Trifluoromethanesulfonic acid, an alternative solvent medium for the direct electrophilic fluorination of DOPA: New syntheses of 6-[18F]fluoro- L-DOPA and 6-[18F]fluoro-D-DOPA

Article abstract of DOI:10.1002/jlcr.1454

Previous work from this laboratory has shown that the direct fluorination of 3, 4-dihydroxy-phenyl-L-alanine (L-DOPA) in anhydrous HF (aHF) or BF ₃/HF with F₂ is an efficient method for the synthesis of 6-fluoro-L-DOPA. Since then, ¹⁸F-labeled 6-fluoro-L-DOPA ([ ¹⁸F]6-fluoro-L-DOPA) has been used to study presynaptic dopaminergic function in the human brain and to monitor gastrointestinal carcinoid tumors. This work demonstrates that the reactivity and selectivity of F₂ toward L-DOPA in CF₃SO₃H is comparable with that in aHF. This new synthetic procedure has led to the production of [¹⁸F] fluoro-L-DOPA and [¹⁸F]fluoro-L-DOPA isomers in 17 ± 2% radiochemical yields (decay corrected with respect to [¹⁸F]F ₂). The 2- and 6-FDOPA isomers were separated by HPLC and subsequently characterized by ¹⁹F NMR spectroscopy. The corresponding [¹⁸F]-FDOPA enantiomers have been obtained in clinically useful quantities by a synthetic approach that avoids the use of aHF. Copyright

Full text of DOI:10.1002/jlcr.1454

Journal of Labelled Compounds and Radiopharmaceuticals
J Label Compd Radiopharm 2007; 50: 1236–1242.
Published online 1 October 2007 in Wiley InterScience
(www.interscience.wiley.com). DOI: 10.1002/jlcr.1454
JLCR
Research Article
Trifluoromethanesulfonic acid, an alternative solvent medium
for the direct electrophilic fluorination of DOPA: new syntheses
of 6-[18F]fluoro-l-DOPA and 6-[18F]fluoro-d-DOPA
BABAK BEHNAM AZAD1,2, RAMAN CHIRAKAL1,2,* and GARY J. SCHROBILGEN2,*
1 Nuclear Medicine, McMaster University Medical Centre, Hamilton Health Sciences, Hamilton, ON L8N 3Z5, Canada
2 Department of Chemistry, McMaster University, Hamilton, ON L8S 4M1, Canada
Received 13 June 2007; Revised 6 July 2007; Accepted 7 August 2007
Abstract: Previous work from this laboratory has shown that the direct fluorination of 3, 4-dihydroxy-phenyl-l-
alanine (l-DOPA) in anhydrous HF (aHF) or BF3/HF with F2 is an efficient method for the synthesis of 6-fluoro-l-
DOPA. Since then, 18F-labeled 6-fluoro-l-DOPA ([18F]6-fluoro-l-DOPA) has been used to study presynaptic
dopaminergic function in the human brain and to monitor gastrointestinal carcinoid tumors.
This work demonstrates that the reactivity and selectivity of F2 toward l-DOPA in CF3SO3H is comparable with
that in aHF. This new synthetic procedure has led to the production of [18F]fluoro-l-DOPA and [18F]fluoro-D-DOPA
isomers in 17 ꢀ 2% radiochemical yields (decay corrected with respect to [18F]F2). The 2- and 6-FDOPA isomers were
separated by HPLC and subsequently characterized by 19F NMR spectroscopy. The corresponding [18F]-FDOPA
enantiomers have been obtained in clinically useful quantities by a synthetic approach that avoids the use of
aHF. Copyright # 2007 John Wiley & Sons, Ltd.
Keywords: electrophilic fluorination; fluorine-18; PET; radiolabeling; DOPA; cyclotron
Introduction
unable to reach the dopaminergic cells of the brain. The
immediate dopamine precursor, l-DOPA, can, however,
cross the blood–brain barrier and can therefore be
employed to monitor in vivo intracerebral dopamine
metabolism when labeled with 18F.3 Garnett et al.
