPCDI)/1-hydroxy-7-azabenzotriazole
(HOAt)21 were respectively more e€ective than TBTU
and DIPCDI/1-hydroxybenzotriazole (HOBt) because
they incorporate into their structure a nitrogen atom
strategically placed at the 7 position of the aromatic
ring.20,21 Incorporation of this nitrogen in the benzene
ring has two consequences. First, the electron with-
drawing in¯uence of the nitrogen e€ects the stabilisation
of the leaving group, leading to greater reactivity. Sec-
ond, placement speci®cally at the seven position results
in classic neighbouring participation. Although HATU
was initially assigned a uronium-type structure, more
recently, the structure has been determined by X-ray
analysis and HATU crystallises as an aminium salt
rather than the corresponding uronium salt.22
The synthesis of the radiotracer was accomplished in
two rapid steps involving the labelling of the peptide-
XAL-PEG-PS with 4-[18F]¯uorobenzoic acid using
HATU within 3 min followed by cleavage of the radio-
tracer peptide 5 from the solid support within 7 min. The
[18F]labelled peptide 6 was analysed by radio HPLC. In
the labelling reaction the reagents and byproducts were
Experiments to optimise cleavage of the 4-FC6H4CO-
KPQVTRGDVFTM-NH2 (4-¯uorobenzoylpeptide) from
the solid support were also carried out. Time course
studies on the cleavage of the ¯uorobenzoylpeptide
indicated that the cleavage of peptide was relatively
slow (ca. 4h) and this was attributed to the dicult
removal of the trityl side chain protection of glutamine
as determined by MS. Since the Gln (Trt) containing
peptides are sometimes cleaved slowly23 the peptide
sequence was re-synthesised without protection of the
glutamine side chain.
The carboxyl group was preactivated as the penta-
¯uorophenyl ester (Fmoc-Gln-OPfp)24 since the action
of activating agents on unprotected glutamine results in
dehydration. Using this modi®cation the peptide was
cleaved from the resin more eciently, however the
cleavage time remained high (Table 2). The use of dif-
ferent scavengers and temperatures were investigated
and optimal cleavage of the ¯uorobenzoylpeptide as
determined by HPLC was achieved within 7 min at
37 ꢀC. Over the short time period, the use of elevated
temperatures did not appear to have a detrimental e€ect
on the integrity of the peptide.
Scheme 1. Solid phase synthesis of [18F] labelled peptides.

J. L. Sutcli€e-Goulden et al. / Bioorg. Med. Chem. Lett. 10 (2000) 1501±1503
1503
removed by simple ®ltration, and in the cleavage the
Acknowledgements
radiotracer was obtained by evaporation of the acid. In this
case, the radiochemical purity of the peptide was >95%
however; in some cases, puri®cation may be required.
We thank the Guy's and St Thomas' Trust for ®nancial
support and to Dr A. E. Theobald and Professor R. C.
Hider for useful comments.
In an attempt to ®nd a more ecient, selective and ver-
satile approach for the [18F]labelling of synthetic pep-
tides a solid phase strategy was developed. This strategy
o€ers a number of advantages over the present peptide
labelling methods in the solution phase. The peptide-
XAL-PEG-PS has been eciently labelled with 4-
[18F]¯uorobenzoic acid within 3 min using HATU as the
activating agent and the [18F]labelled peptide was released
from the solid support within 7 min with TFA with an
overall radiochemical labelling yield of (70±80%). This
percetage yield is based on the [18F] 4-¯ourobenzoic acid
and it is at the end of the synthesis, not decay corrected,
n=20, SEM 10%. During the release of the peptide from
the solid support, scrambling of [18F] with TFA was not
observed. The solid phase strategy is very ecient and the
[18F]labelling was complete within 20 min, excluding the
synthesis of the [18F]F K+ APE 2.2.2. In the solid
phase approach the peptide chemistry was optimised
and all reactions are expected to proceed in 99.99%
yields. Prior to the labelling with the 4-¯uorobenzoic
acid, only one reactive site was made available and
hence the process is completely speci®c and highly
reproducible. In this case, the a-amino group was
deprotected, similarly the E-amino group of lysine can
be deprotected using allyl chemistry and other groups
such as carboxyl or hydroxyl could be deprotected as
required. This approach is readily adapted to the
synthesis of cyclic peptides and peptides containing
disulphide bridges and hence this approach is extremely
versatile. This methodology can also be adapted to Boc-
benzyl chemistry although some of the protecting
groups will have to be changed. Whilst solution phase
chemistry requires time consuming extractions and pur-
i®cation, in the solid phase approach the reagents and
byproducts are removed by simple ®ltration. In this
case, the peptide was very clean and did not require pur-
i®cation, it is possible to release the peptide from the solid
support, evaporate the TFA and inject the crude peptide
directly into a HPLC column for puri®cation. 10±50MBq
of [18F] ¯ouride was used and the speci®c activity of the
peptide was 20±25 GBq/mmol. As with solid phase peptide
synthesis and solid phase combinatorial organic synthesis,
this process may be readily automated and many instru-
ments are commercially available.
