Synthesis of N-(3-[18F]fluoropropyl)-2β-carbomethoxy- 3β-(4-iodophenyl)nortropane ([18F]FP-β-CIT)

Article abstract of DOI:10.1002/jlcr.1043

N-(3-[¹⁸F]fluoropropyl)-2β-carbomethoxy-3β-(4- iodophenyl)nortropane ([¹⁸F]FP-β-CIT) was synthesized in a two-step reaction sequence. In the first reaction, 1-bromo-3-(nitrobenzene-4- sulfonyloxy)-propane was fluorinated with no-carrier-added fluorine-18. The resulting product, 1-bromo-3-[¹⁸F]-fluoropropane, was distilled into a cooled reaction vessel containing 2β-carbomethoxy-3β-(4-iodophenyl)- nortropane, diisopropylethylamine and potassium iodide. After 30 min, the reaction mixture was subjected to a preparative HPLC purification. The product, [¹⁸F]FP-β-CIT, was isolated from the HPLC eluent with solid-phase extraction and formulated to yield an isotonic, pyrogen-free and sterile solution of [¹⁸F]FP-β-CIT. The overall decay-corrected radiochemical yield was 25 ± 5%. Radiochemical purity was >98% and the specific activity was 94 ± 50 GBq/μmol at the end of synthesis. Copyright

Full text of DOI:10.1002/jlcr.1043

JOURNAL OF LABELLED COMPOUNDS AND RADIOPHARMACEUTICALS
J Label Compd Radiopharm 2006; 49: 77–89.
Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/jlcr.1043
Research Article
Synthesis of N-(3-[18F]Fluoropropyl)-
2b-carbomethoxy-3b-(4-iodophenyl)nortropane
([18F]FP-b-CIT)
R. P. Klok*, P. J. Klein, J. D. M. Herscheid and A. D. Windhorst
Nuclear Medicine and PET Research, VU University Medical Centre, De Boelelaan
1085c, 1081 HV Amsterdam, The Netherlands
Summary
N-(3-[18F]fluoropropyl)-2b-carbomethoxy-3b-(4-iodophenyl)nortropane ([18F]FP-b-CIT)
was synthesized in
a two-step reaction sequence. In the first reaction,
1-bromo-3-(nitrobenzene-4-sulfonyloxy)-propane was fluorinated with no-carrier-
added fluorine-18. The resulting product, 1-bromo-3-[18F]-fluoropropane, was
distilled into a cooled reaction vessel containing 2b-carbomethoxy-3b-(4-iodophe-
nyl)-nortropane, diisopropylethylamine and potassium iodide. After 30 min, the
reaction mixture was subjected to a preparative HPLC purification. The product,
[18F]FP-b-CIT, was isolated from the HPLC eluent with solid-phase extraction and
formulated to yield an isotonic, pyrogen-free and sterile solution of [18F]FP-b-CIT.
The overall decay-corrected radiochemical yield was 25 ꢀ 5%. Radiochemical
purity was >98% and the specific activity was 94 ꢀ 50 GBq/mmol at the end of
synthesis. Copyright # 2006 John Wiley & Sons, Ltd.
Key Words: F-18; [18F]FP-b-CIT; dopamine transporter; positron emission
tomography
Introduction
The in vivo visualization and quantification of the dopamine transporter by
positron emission tomography (PET) can be achieved by the utilization of
radiolabelled cocaine derivatives. Cocaine analogues where the benzoic ester
moiety has been replaced by a 4-halogenophenyl group (CFT,1–3 CIT,4–6
CNT,7,8 CBT9) or a 4-methylphenyl group (CMT10) and the N-methyl is
substituted by a N-fluoropropyl or N-fluoroethyl moiety (FP,1,3–6,8–10 FE2,7)
are often used and give satisfactory results.
*Correspondence to: R. P. Klok, Nuclear Medicine and PET Research, VU University Medical Centre,
De Boelelaan 1085c, 1081 HV Amsterdam, The Netherlands. E-mail: rklok@rnc.vu.nl
Copyright # 2006 John Wiley & Sons, Ltd.
Received 4 August 2005
Revised 7 December 2005
Accepted 10 December 2005