pioneered the use of 6-fluoro-l-DOPA, in conjunction
with PET, to visualize regional distributions of intra-
cerebral dopamine in the human brain.2 Fluorine-18-
labeled 6-fluoro-l-DOPA can be produced by the direct
electrophilic fluorination of l-DOPA in anhydrous HF
(aHF), but is formed in admixture with the 2- and 5-
fluoro-l-DOPA isomers, which cannot be utilized to
study the dopaminergic pathways of the brain.4
Although 18F-FDG has been the benchmark for brain
tumor detection, recent studies have shown its use for
the detection of low-grade tumors and, in some cases,
recurrent tumors, is problematic owing to its low
specific to non-specific uptake ratio.5 Amino acids, on
the other hand, usually exhibit higher specific to non-
specific uptake ratios and are consequently making
major contributions to tumor detection by PET.6 As a
result, there has been a growing interest in the use
of 18F-labeled aromatic amino acids and PET for
tumor detection. In particular, [18F]6-fluoro-l-DOPA
Positron emission tomography (PET) is a non-invasive
imaging technique, which has been commonly used for
the in vivo visualization of brain functions such as
blood flow, metabolism, enzyme activity, neurorecep-
tors and neurotransporters.1 Fluorine-18-labeled 6-
fluoro-l-DOPA was initially developed as a routine PET
tracer to assess presynaptic dopaminergic function in
the human brain.2 Dopamine is synthesized in vivo by
the hydroxylation and the subsequent decarboxylation
of the amino acid l-tyrosine, stored intraneuronally in
vesicles from which it is ejected into the synapse during
neurotransmission. The monitoring of intracerebral
dopamine by PET requires the presence of dopamine
in the brain that is labeled with a positron emitter.
Dopamine, however, will not cross the blood–brain
barrier upon injection into the blood stream, and is
*Correspondence to: Raman Chirakal, Nuclear Medicine, McMaster
University Medical Centre, Hamilton Health Sciences, Hamilton, ON
L8N 3Z5, Canada and Gary J. Schrobilgen, Department of Chemistry,
McMaster University, Hamilton, ON L8S 4M1, Canada. E-mail:
chiraklr@mcmaster.ca (R.C.), schrobil@mcmaster.ca (G.J.S.)
Copyright # 2007 John Wiley & Sons, Ltd.

SYNTHESES OF 6-[18F] F-L AND DOPA IN CF3SO3H 1237
appears to have the potential to improve the detection
of neuroendocrine tumors and their metastases.6
Fluorine-18-labeled 6-fluoro-l-DOPA has also been
shown to be an excellent candidate for the visualization
of high-grade and low-grade tumors, and for the
analyses of recurrent low-grade gliomas, which are
difficult to examine by magnetic resonance imaging
and are usually not detected by 18F DG PET.5
method for the syntheses of d- and l-enantiomers
of 6-FDOPA is highly desirable to facilitate further
investigation of their biochemical properties in small
animal studies.
The use of elemental fluorine has been perceived as a
non-selective electrophilic fluorination method for aro-
matic compounds because the strong oxidant nature of
F2 results in exothermic radical chain reactions that
lead to the formation of side products that include tars.
Prior work from this laboratory has shown that direct
electrophilic fluorination of l-DOPA using F2 in aHF
produced 2-, 5- and 6-fluoro-l-DOPA having a total
radiochemical yield of 30% (decay corrected with
respect to [18F]F2),17 which is excellent for electrophilic
radiofluorinations because the theoretical maximum
radiochemical yield is 50% with respect to [18F]F2. The
highly efficient fluorination of l-DOPA and other
aromatic compounds in aHF is partly the result of the
low reaction temperatures employed and the high
solvent polarity, which favor electrophilic substitution
reactions. In addition, aromatic substrates in aHF are
less susceptible to oxidation by F2 because the catechol
oxygens are protonated.18 Chambers19 has shown that
protic solvents promote electrophilic fluorination of
aromatic compounds and that the reactivity of F2, as an
electrophile, varies significantly with solvent acidity. In
a prior work from this laboratory, the regioselectivities
of aromatic electrophilic fluorinations were shown to be
dependent upon the acidity of the reaction medium. For
example, although useful quantities of 18F-labeled
2- and 5-fluoro-l-DOPA have been prepared by the
direct fluorination of l-DOPA in various weak protic
acid solvents such as HCOOH, CH3COOH and
CF3COOH, 6-fluoro-l-DOPA could only be produced
when the superacid, aHF, was used as the solvent.20
The direct electrophilic fluorination of l-DOPA in BF3/
aHF, in particular, was shown to produce 2-, 5- and
6-fluoro-l-DOPA in a total radiochemical yield of 40%.21
