Radiosyntheses and reactivities of novel [18F]2-fluoroethyl arylsulfonates

Article abstract of DOI:10.1002/jlcr.991

[¹⁸F]2-Fluoroethyl tosylate ([¹⁸F]FEOX, X = Ts) is widely used for labeling radiotracers for positron emission tomography (PET). Little work has been reported on syntheses of other [¹⁸F]2- fluoroethyl arylsulfonates ([

Full text of DOI:10.1002/jlcr.991

JOURNAL OF LABELLED COMPOUNDS AND RADIOPHARMACEUTICALS
J Label Compd Radiopharm 2005; 48: 735–747.
Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/jlcr.991
Research Article
Radiosyntheses and Reactivities of Novel [18F]2-fluoro-
ethyl arylsulfonates
John L. Musachio*, Jay Shah and Victor W. Pike
PET Radiopharmaceutical Sciences, Molecular Imaging Branch, National Institute of
Mental Health, National Institutes of Health, Building 10, Room B3 C346, 10 Center
Drive, Bethesda, MD 20892-1003, USA
Summary
[18F]2-Fluoroethyl tosylate ([18F]FEOX, X=Ts) is widely used for labeling radio-
tracers for positron emission tomography (PET). Little work has been reported on
syntheses of other [18F]2-fluoroethyl arylsulfonates ([18F]FEOX) that bear a less
electron-rich aryl group, even though these might offer enhanced reactivities. Thus, a
series of novel [18F]FEOX (X=benzenesulfonyl, brosyl, nosyl, 3,4-dibromobenzene-
sulfonyl) were synthesized and reactivities compared to [18F]FEOTs. Precursors for
radiolabeling (bis-ethylene glycol arylsulfonates) and reference FEOX were synthe-
sized (alcohol+arylsulfonyl chloride+KOSiMe3 in THF). Regardless of substitution
pattern, [18F]FEOX (1108C, 5 min, acetonitrile) were obtained in similar decay-
corrected isolated radiochemical yields (RCY; 47–53%). All [18F]FEOX gave excellent
RCYs (64–87%) of the dopamine uptake radioligand, [18F]FECNT (1308C, 10 min,
acetonitrile). The 3,4-dibromobenzensulfonate gave the highest RCY of [18F]FECNT
(87%) and this HPLC-purified labeling agent was used directly for efficient
[18F]FECNT production. When the secondary aniline of an amyloid probe (HM-
IMPY) or p-nitrophenol was reacted with [18F]FEOX, RCYs were appreciably higher
for brosylate and nosylate than for tosylate, while 3,4-dibromobenzenesulfonate again
gave the highest RCY. Owing to the high reactivity of the new [18F]FEOX and their
ease of syntheses via stable precursors, such agents (particularly 3,4-dibromobenze-
nesulfonate) should be considered as alternatives to [18F]FEOTs. Copyright # 2005
John Wiley & Sons, Ltd.
Key Words: fluorine-18; sulfonate ester; labeling agent
*Correspondence to: John L. Musachio, PET Radiopharmaceutical Sciences, Molecular Imaging Branch,
National Institute of Mental Health, National Institutes of Health, Building 10, Room B3 C346, 10 Center
Drive, Bethesda, MD 20892-1003, USA. E-mail: john.musachio@nih.gov
Copyright # 2005 John Wiley & Sons, Ltd.
Received 24 April 2005
Revised 19 May 2005
Accepted 23 May 2005