78
R. P. KLOK ET AL.
18F
O
O
N
I
Figure 1. N-(3-[18F]fluoropropyl)-2b-carbomethoxy-3b-(4-iodophenyl)nortro-
pane ([18F]FP-b-CIT)
Table 1. Yields and specific activity of literature methods
Entry
Reference
Radiochemical
yield (%)
Specific activity
(EOS, GBq/mmol)
Remarks
1
2
3
Kazumata6
Lundkvist4
Chaly5
NR
2-4
1-2
37
55
NR
90% chemically pure
NR, not reported.
We focussed our efforts on the synthesis of N-(3-[18F]fluoropropyl)-2b-
carbomethoxy-3b-(4-iodophenyl)nortropane ([18F]FP-b-CIT, Figure 1). The
I-123-labelled analogue (DATScan) is a well known and investigated SPECT
ligand. We chose to label this compound with 18F for use in PET. Methods
for the synthesis of [18F]FP-b-CIT described in literature thus far showed
either low yields or moderate specific activities, or a combination of the both
(see Table 1).
We assume that the moderate specific activity is caused by side-products,
since the employed no-carrier-added methods cannot cause that much
[18F]FP-b-CIT carrier. One can only speculate on the nature of such side-
products since the authors have not explored these possibilities. However, it
must be assumed that these are caused by alkylation of nor-b-CIT by
unreacted starting material of the fluorination reaction. It is beyond the scope
of this paper to elaborate on such speculations, but the mediocre results urged
us to explore a different synthesis route with a higher yield, reproducibility and
specific activity.
It was our hypothesis that splitting the synthesis of [18F]FP-b-CIT in two
separate reactions steps, with intermediate purification of o-[18F]fluorohalo-
genopropane by distillation and subsequent alkylation of 2b-carbomethoxy-
3b-(4-iodophenyl)-nortropane (nor-b-CIT), would be an improvement on the
described synthesis procedures in literature.
Results and discussion
Synthesis of o-[18F]fluorohalogenopropane
An effective synthesis of a o-[18F]fluorohalogenopropane as a radiofluorinated
intermediate in the synthesis of [18F]FP-b-CIT would be beneficial for two
Copyright # 2006 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 77–89

SYNTHESIS OF N-(3-[18F]FLUOROPROPYL)
79
reasons. Firstly, the boiling point of such intermediates is in such a range that
distillation is feasible and secondly the starting material will be retained in the
reaction vessel, due to their very high boiling points, thus eliminating the risk
of formation of pseudo-carrier. Moreover, the fluorination chemicals will not
be present during the alkylation reaction, in contrast to the one-pot synthesis
procedure described in literature,6 resulting in a less tedious purification
procedure as well as a possibly higher alkylation yield.
Fluorination was performed as depicted in Scheme 1, starting from
compounds 1a–h, which were synthesized according to literature procedures.
Fluorination yielded a mixture of the products 2 and 3 of which the
individual yields were dependent on the characteristics of the leaving groups.
The tosylates and nosylates were preferably substituted by 18F, while the
bromide was substituted in favor of the mesylate as the leaving group. In this
way the yield of 2 could be controlled in such a way that the yield increased up
to 75% (decay corrected) for 1g Results are shown in Table 2.
Reaction time and temperature were also investigated. The fluorination is a
fast reaction, already after 5 min yields are at their maximum. Upon
elongation of the reaction time, yields decreased significantly, probably caused
by decomposition of the product. The temperature was varied between 50 and
1258C. Best results were obtained between 100 and 1108C, at higher
temperatures the volatile product can leak away from the reaction vessel,
due to the elevated pressure inside.
18F/K222/K2CO3
R
18F
18F
R'
+
R
'
R
CH3CN
Precursor (1)
Main product (2)
Side product (3)
Compound
R
R’
Compound
R
R’
1a
1b
1c
1d
Br
Cl
Br
Cl
Mesylate
Triflate
Triflate
Tosylate
1e
1f
1g
Br
I
Cl
Br
Tosylate
Tosylate
Nosylate
Nosylate
1h
Scheme 1. Fluorination reaction
Table 2. Results of the fluorination of 1a–h
Precursor
R
R0
% (2)
% (3)
% Unreacted 18F
% Undefined
1a
1b
1c
1d
1e
1f
Br
Cl
Br
Cl
Br
I
Mesylate
Triflate
Triflate
Tosylate
Tosylate
Tosylate
Nosylate
Nosylate
25
16
7
58
41
31
75
57
46
0
5
1
5
0
1
1
21
75
83
19
37
41
19
24
8
9
5
21
17
28
5
1g
1h
Cl
Br
18
Copyright # 2006 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 77–89