Anhydrous HF does not readily lend itself to use in
most hospital environments because of its hazardous
nature and the specialized fluoroplastic equipment and
expertise required for its handling. An alternative route
to the production of 6-[18F]fluoro-l-DOPA and 6-
Although the applications of 6-fluoro-l-DOPA have
been well studied and are extensive, drawbacks still
remain because 3-O-methyl-[18F]6-FDOPA is formed in
the blood and brain by the action of catechol-O-methyl
transferase (COMT). It has been shown that the
presence of 3-O-methyl-[18F]6-FDOPA causes non-
specific accumulation of radioactivity in the brain and
blood, resulting in lower signal-to-noise ratios in PET
images.7 Because aromatic amino acid decarboxylase
and COMT are specific to l-DOPA, it is speculated that
[
18F]6-fluoro-d-DOPA may not be as extensively meta-
bolized in the brain, leading to higher signal-to-noise
ratios in tumor images when compared with 6-
[
18F]fluoro-l-DOPA. In recent in vivo and in vitro
studies, Bauwens et al.8 showed that the tumor-uptake
to background ratio of 123/125I-labeled 2-iodo-d-tyro-
sine is similar to that of its l-analogue in tumor cells
(in vitro in LAT1-expressing R1M rat rhabdomyosarco-
ma cells and in vivo in R1M tumor-bearing Wag/Rij
rats). These authors also reported similar results for
2-[123I]phenyl-d-alanine and its l-analogue.9 Further-
more, van Langevelde et al.10 have demonstrated that
14C- and 11C-labeled dl- and d-DOPA exhibit high
uptakes in the tumor tissue 1 h after intravenous
injection into Syrian golden hamsters with Greene
melanoma. Results from the above studies suggest
that both the d- and l-enantiomers of 6-[18F]FDOPA
may also display comparable uptakes in tumors.
Among the various methods that have been used for
the synthesis of 6-[18F]fluoro-DOPA,11,12 the most
common synthetic route is regioselective fluorodestan-
nylation,13 although, in the past, fluorodemercura-
tion14,15 reactions have also been extensively used.
Although both methods provide high radiochemical
yields, they require costly functionalized precursors,
prepared and purified through tedious multi-step
procedures, and considerable expertise. Moreover,
metal contamination of the product may occur, requir-
ing further purification prior to clinical use.16 Fluor-
odestannylation is preferred over fluorodemercuration
because of the toxicities associated with the mercury
derivatives employed in this procedure. Syntheses
using fluorodestannylation have the added drawback
of producing insoluble (CH3)3SnF which can obstruct
tubing and valves used in automated synthetic proce-
dures.16 Consequently, a simplified and more routine
[
18F]fluoro-d-DOPA by the direct radiofluorination of
their respective precursors in the superacidic medium,
trifluoromethanesulfonic acid (triflic acid, CF3SO3H),
that largely circumvents these difficulties is reported in
this article.
Results and discussion
In choosing an alternative solvent medium for electro-
philic fluorination of l-DOPA, a high solvent acidity is
required to facilitate and enhance the regioselectivity of
Copyright # 2007 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2007; 50: 1236–1242
DOI: 10.1002.jlcr

1238 B. BEHNAM AZAD ET AL.
the reaction. Triflic acid was selected because
of its high Hammett acidity (H0 ¼ ꢁ13:8), which places
it in the superacidic category along with aHF
(H0 ¼ ꢁ15:1),22 and its resistance to oxidation by F2.
Moreover, CF3SO3H has a favorable liquid range (f.p.
ꢁ40 to ꢁ458C; b.p. 1628C),23 both it and its conjugate
base are not sources of fluoride ions, even in the
presence of very strong Lewis acid fluoride ion accep-
tors, and the solubilities of amino acids are high in both
CF3SO3H and aHF, which is expected to promote
greater product yields. There have been numerous
applications of CF3SO3H in organic syntheses such as
salt formation, polymerization, ester formation and
Friedel–Crafts reactions.24 Moreover, Coe et al.25 used
10% CF3SO3H in CFCl3 solvent to promote electrophilic
fluorinations of aromatic compounds containing the
(CH3)3Si group. The use of CF3SO3H minimizes side
reactions encountered in direct electrophilic fluorina-
tions that utilize F2 and reduces the difficulties and
hazards associated with the use of aHF.23 Prior studies
from this laboratory have shown increased regioselec-
tive fluorination of the C6 position in electrophilic
fluorinations of l-DOPA, with increasing solvent acid-
ity.20 Consequently, in the present study, the direct
fluorination of l-DOPA was carried out in 99%
CF3SO3H to maximize the yield of 6-fluoro-l-DOPA.