736
J. L. MUSACHIO ET AL.
Introduction
1
Fluorine-18 (t =109.8 min) is a radionuclide used extensively with positron
2
emission tomography (PET). The utility of 18F can be ascribed, in part, to
the numerous methods that allow for its efficient incorporation into a variety
of radiotracers. Of recent interest to our group are radiotracers that possess
an [18F]2-fluoroethyl group such as the dopamine uptake radioligand,
[18F]FECNT (3),1 and the b-amyloid probe, [18F]FEM-IMPY2 (4, Scheme 1).
[18F]2-Fluoroethyl tosylate (2e, Scheme 1) has featured as a labeling agent in
this work. This arylsulfonate ester has been used extensively owing to its ease
of radiosynthesis via a stable, commercially available precursor (ethylene
glycol di-p-tosylate), stability towards intermediate purification, and accep-
table reactivity with a variety of nucleophiles.3–6 Little work has been reported
on the synthesis of other [18F]2-fluoroethyl arylsufonates that bear a less
electron-rich aryl group, even though these might offer enhanced reactivities
towards weak nucleophiles. Here we report a series of such labeling agents.
Specifically, we provide details of the syntheses of the bis-ethylene glycol
arylsulfonate ester precursors used for radiolabeling as well as the syntheses of
the 2-fluoroethyl esters of benzenensulfonate (2a), brosylate (2b), nosylate (2c),
and 3,4-dibromobenzenesulfonate ester (2d) and their 18F-labeled counterparts
(Scheme 1). Of note, 2a, 2c, and 2d, are novel compounds that have not been
prepared with an 18F label.7 In this report we examine the reactivities of these
new 18F-fluoroethylating agents towards assorted nucleophiles (secondary
amine, aniline nitrogen, and phenol) in comparison to the widely used
arylsulfonate, [18F]2-fluoroethyl tosylate (Scheme 1). Finally, we detail a
highly efficient method for the radiosynthesis of [18F]FECNT where the
18F
H
N
N
COOMe
COOMe
Cl
Cl
OSO2
X
18F
3
6
[
18F]2a-2e
Y
I
I
N
N
N
N
X = Y = H, 2a
H
N
X = Br, Y = H, 2b
X = NO2, Y = H, 2c
X = Y = Br, 2d
N
4
7
18F
X = Me, Y = H, 2e
O2N
O
18F
O2N
OH
5
8
Scheme 1. Reaction of novel [18F]fluoroethyl arylsulfonates 2a–2d and widely
used 2e with assorted nucleophiles
Copyright # 2005 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2005; 48: 735–747

[
18F]2-FLUOROETHYL ARYLSULFONATES
737
reactive and lipophilic alkylating agent [18F]2d is quickly purified by reverse
phase HPLC and used directly to prepare [18F]FECNT in high radiochemical
yield and chemical purity.
Results and discussion
bis-Ethylene gylcol arylsulfonates (1a–1d) and 2-fluoroethyl arylsulfonates
(2a–2d) were prepared by treatment of the appropriate alcohol with
commercially available arylsulfonyl chloride using potassium trimethylsilano-
late as base in tetrahydrofuran at ice-bath temperature (Scheme 2).
Unoptimized isolated yields for 1a–1d and 2a–2d ranged from 32 to 64%.
We found the use of potassium trimethylsilanolate to be superior, in terms of
ease of work-up and yield, to the use of pyridine as previously reported in the
synthesis of 1c.8 bis-Sulfonate esters were stored at room temperature and
showed minimal decomposition over the course of this work (6 months) as
monitored by HPLC. To explore the reactivity of the bis-sulfonate esters (1a–
1e) in a uniform manner, a Synthia radiochemistry apparatus9 possessing an
x,y,z robotic arm, computer-controlled heaters, fraction collector, solid phase
extraction (SPE) module, HPLC injector/column switcher, and gas flow
controllers (He and N2) was employed. [18F]Fluoride ion contained in
[18O]water and potassium carbonate-Kryptofix1 222 was dried by cycles of
acetonitrile addition and evaporation at 1108C under a nitrogen stream
(200 ml/min). The dried, activated [18F]fluoride ion complex was subsequently
treated with 8.1 mmol of one of the precursors 1a–1e in acetonitrile (1108C,
5 min) and the labeled product isolated by reverse-phase HPLC. Average
decay-corrected isolated radiochemical yields of [18F]2a–2e ranged from 47 to
53% (range of individual yields: 45–55%), regardless of the substitution
pattern on the aryl ring of the sulfonate ester (Table 1). After HPLC
purification of each product [18F]2a–2e, the radioalkylating agents were
recovered (>85%) in a 1 ml volume of acetonitrile by SPE. The radiochemical
Y
X
O2SO
HO
F
X
SO2Cl
OSO2
X
OH
Y
1a-1d
or
Y
F
K+ -OSiMe3
THF
OSO2
X
OH
2a-2d
ice-bath
Y
X = Y = H, 1a or 2a
X = Br, Y = H, 1b or 2b
X = NO2, Y = H, 1c or 2c
X = Y = Br, 1d or 2d
Scheme 2. Syntheses of bis-ethylene glycol arylsulfonates (1a–1d) and authentic
2-fluoroethyl arylsulfonates (2a–2d)
Copyright # 2005 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2005; 48: 735–747