80
R. P. KLOK ET AL.
The precursors 1a–h in acetonitrile were added to the dried 18F/Kryptofix
K222/K2CO3 solution (see Experimental section). This mixture was heated for
5 min at 1008C, after which the products were distilled, together with the
solvent acetonitrile, into a cold vessel. The distillate was subjected to HPLC to
determine the yield of product 2. The amount of unreacted fluoride and all
other non-volatile products 3 (identified with cold reference compounds) was
determined by adding acetonitrile to the residue in the reaction vessel for
subsequent HPLC analysis.
Best results were obtained with the nosylate precursors. Giving the fact of a
decreasing yield of 1d–1f and a decreasing yield of 1g–1h, the iodopropylno-
sylate precursor was not investigated, since it was anticipated that this
compound would result in lower yields. Furthermore, the iodopropylfluoride
has a higher boiling point than the chloro- and bromopropylfluoride, making
the distillation of this intermediate less feasible.
Synthesis of N-(3-[18F]fluoropropyl)-2b-carbomethoxy-3b-(4-iodophenyl)
nortropane (5a)
Initially, compound 1g was selected to be used for the alkylation of the nor-
b-CIT 4, Scheme 2. However, alkylation yields were low, despite increasing the
reaction temperature (80–1608C). Also the addition of potassium iodide as a
catalyst did not improve the alkylation yield. However, 1-bromo-3-[18F]fluor-
opropane gave better results, caused by the better leaving group characteristics
of the bromine atom, which also has been shown by Bauman et al.11
Compound 1h was therefore selected for optimization of the alkylation
reaction (Table 3).
Best results for the alkylation were achieved in 1 ml of acetonitrile with 5 ml
diisopropylethylamine as a base at 1508C. Also DMSO at 1308C gave good
results, but we did not investigate higher temperatures with DMSO because at
1308C already no intact fluorobromopropane was left and polar side-products
are formed. We believe that at higher temperatures the reaction yield will not
increase, but decrease and more polar side-products will be formed.
In contrast to all literature methods1,2,4,6–10 where DMF is used as solvent
for the alkylation reaction, the alkylation in acetonitrile shows higher yields
18F
O
O
O
O
HN
N
18F
Br
+
i
I
I
4
5a
2h
Scheme 2. Alkylation of nor-b-CIT (4). Reaction conditions: (i) potassium
iodide, diisopropylethylamine, acetonitrile, 1508C, 30 min
Copyright # 2006 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 77–89

SYNTHESIS OF N-(3-[18F]FLUOROPROPYL)
81
Table 3. Optimization of the alkylation reaction conditionsa
Solvent
Temperature
(8C)
Yield 5a
(%)
Yield undefined
product (%)
DMF
DMF
DMF
THF
DMSO
CH3CN
CH3CN
100
130
150
130
130
130
150
23
30
29
39
49
41
52
35
44
71
10
46
27
22
a Three milligrams of 4, 30 min reaction time; yields relative to 1-bromo-3-[18F]fluoropropane.
with less impurities. Diisopropylethylamine is a well known, sterically
hindered base, which itself is non-nucleophilic, and enhances the alkylation
reaction also in this case, improving the yield by 10%. The addition of
potassium iodide proved to be essential for the alkylation. Without the in situ
formation of iodo 2h via exchange of Br for I, alkylation occurred only in very
low yields of 5–10%. The iodo 2h could clearly be recognized on HPLC. The
formation of the iodo intermediate is probably also the reason why 2g shows
much less alkylation, since in that case no exchange of the Cl by the I occurs
and therefore a low alkylation yield was observed.
The alkylation reaction mixture was subjected to semi-preparative HPLC
for the purification of 5a, on a Kromasil 250 ꢁ 21 mm column with methanol/
water/diisopropylamine 90/10/0.2 as the eluent. At a flow rate of 10 ml/min the
retention time of [18F]FP-b-CIT was 15 min. However, from the chromato-
gram (Figure 2), the presence of a contamination in the UV channel which
co-eluted with [18F]FP-b-CIT was observed.
One likely possibility for this contamination could be N-(1-propene-3-yl)-
2b-carbomethoxy-3b-(4-iodophenyl)nortropane 5b. This product could result
from the alkylation of 4 by 3-bromo-1-propene 6 and might even be a potent
dopamine transporter ligand, and can as such act as a pseudo-carrier. The very
reactive alkylating agent 6 could be formed under the alkaline reaction
conditions of the fluorination by elimination of the nitrophenylsulphonyl
group (Scheme 3).
In order to verify this hypothesis, commercially available 3-bromo-
1-propene, was reacted with (4), and a single reaction product was formed
within 5 min at room temperature. The obtained compound from this
synthesis co-eluted on HPLC analysis with the observed contamination in
the synthesis of [18F]FP-b-CIT proving that this compound was indeed the
contamination. The chemical structure of the reaction product was determined
with NMR and proved to be compound 5b. The product 5b is also formed
when other sulfonate esters 1a–1g, were used.
Copyright # 2006 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 77–89