Table 1 Radiochemical yields (RCY) of FDOPA resulting from
direct fluorination of DOPA in CF3SO3Ha
Temperature RCY
(8C)
Number of Relative ratioc
(%)b
trials
(2-, 5- and 6-FDOPA)
RT
4
ꢁ30
ꢁ40
19.3 ꢀ 2
15.6 ꢀ 3
11.4 ꢀ 2
14.5 ꢀ 2
6
6
2
2
38:15:47
36:7:57
36:5:59
33:0:67
a Reactions were carried out in 99% CF3SO3H.
CF3SO3H as a direct fluorination medium
The direct regioselective fluorination of DOPA in 99%
CF3SO3H (Table 1) produced clinically useful quanti-
ties of 18F-labeled 2- and 6-FDOPA. Fewer side
products and higher relative isomeric ratios of 6-
FDOPA (Table 1) were observed by preparative HPLC
(Figure 1) at lower reaction temperatures, suggesting
that 6-FDOPA is a kinetically favored product. It is
noteworthy that the highest radiochemical yield was
obtained when a 3:1 molar ratio of DOPA:F2 was used.
A typical fluorination of DOPA in CF3SO3H at 48C
resulted in a radiochemical yield of 19.7% compared
with 15.6% when a 1.6:1 molar ratio of DOPA:F2 was
used. The preparative HPLC findings are in agreement
with the analytical HPLC UV and radiochromatograms
obtained for the products of the direct fluorination of
l-DOPA in CF3SO3H at ꢁ408C (Figure 2).
Figure 1 Typical preparative HPLC radiochromatograms for
reaction mixtures resulting from the direct fluorination of l-
DOPA in 99% CF3SO3H at (a) ꢁ408C and (b) 48C. Small
differences in retention times result from minor variations in
mobile phase concentrations or flow rates. Peak splitting
resulted from detector saturation.
Reduction of the number of side products is a major
advantage of direct radiofluorination of l-DOPA in
CF3SO3H (Figure 1(a)) over radiofluorinations carried
out in BF3/aHF (Figure 4) or aHF15 and serves to
facilitate the isolation of 6-FDOPA from the final
reaction mixture. The use of CF3SO3H as the solvent
medium resulted in a decay-corrected total radio-
chemical yield of 19.7% for the FDOPA isomers, which
is sufficient for clinical use.
The 19F NMR spectra of 2-FDOPA (ꢁ139.6 ppm, broad
singlet, Dn1=2 ¼ 16 Hz), 5-FDOPA (ꢁ135.6ppm, doublet,
3J(F–H6) =11.3Hz) and 6-FDOPA (ꢁ126.4 ppm, doublet
of doublets, 3J(F–H5) ꢂ 4J(F–H2) = 9.0 Hz) are in agree-
ment with those reported in the literature.26 The peak
integrations corresponding to the FDOPA isomers in the
19F NMR spectra matched with those of the 18F-labeled
FDOPA isomers in the HPLC radiochromatograms.
Copyright # 2007 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2007; 50: 1236–1242
DOI: 10.1002.jlcr

SYNTHESES OF 6-[18F] F-L AND DOPA IN CF3SO3H 1239
anion exchange resin with an additional 5 mL of
0.1 M HCl, followed by HPLC analysis of the eluate.
Attempts to recover the remaining FDOPA from the
column by using higher HCl concentrations (0.5, 1 and
2 M) resulted in the co-elution of CF3SO3H with
FDOPA.
Other direct fluorination media
CF3SO3H/HCOOH. Because the direct removal of
CF3SO3H from the product under dynamic vacuum is
problematic owing to its low vapor pressure, solutions
of CF3SO3H in formic acid were also investigated in an
attempt to reduce the amount of CF3SO3H. Formic acid
has been used as a solvent in the selective fluorination
of substrates containing carbon centers of high elec-
tron density.27 In the present study, it was shown that
increasing the solvent acidity results in fewer side
products and higher yields of 6-FDOPA (Table 3). In
addition, smaller amounts of 5-FDOPA were produced
at higher concentrations of CF3SO3H which is signifi-
cant because 5-FDOPA cannot be fully separated from
6-FDOPA using HPLC, because of their very similar
retention times. As a result, CF3SO3H is preferred over
CF3SO3H/HCOOH as the solvent medium for the
production of 6-FDOPA.