738
J. L. MUSACHIO ET AL.
Table 1. Average isolated, decay-corrected radiochemical yields (n53) of [18F]2-
fluoroethyl arylsulfonatesa
Precursor
[18F]Product
HPLC isolated radiochemical yield
(RCY; %)
1a
1b
1c
1d
1e
2a
2b
2c
2d
2e
53
51
47
53
53
a 8.1 mmol precursor + 18Fꢀ/K222/K2CO3 in 1 ml MeCN, 5 min, 1108C.
yields of [18F]2a–2e were not further optimized since the principal objective of
this work was to compare the reactivities of [18F]2a–2e under a standard set of
conditions.
The secondary amine, 2b-carbomethoxy-3b-(4-chlorophenyl)nortropane
(CNT, 6) (2.7 mmol), was reacted with each of the purified [18F]2-fluoroethyl
arylsulfonates under the standard conditions (acetonitrile, 1308C, 10 min in
open vessel). Resulting average isolated radiochemical yields for [18F]FECNT
3 are presented in Table 2 and range from 64 to 87%. Use of the novel 3,4-
dibromobenzenesulfonate ester, [18F]2d, afforded 3 in highest radiochemical
yield (87%). It was found that evaporation of the acetonitrile reaction solvent
to dryness in the 10 min reaction time span, thereby effectively increasing the
concentration of nucleophile, had a favorable effect on radiochemical yield.
When 18F-fluoroethylation of 6 was conducted in a septum-sealed V-vial using
a 9-fold greater amount of nucleophile (23.4 mmol) in about 1 ml of
acetonitrile, the radiochemical yields of 3 from [18F]2d and [18F]2e were only
54 and 18%, respectively (Table 2, entries 6 and 7). The feasibility of using an
‘open vessel’ radiolabeling procedure, which in the current case offers the
advantages of minimizing the amount of costly precursor and increasing
radiochemical yields, is dependent, in part, on the volatility of the 18F-
fluoroethylating species. To assess volatility in a practical manner, [18F]2a–2e
of known starting radioactivities were each heated in an open vessel, with
helium sweep gas (10 ml/min) to minimize bumping, under conditions identical
to those used to produce 3, but with nucleophile (e.g. 6) absent from the
acetonitrile solution. After the 10 min concentration procedure, residual
radioactivity was measured. The results (Table 3) revealed [18F]2c and [18F]2d
to be less volatile than the widely employed [18F]2e and in this regard to be
well-suited for open vessel radiolabeling procedures. Volatility differences
alone cannot account for increased radiochemical yields when using [18F]2d
compared to [18F]2e to produce 3; this is shown by the sealed reaction
radiolabelings that minimize volatility differences in the agents [18F]2d
and [18F]2e.
Copyright # 2005 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2005; 48: 735–747

[
18F]2-FLUOROETHYL ARYLSULFONATES
739
Table 2. Decay-corrected radiochemical yields of 3–6 upon reaction with novel
fluoroethylating agents ([18F]2a–2d) and fluoroethyl tosylate ([18F]2e)
Entry
Precursor
18F-labeling agent
[18F]Product
HPLC isolated RCY (%)a
1
2
3
4
6
6
6
6
6
6
6
6
6
7
7
7
7
7
8
8
8
8
8
2a
2b
2c
2d
2e
2d
2e
2d
2e
2b
2c
2d
2e
2d
2a
2b
2c
2d
2e
3
3
3
3
3
3
3
3
3
4
4
4
4
4
5
5
5
5
5
64
73
81
87
75
54
18
84
37
17
29
38
3
5
6b
7b
8c
9d
10
11
12
13
14e
15
16
17
18
19
5
23
34
50
63
19
a Average (n52).
b 23.4 mmol of 6 in 1 ml total volume in sealed V-vial.
c No SPE of [18F]2d (Method B, see experimental).
d No SPE of [18F]2e-isolated in 7:3 v/v acetonitrile–water.
e No SPE of [18F]2d-isolated in 95:5 v/v acetonitrile–water.
Table 3. Percentage of starting [18F]2-fluoroethyl arylsulfonate radio-
activity remaining in an open test tube upon heating 1 ml acetonitrile
solutiona
18F-labeling agent
Decay-corrected radioactivity remaining (%)
2a
2b
2c
2d
2e
11
23
68
67
12
a 10 min, 1308C with He sweep gas (10 ml/min).
Differences in the reactivities of [18F]2a–2e became more pronounced as the
aniline nitrogen of HM-IMPY 7 was employed as a weaker nucleophile, using
the same radiolabeling conditions described above for 3 (Table 2, entries
10–13). Again, use of [18F]2d resulted in the highest radiochemical yield of
4 (38%) while use of the tosylate [18F]2e gave the lowest yield (3%).
Intermediate radiochemical yields of 17 and 29% were noted for the brosylate
and nosylate 18F-labeling agents, respectively. In the isolation of 4, the only
Copyright # 2005 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2005; 48: 735–747