82
R. P. KLOK ET AL.
Figure 2. Semi-prep HPLC of [18F]FP-b-CIT, using a Kromasil 250 ꢁ 21 mm
column
Nos
Br
Elimination
O
O
O
O
HN
N
i
+
Br
I
I
6
4
5b
Scheme 3. Alkylation of the nor-b-CIT precursor yielding the pseudo-carrier 5b.
Reaction conditions: (i) potassium iodide, diisopropylethylamine, acetonitrile,
1508C, 30 min
Semi-preparative HPLC purification
Compound 5b shows a remarkable resemblance with [18F]FP-b-CIT on
various semi-preparative HPLC systems. Results are depicted in Table 4, these
were the maximum achieved resolution for these columns, despite the
variation of the eluent. Only by using a Phenomenex Luna C18(2) 5 mm,
250 ꢁ 10 mm and acetonitrile/water/diisopropylamine 57.5/42.5/0.2 (v/v/v) as
eluent, sufficient separation could be achieved between FP-b-CIT (Rt 52 min)
and the pseudo-carrier (Rt 48 min) (Figure 3).
Copyright # 2006 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 77–89

SYNTHESIS OF N-(3-[18F]FLUOROPROPYL)
83
Table 4. Selection of the investigated HPLC purification systems
Column
Eluent
Rt
FP-b-
CIT
Rt
propene- (Rs)
Resolution
CIT
between
investigated
peaksa
Kromasil
MeOH/H2O/DiPA 90/10/0.2
MeCN/NH4Ac 75/25
15
17
16
16
19
20
21
29
48
0.64
1.00
0.93
0.28
0.88
0.29
1.82
250 ꢁ 21 mm
Spherisorb
250 ꢁ 8 mm
Sperimage
5 mM NH4H2PO4 pH 7.4/MeCN 15/85 17
250 ꢁ 16 ODS2
Xterra
MeOH/H2O/DiPA 75/25/0.2
0.01 M H3PO4/MeCN 75/25
MeCN/H2O/DiPA 65/35/0.2
MeCN/H2O/DiPA 57.5/42.5/0.2
19
27
30
52
150 ꢁ 3.9 mmb
Bondapak
300 ꢁ 7.8 mm
Luna C18(2)
250 ꢁ 10 mm
Luna C18(2)
250 ꢁ 10 mm
a Rs51.25 no baseline separation; Rs>1.5 baseline separation between UV propene-CIT and radioactive
FP-b-CIT.
b Analytical amount.
Figure 3. Representative semi-preparative HPLC purification of [18F]FP-b-CIT,
using a Phenomenex Luna C18(2) 250 ꢁ 10 mm column
Copyright # 2006 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 77–89