BF3/CF3SO3H. The presence of a Lewis acid, such as
BF3 or AsF5, in aHF has been shown to increase
radiochemical yields of the monofluorinated amino
acids in direct fluorinations.21 In particular, BF3/HF
media have been utilized to significantly improve
radiochemical yields of 6-FDOPA.21 The radiofluorina-
tion of l-DOPA in BF3/CF3SO3H at 48C did not,
however, result in a significantly higher radiochemical
yield (Table 3). This is in agreement with a previous
study by Coenen et al.28 in which the use of BF3/
CF3COOH as a solvent medium for the fluorination of
phenylalanine did not result in a significant increase in
the radiochemical yield when compared with yields
obtained in CF3COOH.
Figure 2 Typical analytical HPLC UV (a) and radiochromato-
gram (b) for a reaction mixture resulting from the direct
fluorination of l-DOPA in 99% CF3SO3H at ꢁ408C.
Figure 3 shows a typical 19F NMR spectrum after
isolation of 6-FDOPA from the final product mixture
and Table 2 lists the relative intensities of fluorinated
species in the aforementioned sample. Higher purity 6-
FDOPA samples can be obtained when the correspond-
ing peak in the radiochromatogram is collected at
longer HPLC retention times; this, however, results in
lower activities.
Removal of CF3SO3H by evaporation proved to
be difficult and time consuming owing to its high
boiling point (1628C) and low vapor pressure (1 Torr at
428C).22 Anion exchange proved to be an efficient
method for the removal of CF3SO3H over a short time
period (55 min). Approximately 42 ꢀ 2% (decay cor-
Materials and methods
Reagents and chemicals
rected with respect to
[
18F]F2) of the theoretical
Enriched [18O]O2 18O, 99 atom%, Isotec), neon (Air
(
maximum activity was eluted from the column using
30 mL of 0.1 M HCl. The 19F NMR spectrum of the
eluate (pH, 4–5) indicated removal of most of the
triflic acid in the reaction mixture. The retention of a
small amount of FDOPA remaining on the column
(typically 50.5%) was confirmed by elution of the
Products, 99.999%), 1% F2 in neon (Canadian Liquid
Air), boron trifluoride (Matheson, 99.5%), helium
(Matheson, 99.9999%), CF3SO3H (Fluka, 99.8%) and
HPLC-grade CH3CN (Caledon) were used without
further purification and/or drying. Sterile, deionized
water was used in all aqueous procedures.
Copyright # 2007 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2007; 50: 1236–1242
DOI: 10.1002.jlcr

1240 B. BEHNAM AZAD ET AL.
Figure 3 Fluorine-19 NMR spectrum of 6-FDOPA isolated from the product mixture using preparative HPLC. Peaks B, C and E
correspond to 6-, 5- and 2-FDOPA, respectively. Small amounts of unidentified fluorinated side products were also observed (A and D).
The 19F NMR spectrum also showed two singlets at ꢁ75.7 (unidentified) and ꢁ79.9ppm (residual CF3SO3H) which are not shown.
Table 2 Relative intensities of fluorinated species in an
HPLC-purified 6-FDOPA sample resulting from direct fluor-
ination of DOPA in CF3SO3H at 48C
Fluorinated product
Relative ratioa
2-FDOPA
5-FDOPA
7.3
6.8
6-FDOPA
Combined unassigned byproducts
79.7
6.2
a Relative intensities were determined from integrated 19F NMR
spectra.
into a cryo-trap consisting of molecular sieves (Varian
1
˚
VacSorb, 5 A) contained in a 316 stainless steel Whitey
Figure 4 Typical preparative HPLC radiochromatograms for
reaction mixtures resulting from the direct fluorination of l-
DOPA in BF3/HF at ꢁ658C. Small differences in retention times
result from minor variations in mobile phase concentrations or
flow rates. Peak splitting resulted from detector saturation.