740
J. L. MUSACHIO ET AL.
other radioactive compound noted in the chromatogram was unreacted [18F]2-
fluoroethyl arylsulfonate ester. Increasing the precursor concentration and/or
prolonged heating, as reported previously,2 might result in even higher yields
of 4. Such experiments were not undertaken as it was our main focus to
demonstrate the enhanced reactivities of the novel 18F-alkylating agents versus
the widely used [18F]2e under a standard (but not fully optimized) set of
conditions.
A similar reactivity profile of [18F]2d>[18F]2c>[18F]2b>[18F]2a ꢁ [18F]2e
was observed when the agents were reacted with the deactivated phenol 8 to
give the radiolabeled ether 510 (Table 2, entries 15–19). An initial solvent
composition of acetoniltrile (1 ml) containing the 18F-labeling agent and DMF
(0.2 ml) containing 8 and 1 equivalent of methanolic tetrabutylammonium
hydroxide was used for production of 5.
Alternative methods of labeling by fluoroethylation using novel arylsulfo-
nate leaving groups and/or [18F]2a–d have yet to be explored. Such methods
include direct [18F]fluoride displacement of the mono arylsulfonate ester11,12 or
a one-pot procedure where the intermediate [18F]2-fluoroethyl arylsulfonate is
used without purification.2,12 In general, however, both of these alternative
methods may suffer from erratic yields and difficult HPLC purifications owing
to the copious amounts of byproducts formed.
For production of [18F]FECNT, the two-step radiolabeling sequence
involving synthesis and purification of a labeling agent [18F]2a–2e followed
by alkylation of CNT proved most reliable. In addition to the high reactivity
of [18F]2d, a distinct advantage of this labeling agent is that it may be used
directly for alkylation without intermediate SPE isolation. Owing to the high
lipophilicity of [18F]2d, this alkylating agent can be isolated within 6 min in a
high organic content HPLC mobile phase (9:1 v/v, acetonitrile–water) from a
semi-preparative C-18 column that is able to remove >99.9% of starting bis-
sulfonate ester 1d. It was found that the HPLC fraction containing [18F]2d
(ca 1.1 ml) could be combined directly with secondary amine 6 to give 3 in
radiochemical yield (84%, Table 2, entry 8), similar to the yield from the
method where HPLC-purified [18F]2d is first isolated in neat acetonitrile via
SPE (87%). For the purpose of comparison, direct reaction of the tosylate,
[18F]2e, in acetonitrile–water (7:3 v/v) with 6 gave 3 in only 37% radiochemical
yield (Table 2, entry 9). Solid phase extraction of radiolabeling agents often
requires manual manipulation of the radioactive solution, development of a
custom-made SPE module,13 or use of a commercial apparatus often
possessing a lone SPE module that is primarily intended for isolation of the
final radiopharmaceutical rather than a radiolabeled intermediate. In the
production of 3 via [18F]2d, the need for intermediary SPE can be avoided and
thus the radiochemistry sequence is simplified. Direct use of the HPLC
collected fraction of [18F]2d versus an intermediary SPE is not always
Copyright # 2005 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2005; 48: 735–747