84
R. P. KLOK ET AL.
Stability of [18F]FP-b-CIT
QC HPLC analysis of the formulated [18F]FP-b-CIT showed that radio-
chemical purity of [18F]FP-b-CIT was irreproducible. Sometimes the
radiochemical purity was even less than 98%. QC showed a side-product
(SP1) which was formed during the formulation of [18F]FP-b-CIT (Figure
4(A)) and did not increase in time. We hypothesized that the impurity might be
the product of the hydrolysis of the ester function of [18F]FP-b-CIT.
However, upon addition of 1 M NaOH to the formulated solution of
[18F]FP-b-CIT and heating at 508C for 10 min, a different side-product was
formed instead of SP1, since the retention time of the alkaline hydrolysis
product was clearly different from the retention time of SP1 (Figure 4(B)). One
must conclude from this experiment that SP1 is not the product of the
hydrolysis of the ester moiety of [18F]FP-b-CIT, but another, yet unidentified
impurity. When the HPLC fraction itself was analysed, it was shown that
[18F]FP-b-CIT remained 100% pure in time (Figure 4(C)). One can only
speculate on the nature of this impurity, but Chaly et al.5 reported a
decomposition or epimerization into the 2-a isomer on a silica cartridge, also
Xing et al.12 reported an epimerization in the [18F]FP-b-CIT synthesis. We
assume that SP1 is also this epimerization product, since in the formulation of
2
CPS *1000
CPS *1000
2
1
1
(A)
0
8
min
(B)
0
2
6
8
min
8
2
4
6
4
4
CPS *1000
CPS *1000
2
4
(C)
(D)
0
min
min
2
6
8
2
6
8
4
0
4
Figure 4. QC radiochromatograms of formation of SP1. (A) Formulated
product, with side-product. (B) Formulated product partially hydrolysed. (C)
Collected HPLC fraction. (D) Ethanol fraction of solid-phase extraction
Copyright # 2006 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 77–89

SYNTHESIS OF N-(3-[18F]FLUOROPROPYL)
85
Table 5. Purity of [18F]FP-b-CIT during formulation
Step in SPE
[18F]FP-b-CIT (%)
SP1 (%)
HPLC fraction
HPLC fraction, after 1 h
Washing
Ethanol fraction
Formulated product
100
100
ND
95
0
0
ND
5
5
95
ND, not determined; products present on SPE cartridge; HPLC analysis not possible.
[18F]FP-b-CIT, also a solid-phase extraction (SPE) is employed. Table 5 shows
the results of the analysis of the various intermediate steps in the SPE
procedure.
From these results it was concluded that SP1 originates after the washing of
the SPE cartridge and before the elution with ethanol (Figure 4(D)). In
between these two successive steps, the SPE cartridge runs dry and during this
period SP1 originates. Indeed it was shown that upon elongation of the period
the SPE cartridge runs dry, the amount of SP1 impurity also increases.
Preventing the SPE cartridge to run dry during the formulation resulted in a
reproducible purity of [18F]FP-b-CIT which exceeded 98%.
Experimental
Material and methods
Acetonitrile, Kryptofix 222, potassium carbonate and potassium iodide were
obtained from Merck, other chemicals were obtained from Aldrich. All
chemicals were used as received.
HPLC for semi-preparative separation was performed with a Jasco PU1587
pump, an in-line Jasco UV1575 UV detector (wavelength 235 nm), a flow-
through radioactivity detector with GM tube (home-made) and Jasco-Borwin
software for data acquisition and processing. HPLC Column Phenomenex
Luna C18(2) 5 mm, 250 ꢁ 10 mm (Bester). Seppak C18 plus (Waters). Radio-
activity was quantified with a VDC-405 dose calibrator (Veenstra instru-
ments). All reactions were carried out utilizing a home-made, remotely
controlled apparatus.13 Analytical HPLC was performed with a Jasco PU1580
pump, an in-line Jasco UV2075 UV detector (wavelength 235 nm), a flow-
through NaI (Tl) crystal scintillation detection system and Raytest Gina-NT
for data acquisition.
Synthesis of sulphonic esters
All esters, except for the 3-chloropropyl-trifluoromethanesulfonate, were
synthesized according to literature.14 This was successful for all the esters
Copyright # 2006 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 77–89