cylinder (75mL). The target was pumped to remove
trace amounts of [18O]O2 and subsequently filled with
1% F2 (40–50 mmol) in neon, pressurized to 20 atm with
neon, and irradiated for 10 min in a 15 mA proton beam
(recovery shoot). Portions of the [18F]F2/Ne mixture
were periodically released from the target into
a
Fluorine-18-labeled F2 was produced by the nuclear
reaction, 18O(p,n)18F, using a Siemens RDS 112 proton
cyclotron, operating at 11 MeV, and the ‘double shoot’
method29,30 in the Nuclear Medicine Department of
Hamilton Health Sciences. An aluminum target (11mL)
was pressurized to 14–16 atm with 99% enriched
continuous stream of helium until the target pressure
dropped to 2 atm. Helium was used as the sweep gas to
transfer [18F]F2 from the target into the hot cell.
Electrophilic fluorinations in CF3SO3H, CF3SO3H/
HCOOH and BF3/ CF3SO3H using [18F]F2 and F2
[
18O]O2 and irradiated for 20 min using a 30 mA proton
beam (production shoot). After irradiation, [18O]O2 was
DOPA (13 mg, 66 mmol), dissolved in 0.5 mL of the
recovered from the target by condensation at ꢁ1968C
appropriate solvent, was loaded into
a
5/16 in
Copyright # 2007 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2007; 50: 1236–1242
DOI: 10.1002.jlcr

SYNTHESES OF 6-[18F] F-L AND DOPA IN CF3SO3H 1241
Table 3 Radiochemical yield (RCY) of FDOPA resulting from direct fluorination of DOPA in CF3SO3H/HCOOH and BF3/CF3SO3H
Solvent
Temperature (8C)
RCYa (%)
Relative ratio (2-, 5- and 6-FDOPA)
99% CF3SO3H
4
4
ꢁ15
4
19.7
25.8
8.8
36:7:57
38:26:36
4:60:36
39:4:57
30% CF3SO3H in HCOOH
20% CF3SO3H in HCOOH
BF3/CF3SO3H
20.5
a Radiochemical yields have been decay corrected with respect to [18F]F2. A 3:1 molar ratio of DOPA:F2 was used in each reaction.
o.d. ꢃ 5/32 in i.d. FEP (tetrafluoroethylene/hexafluor-
opropylene copolymer) reaction vessel connected to an
FEP Y-piece. A 1/16 in o.d. ꢃ 1/32 in i.d. FEP tube,
connected to the [18F]F2 target at one end, was fed
through the sidearm of the Y-piece into the reaction
vessel. The other arm of the Y-piece was connected to a
separate 1/16 in o.d. FEP tube, which was immersed in
1 M NaOH. The reaction vessel and contents were
allowed to equilibrate at selected temperatures in a
liquid nitrogen cooled CH3OH bath. The BF3/CF3SO3H
solvent mixture was prepared by bubbling BF3 into the
CF3SO3H/DOPA solution at 48C until saturation was
achieved, and was followed by equilibration at 48C for
15 min prior to fluorination.
identified by injection of a standard solution that
eluted at 11 min. The peaks appearing at 12 and
14 min corresponded to those appearing in the radio-
chromatogram at 13 and 15 min, respectively. Each
peak was collected and assayed for radiochemical
yield. After a 24-h decay period, both HPLC collected
samples were combined and analyzed by 19F NMR
spectroscopy to obtain the relative molar amounts of
products which were shown to be 2- and 6-FDOPA. The
isomeric ratios of the mono-fluorinated aromatic amino
acids were determined by integrations of the HPLC
radiochromatogram peaks and 19F NMR spectra.
Fluorine-18-labeled DOPA isomers were also ana-
lyzed by the use of a reverse-phase preparative HPLC
column (Keystone Scientific, Inc., Bellefonte, PA, USA
16823, Fluophase PFP, 5 mm, 250 ꢃ 10 mm). A solution
of 17 mg of ascorbic acid in 500 mL of 0.1% CH3CO2H
was used as the mobile phase with a flow rate of
3.5 mL minꢁ1. The eluate from the column was mon-
itored using a UV detector set at 280 nm and a Geiger-
Fluorine-18-labeled F2 gas (typically 40 mmol) was
passed through a solution of DOPA in CF3SO3H and
the effluent gas was passed through 1 M NaOH before it
was vented into the hot cell. The amount of [18F]F2 that
had reacted was determined by counting the amount of
radioactivity present in the reaction mixture.