References
1. Vaidyanathan, G.; Zalutsky, M. R. Nuclear Med. Biol.
1997, 24, 171.
2. Vaidyanathan, G.; Zalutsky, M. R. Nuclear Med. Biol.
1995, 22, 759.
3. Guhlke, S.; Wester, H. J.; Bruns, C.; Stocklin, G. Nuclear
Med. Biol. 1994, 21, 819.
4. Moody, T. W.; Leyton, J.; Unsworth, E.; John, C.; Lang,
L.; Eckelman, W. C. Peptides 1998, 19, 585.
5. Vaidyanathan, G.; Bigner, D. D.; Zalutsky, M. R. J.
Nuclear Med. 1992, 33, 1533.
6. Wester, H. J.; Hamacher, K.; Stocklin, G. Nuclear Med.
Biol. 1996, 23, 365.
7. Dolle, F.; Hinnen, F.; Vaufrey, F.; Tavitian, B.; Crouzel, C.
J. Lab. Comp. Radiopharm. 1997, 39, 319.
8. Lang, L.; Eckelman, W. C. Appl. Radiat. Isot. 1994, 44,
1085.
9. Lang, L.; Eckelman, W. C. Appl. Radiat. Isot. 1997, 48,
169.
10. Guhlke, S.; Coenen, H. H.; Stocklin, G. Appl. Radiat. Isot.
1994, 45, 715.
11. Vaidyanathan, G.; Zalutsky, M. R. Nuclear Med. Biol.
1992, 19, 275.
12. Wester, H. J.; Krommeich, C. H.; Fixmann, A.; Former,
W.; Muller-Gartner, H. W.; Stocklin, D. Lab. Comp. Radio-
pharm. 1995. 37, 513.
13. Barlos, K.; Gatos, D.; Koutsogianni, S. J. of Pept. Res.
1998, 51, 194.
14. Fields, C. G.; Fields, G. B. Tetrahedron Lett. 1993, 34,
6661.
15. Carpino, L. A.; Shro€, H.; Triolo, S. A.; Mansour, E. M.
E.; Wenschuh, H.; Albericio, F. Tetrahedron Lett. 1993, 34,
7829.
16. Aletras, A.; Barlos, K.; Gatos, D.; Koutsogianni, S.;
Mamos, P. Int. J. of Pept. Prot. Res. 1995, 45, 488.
17. Han, Y.; Bontems, S. L.; Hegyes, P.; Munson, M. C.;
Minor, C. A.; Kates, S. A.; Albericio, F.; Barany, G. J. Org.
Chem. 1996, 61, 6326.
18. Knorr, R.; Trzeciak, A.; Bannwarth, W.; Gillessen, D.
Tetrahedron Lett. 1989, 30, 48.
19. Pearson, D. A.; Blanchette, M.; Baker, M. L.; Guindon, C.
A. Tetrahedron Lett. 1989, 30, 2739.
20. Carpino, L. A.; Elfaham, A.; Minor, C. A.; Albericio, F. J.
Chem. Soc. Ð Chem. Commun. 1994, 201.
21. Carpino, L. A. J. Am. Chem. Soc. 1993, 115, 4397.
22. Bo®ll, J. M.; Albericio, F. Tetrahedron Lett. 1999, 40,
2641.
23. Sieber, P.; Riniker, B. Tetrahedron Lett. 1991, 32, 739.
24. Gausepohl, H.; Kraft, M.; Frank, R. W. Int. J. Pept. Pro-
tein Res. 1989, 34, 287.
In conclusion, the current study has demonstrated that
the solid phase strategy presented has a number of
advantages in the labelling of peptides with [18F]. It is
fast, ecient, clean, selective, highly versatile and may
be readily automated. The [18F]labelled peptides were
synthesised within 20 min with a high speci®c activity.
25. Haka, M. S.; Kilbourn, M. R.; Watkins, G. L.; Toor-
angian, S. A. J. Lab. Comp. Radiopharm. 1989, 27.