86
R. P. KLOK ET AL.
except for 3-chloropropyl-trifluoromethanesulfonate, this compound was
prepared according to Chi et al.15
NMR data
1
3-Bromoprophyl-methanesulfonate. H NMR (200 MHz, CDCl3) d 4.36 (t, 2H,
J ¼ 5:8 Hz, H-1), 3.50 (t, 2H, J ¼ 6:2 Hz, H-3), 3.02 (s, 3H, methyl), 2.25
(m, 2H, H-2).
1
3-Chloropropyl-trifluoromethanesulfonate. H NMR (200 MHz, CDCl3) d 4.69
(t, 2H, J ¼ 5:9 Hz, H-1), 3.65 (t, 2H, J ¼ 6:0 Hz, H-3), 2.25 (m, 2H, H-2).
1
3-Bromopropyl-trifluoromethanesulfonate. H NMR (200 MHz, CDCl3) d 4.63
(t, 2H, J ¼ 5:8 Hz, H-1), 3.44 (t, 2H, J ¼ 6:2 Hz, H-3), 2.28 (m, 2H, H-2).
1
1-Chloro-3-(toluene-4-sulfonyloxy)-propane. H NMR (200 MHz, CDCl3)
d 7.76 (d, 2H, J ¼ 8:3 Hz, aryl), 7.32 (d, 2H, J ¼ 8:3 Hz, aryl), 4.15 (t, 2H,
J ¼ 5:9 Hz, H-3), 3.53 (t, 2H, J ¼ 6:2 Hz, H-1), 2.42 (s, 3H, methyl), 2.07
(m, 2H, H-2).
1
1-Bromo-3-(toluene-4-sulfonyloxy)-propane. H NMR (200 MHz, CDCl3)
d 7.75 (d, 2H, J ¼ 8:0 Hz, aryl), 7.30 (d, 2H, J ¼ 8:0 Hz, aryl), 4.12 (t, 2H,
J ¼ 6:0 Hz, H-3), 3.35 (t, 2H, J ¼ 6:0 Hz, H-1), 2.40 (s, 3H, methyl), 2.12
(m, 2H, H-2).
1
1-Iodo-3-(toluene-4-sulfonyloxy)-propane. H NMR (200 MHz, CDCl3) d 7.70
(d, 2H, J ¼ 8:0 Hz, aryl), 7.32 (d, 2H, J ¼ 8:0 Hz, aryl), 4.08 (t, 2H,
J ¼ 6:0 Hz, H-3), 3.14 (t, 2H, J ¼ 6:0 Hz, H-1), 2.45 (s, 3H, methyl), 2.10
(m, 2H, H-2).
1
1-Chloro-3-(nitrobenzene-4-sulfonyloxy)-propane. H
NMR
(200 MHz,
CDCl3) d 8.41 (d, 2H, J ¼ 8:8 Hz, aryl), 8.12 (d, 2H, J ¼ 8:8 Hz, aryl), 4.29
(t, 2H, J ¼ 5:8 Hz, H-3), 3.57 (t, 2H, J ¼ 6:1 Hz, H-1), 2.13 (m, 2H, H-2).
1
1-Bromo-3-(nitrobenzene-4-sulfonyloxy)-propane. H
NMR
(200 MHz,
CDCl3) d 8.40 (d, 2H, J ¼ 8:9 Hz, aryl), 8.11 (d, 2H, J ¼ 8:8 Hz, aryl), 4.28
(t, 2H, J ¼ 5:8 Hz, H-3), 3.41 (t, 2H, J ¼ 6:1 Hz, H-1), 2.21 (m, 2H, H-2).
Synthesis of N-(1-propene-3-yl)-2b-carbomethoxy-3b-(4-iodophenyl)nortropane
(5b)
3-Bromo-1-propene (6) (2 ml, 23 mmol), was reacted with 5 mg (13 mmol)
nortropane (4) in acetonitrile, for 30 min at 508C. Compound 5b was isolated
with the same semi-prep and SPE procedure as for [18F]FP-b-CIT.
1H NMR (200 MHz, CDCl3) d 7.55 (d, 2H, J ¼ 8:4 Hz, aryl), 7.00 (d, 2H,
J ¼ 8:3 Hz, aryl), 5.75 (m, 1 H, ¼¼ CH), 5.10 (t, 2H, CH2 ¼¼), 3.67 (s, 1 H,
Copyright # 2006 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 77–89