¨
Removal of residual CF3SO3H from the reaction
mixture was achieved using a 250 ꢃ 10 mm anion
exchange column (Bio-Rad AG 1–X8 in acetate form).
The reaction mixture was loaded onto the column and
20 mL of 0.1 M HCl was used as the eluent. The eluate
was then evaporated on a rotary evaporator and was
subsequently isolated and analyzed by preparative and
analytical HPLC, respectively.
Muller counter (Bicron SWGM B980C) coupled to a rate
meter (Bicron Erick-TechTM). The 2- and 6-FDOPA
isomers (typically appearing at 12 and 14 min on the
radiochromatogram) were collected, radio-assayed and
subsequently analyzed using 19F NMR spectroscopy.
Nuclear magnetic resonance spectroscopy
The 19F NMR spectra were recorded on a Bruker Avance
200 (4.6976 T) or DRX-500 (11.7440 T) spectrometer
using pulse widths of 1 ms corresponding to bulk
magnetization tip angles of ꢄ908. Typical 19F NMR
spectra, obtained at 11.7440 T, were accumulated over
spectral widths of 14 kHz (acquisition time, 1.16 s),
using 300 scans and 32 K memories, yielding data
point resolutions of 0.35 Hz/point. Fluorine-19 NMR
spectra, obtained at 4.6976 T, were accumulated over
spectral widths of 17 kHz (acquisition time, 0.94 s),
using 200 scans and 32 K memories, yielding data
point resolutions of 0.53 Hz/point. Spectra were refer-
enced at room temperature to external CFCl3. The
chemical shift convention used is that positive and
negative signs indicate chemical shifts to high and low
frequencies relative to that of the reference compound.
Analyses of reaction mixtures using HPLC
The ring-fluorinated isomers of d- and l-DOPA were
analyzed using a reverse-phase analytical HPLC col-
umn (Keystone Scientific, Inc., Bellefonte, PA, USA
16823, Fluophase PFP, 5 mm, 150 ꢃ 10 mm). A solution
of 0.2% CF3CO2H in water containing 7% CH3CN was
used as the mobile phase with
a flow rate of
2.5 mL minꢁ1. The column eluate was monitored by
using a Waters 490E Programmable Multi-wavelength
Detector set at 280 and 230 nm in conjunction with a
Beckman 170 Radioisotope Detector. A typical UV
chromatogram of the reaction mixture showed peaks
at 11, 12 and 14 min corresponding to DOPA, 2-FDOPA
and 6-FDOPA, respectively. The DOPA peak was
Copyright # 2007 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2007; 50: 1236–1242
DOI: 10.1002.jlcr

1242 B. BEHNAM AZAD ET AL.
Conclusions
9. Kersemans V, Cornelissen B, Kersemans K,
Bauwens M, Dierckx RA, De Spiegeleer B, Mertens
J, Slegers G. Eur J Nucl Med Mol Imaging 2006; 33:
919–927.
A fast, efficient and versatile method for the production
of 6-fluoro-l-DOPA and 6-fluoro-d-DOPA, in high
yields, in CF3SO3H has been developed as an
alternative to aHF for use in a hospital environment.
It has been shown that the direct fluorination of d-
and l-DOPA in CF3SO3H is a viable method for the
production of mono-fluorinated aromatic amino
acid isomers in clinically useful quantities. In a typical
fluorination reaction, 12 ꢀ 2 mCi of 6-[18F]fluoro-DOPA
was produced starting with 200 mCi of [18F]F2. The
radiochemical yield (17 ꢀ 2%) obtained from the cur-
rent method is not only sufficient for investigation of
the biochemical behavior of 6-[18F]fluoro-d-DOPA
using small animal imaging but is also sufficient for
clinical use in human subjects. The application of
6-fluoro-d-DOPA as a PET tracer for brain tumor
imaging is currently under investigation.
10. Van Langevelde A, Van Der Molen HD, Journee´-de
Korver JG, Paans AMJ, Pauwels EKJ, Vaalburg W.
Eur J Nucl Med 1988; 14: 382–387.
11. Snyder SE, Kilbourne MR. Chemistry of Fluorine-18
Radiopharmaceuticals. In Handbook of Radiophar-
maceuticals, Welch MJ, Redvanly CS (eds). John
Wiley and Sons, Ltd.: Chichester, UK, 2003; 195–
229.