SYNTHESIS OF N-(3-[18F]FLUOROPROPYL)
87
H-1), 3.48 (s, 3H, OCH3), 3.41 (s, 1 H, H-5), 3.0 (m, 2H, H-3 and H-2), 2.55
(m, 1 H, H-4a), 2.05 (m, 2H, H-6a and H-7a), 1.65 (m, 5 H, CH2, H-6b, H-7b
and H-4b).
ꢂ
Production of [18F]F
[18F]Fꢂ was produced by the 18O(p,n)18F nuclear reaction with an IBA 18/9
cyclotron. An irradiation of 40 min with 40 mA of 18 MeV protons yielded
55–65 GBq of [18F]Fꢂ at EOB. After irradiation the 18F was trapped on a
BioRad AG-1-X8 column (carbonate form), while the enriched water was
collected for re-use.
Production of 1-bromo-3-[18F]-fluoropropane
The 18F was eluted from the anion exchange column by using 1 ml acetonitrile/
water (9/1 v/v) containing 13 mg (34 mmol) of Kryptofix 222 and 2 mg
(14 mmol) of potassium carbonate. The solution was dried under a Helium flow
and reduced pressure at 908C. To remove residual water, 0.5 ml of acetonitrile
was added, and the solution was dried again. At room temperature the
precursor 1-bromo-3-(nitrobenzene-4-sulfonyloxy)-propane in 0.5 ml acetoni-
trile was added. Fluorination reaction was carried out for 5 min at 1108C. At
the end of the reaction, the formed 1-bromo-3-[18F]fluoropropane was
distilled, together with the solvent, acetonitrile, into a cooled (ꢂ308C) second
reaction vessel, which contained 5.5 mg (15 mmol) of 2b-carbomethoxy-
3b-(4-iodophenyl)-nortropane (4), and 4 mg (24 mmol) of potassium iodide in
0.5 ml acetonitrile and 5 ml (29 mmol) diisopropylethylamine.
Production of [18F]-N-(3-fluoropropyl)-2b-carbomethoxy-3b-(4-iodophenyl)-
nortropane
After the distillation step, the reaction vessel was closed and heated at 1508C
for 30 min. The reaction mixture was then cooled to room temperature and
quenched with 0.5 ml of water and subsequently transferred to the HPLC unit
using a remote-control-operated HPLC injection system and subjected to a
semi-preparative HPLC purification using a Phenomenex Luna C18(2) 5 mm,
250 ꢁ 10 mm column. Eluent was acetonitrile:water:diisopropylamine
57.5:42.5:0.2 (% v/v/v) at a flow of 5 ml/min.
[18F]-N-(3-fluoropropyl)-2b-carbomethoxy-3b-(4-iodo-phenyl)nortropane
had a retention time of 52 min, as detected by UV (235 nm) and radioactivity
flow-through GM detector. Specific activity was determined by calculation of
the peak area of radioactive product combined with the peak area of the UV
peak, quantified with a calibration curve of the non-radioactive FP-b-CIT.
The collected fraction was diluted with 60 ml of sterile water in the
formulation unit.13 The diluted product was passed over a Sep-Pak plus C18
Copyright # 2006 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 77–89

88
R. P. KLOK ET AL.
cartridge using helium overpressure, followed by a washing with 60 ml sterile
water. The desired product was eluted with 2 ml of sterile ethanol, followed by
17 ml of sterile citrate/acetate buffer (0.038 M citric acid, 0.054 M sodiumci-
trate, 0.051 M sodiumacetate, pH 5) and filtered through a sterile 0.22 mm
Millipore GV filter using helium overpressure, yielding a sterile and pyrogen-
free solution, ready for injection. Total synthesis time was 125 min including
the purification and the formulation. The product was obtained in 25 ꢀ 5%
(n ¼ 34) decay-corrected isolated radiochemical yield with a specific activity at
the end of synthesis of 94 ꢀ 50 GBq/mmol (n ¼ 34).
An aliquot was analyzed by HPLC using a Kromasil C18, 10 mm,
250 ꢁ 4.6 mm column, with acetonitrile:water:diisopropylamine 90:10:0.2
(% v/v/v) as mobile phase, at a flow rate of 1 ml/min. Radiochemical purity
was >98%.
Residual solvent analysis on GC showed less than 18, 18, 26 and 1 ppm of
acetone, acetonitrile, diisopropylamine and diisopropylethylamine, respec-
tively. The ethanol concentration was found to be 10%.16
Conclusion
A fully automated, GMP compliant, reproducible procedure for the synthesis
of [18F]-FP-b-CIT has been developed. The product was obtained in 25 ꢀ 5%
(n ¼ 34) decay-corrected isolated radiochemical yield in 125 min after end of
bombardment. The specific activity at the end of synthesis was 94 ꢀ 50 GBq/
mmol (n ¼ 34). Radiochemical purity was >98%, and no other mass peaks
other than the carrier product were detected. Although the synthesis time is
longer than the previously described synthesis routes, the yield and specific
activity are significantly improved.
Acknowledgements
The authors wish to thank the Janssen Research Foundation for kindly
donating 2b-carbomethoxy-3b-(4-iodophenyl)nortropane. P.J. van Leuffen
and the personnel of the BV Cyclotron VU are greatly acknowledged for
providing the 18F.
References
1. Kamarainen EL, Kyllonen T, Airaksinen A, Lundkvist C, Yu MX, Nagren K,
Sandell J, Langer O, Vepsalainen J, Hiltunen J, Bergstrom K, Lotjonen S,
Jaakkola T, Halldin C. J Label Compd Radiopharm 2000; 43: 1235–1244.
2. Harada N, Ohba H, Fukumoto D, Kakiuchi T, Tsukada H. Synapse 2004; 54:
37–45.
3. Koivula T, Perhola O, Kamarainen EL, Lipponen T, Vepsalainen J, Solin O.
J Label Compd Radiopharm 2005; 48: 463–471.
Copyright # 2006 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 77–89

SYNTHESIS OF N-(3-[18F]FLUOROPROPYL)
89
4. Lundkvist C, Halldin C, Ginovart N, Swahn CG, Farde L. Nucl Med Biol 1997;
24: 621–627.
5. Chaly T, Dhawan V, Kazumata K, Antonini A, Margouleff C, Dahl JR,
Belakhlef A, Margouleff D, Yee A, Wang SY, Tamagnan G, Neumeyer JL,
Eidelberg D. Nucl Med Biol 1996; 23: 999–1004.
6. Kazumata K, Dhawan V, Chaly T, Antonini A, Margouleff C, Belakhlef A,
Neumeyer J, Eidelberg D. J Nucl Med 1998; 39: 1521–1530.
7. Goodman MM, Kilts CD, Keil R, Shi B, Martarello L, Xing DX, Votaw J,
Ely TD, Lambert P, Owens MJ, Camp VM, Malveaux E, Hoffman JM. Nucl Med
Biol 2000; 27: 1–12.
8. Goodman MM, Keil R, Shoup TM, Eshima D, Eshima L, Kilts C, Votaw J,
Camp VM, Votaw D, Smith E, Kung MP, Malveaux E, Watts R, Huerkamp M,
Wu D, Garcia E, Hoffman JM. J Nucl Med 1997; 38: 119–126.
9. Chaly T, Baldwin RM, Neumeyer JL, Hellman MJ, Dhawan V, Garg PK,
Tamagnan G, Staley JK, Al Tikriti MS, Hou YK, Zoghbi SS, Gu XH, Zong R,
Eidelberg D. Nucl Med Biol 2004; 31: 125–131.
10. Chaly T, Matacchieri R, Dahl R, Dhawan V, Eidelberg D. Appl Radiat Isot 1999;
51: 299–305.
11. Bauman A, Piel M, Schirrmacher R, Rosch F. Tetrahedron Lett 2003; 44:
9165–9167.
12. Xing DX, Chen P, Keil R, Kilts CD, Shi B, Camp VM, Malveaux G, Ely T,
Owens MJ, Votaw J, Davis M, Hoffman JM, BaKay RAE, Subramanian T,
Watts RL, Goodman MM. J Med Chem 2000; 43: 639–648.
13. Windhorst AD, Ter Linden T, De Nooij A, Keus JF, Buijs FL, Schollema PE,
Van Rooij LF, Herscheid JDM. J Label Compd Radiopharm 2001; 44:
S1052–S1054.
14. Jarosinski MA, Reddy PS, Anderson WK. J Med Chem 1993; 36: 3618–3627.
15. Chi DY, Kilbourn MR, Katzenellenbogen JA, Welch MJ. J Org Chem 1987;
52: 658–664.
16. Klok RP, Windhorst AD. Nucl Med Biol, submitted for publication.
Copyright # 2006 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 77–89