12. Luxen A, Guillaume M, Melega WP, Pike VW, Solin
O, Wagner R. Nucl Med Biol 1992; 19: 149–158.
13. Namavari M, Bishop A, Satyamurthy N, Bida G,
Barrio JR. Appl Radiat Isot 1992; 43: 989–996.
14. Luxen A, Barrio JR, Bida GT, Satyamurthy N.
J Labelled Cmpd Radiopharm 1986; 23: 34–35.
15. Luxen A, Bida GT, Phelps ME, Barrio JR. J Nucl
Med 1987; 28: 624.
¨
16. De Varies EFJ, Luurtsema G, Brussermann M,
Acknowledgements
Elsinga PH, Vaalburg W. Appl Radiat Isot 1999; 51:
389–394.
We thank Dr Karen Gulenchyn, Chief of Nuclear
Medicine, and Ms Carol Dunne, Manager of Nuclear
Medicine, Hamilton Health Sciences, for granting
permission to use hospital facilities for this research.
We also thank Dr He´le`ne P. A. Mercier for her
assistance in preparing the figures for this article.
17. Firnau G, Chirakal R, Garnett ES. J Nucl Med
1984; 25: 1228–1233.
18. Chirakal R. Ph.D. Dissertation, McMaster Univer-
sity, Hamilton, Ontario, Canada, 1991.
19. Chambers RD. Fluorine In Organic Chemistry.
Blackwell Publishing, Ltd.: Oxford, UK, 2004.
20. Chirakal R, Vasdev N, Schrobilgen GJ, Nahmias C.
J Fluorine Chem 1999; 99: 87–94.
REFERENCES
1. Herscovitch P, Kimura Y, Senda M. Brain
Imaging Using PET. Academic Press: San Diego,
CA, 2002.
21. Chirakal R, Firnau G, Garnett ES. J Nucl Med
1986; 27: 417–421.
2. Nahmias C, Firnau G, Garnett ES. Nature 1983;
305: 137–138.
22. Gillespie RJ, Liang J. J Am Chem Soc 1988; 110:
6053–6057.
3. Firnau G, Garnett ES, Chirakal R, Sood S,
Nahmias C, Schrobilgen GJ. Appl Radiat Isot
1986; 37: 669–675.
23. Haszeldine RN, Kidd JM.
4228–4232.
J Chem Soc 1954;
24. Howells RS, Cown JD. Chem Rev 1977; 77: 69–90.
25. Coe PL, Stuart AM, Moody DJ. J Chem Soc Perkin
Transactions 1 1998; 11: 1807–1812.
4. Nahmias C, Schrobilgen GJ, Asselin M, Vasdev N,
Chirakal R. J Fluorine Chem 2002; 115: 33–39.
5. Chen WJ, Silverman DHS, Delaloye S, Czernin J,
Kamdar N, Pope W, Satyamurthy N, Schiepers C,
Cloughesy T. J Nucl Med 2006; 47: 904–911.
6. Becherer A, Szabo´ M, Karanikas G, Wunderbaldinger
P, Angelberger P, Raderer M, Kurtaran A, Dudczak R,
Kletter K. J Nucl Med 2004; 45: 1161–1167.
7. Firnau G, Sood S, Chirakal R, Nahmias C, Garnett
ES. J Nucl Med 1988; 29: 363–369.
26. Deng W, Wong KL, Kirk KL. Tetrahedron: Asymmetry
2002; 13: 1135–1140.
27. Chambers RD, Greenhall MP, Hutchinson J.
J Chem Soc Chem Comm 1995; 21–22.
¨
28. Coenen HH, Franken K, Kling P, Stocklin G. Appl
Radiat Isot 1988; 39: 1243–1250.
29. Nickels RJ, Daube ME, Ruth TJ. Int J Appl Radiat
Isot 1984; 35: 117–122.
8. Bauwens M, Lahoutte T, Kersemans K, Gallez C,
Bossuyt A, Mertens J. Nucl Med Biol 2006; 33:
735–741.
30. Chirakal R, Adams RM, Firnau G, Schrobilgen GJ,
Coates G, Garnett ES. Nucl Med Biol 1995; 22:
111–116.
Copyright # 2007 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2007; 50: 1236–1242
DOI: 10.1002.jlcr