[
18F]2-FLUOROETHYL ARYLSULFONATES
741
comparable. For example, it was found that reaction of the HPLC fraction
containing [18F]2d in 1 ml of acetonitrile–water (95:5 v/v) with 7 resulted in
lower radiochemical yield of 4 (5%, Table 2, entry 14) as compared to when
[18F]2d was isolated in neat acetonitrile via SPE (38%). Thus, the use of the
HPLC fraction of [18F]2d containing 5–10% water for direct reaction with
nucleophiles needs to be assessed on a case by case basis.
Conclusions
Novel 2-fluoroethyl arylsulfonates possessing non-electron-donating substitu-
tents about the aryl ring were synthesized and also labeled with fluorine-18
(2a–2d). The reactivities of the novel labeling agents were compared to the
widely used [18F]2-fluoroethyl tosylate (2e). Substituent pattern on the phenyl
ring had minimal effect on radiochemical yield of the radiolabeled
intermediates. The radiolabeled alkylating agents, prepared from stable, easily
synthesized bis-ethylene glycol arylsulfonates, showed enhanced reactivity
toward weak nucleophiles in comparison to 2e. In particular, [18F]2-
fluoroethyl 3,4-dibromobenzenesulfonate demonstrated high reactivity, low
volatility, and ease of purification making it an attractive alternative to [18F]2-
fluoroethyl tosylate.
Experimental
Materials
All chemicals were obtained from commercial sources and used as received
unless otherwise stated. Of note, 3,4-dibromobenzenesulfonyl chloride was
obtained from Oakwood Products Inc. (West Columbia, SC). Standard
samples of FECNT, 3, and CNT, 6, were obtained from Dr Mark Goodman
(Emory University). Authentic standards of FEM-IMPY, 4, HM-IMPY, 7,
2-fluoroethyl 4-nitro-phenyl ether, 5, and 2-fluoroethyl tosylate, 2e, were
available based on published methods from our laboratory.2,10
General methods
HPLC analysis was performed on a Beckman Coulter (Fullerton, CA) Gold
HPLC module (System Gold 126 gradient solvent module coupled with a
variable wavelength 166 UV absorbance detector). In-line HPLC detection of
radioactivity used a Bioscan (Washington, DC) flow count radioactivity
detector (Pin diode). Radioactivity was measured using a Biodex Medical
Systems (Shirley, NY) AtomlabTM 300 dose calibrator. Isolated radiochemical
yields reported in Tables 2 and 3 are decay-corrected and an average from 2 or
more reactions. Uncorrected melting points were determined with a capillary
apparatus (Mel-Temp; Barnstead Thermolyne, Dubuque, IA). Analytical TLC
was conducted on Macherey-Nagel silica gel 60 G-254 plates (250 mm; Alltech
Copyright # 2005 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2005; 48: 735–747

742
J. L. MUSACHIO ET AL.
Associates, Deerfield, IL). Normal phase column chromatography, using
disposable silica gel cartridges (12–40 mm ID ꢂ 150 mm length), was
performed with a Horizon high-performance flash chromatography (HPFC)
system (Biotage, Inc, Charlottsville, VA) equipped with UV detector (254 nm)
and fraction collector. Elemental analyses were determined by Midwest
Microlab (Indianapolis, IN). High-resolution mass spectrometry (HRMS) was
1
performed at the University of Minnesota Mass Spectrometry facility. H
NMR (400 MHz) and 13C NMR (100 MHz) spectra were obtained with a
Bruker instrument. Chemical shifts are reported in ppm (d) downfield relative
to the signal from internal tetramethylsilane.
No-carrier-added [18F]fluoride ion was prepared by the 18O(p,n)18F nuclear
reaction on a PETtrace cyclotron (GE Medical Systems, Milwaukee, WI). 18O-
enriched water (1.8 ml, >95% isotopic enrichment) was added to a titanium
target equipped with titanium foils. The [18O]water was bombarded (typically
20 mA ꢂ 120 min) with 18 MeV protons. Portions of the irradiated water (0.02–
0.5 ml) were used for individual experiments. All radiolabeling reactions were
conducted with the use of Synthia, a computer program driven synthesis
module, developed at Uppsala PET Center, that enabled fine control of
reaction parameters (e.g. oven temperature, reaction times, liquid transfers,
HPLC injections, SPE).
General method: syntheses of ethylenglycol-1,2 diarylsulfonates (1a ꢀ 1d) and
2-fluoroethyl arylsulfonates (2a ꢀ 2d)
To a round-bottomed flask cooled in an ice bath and kept under argon was
added either 0.125–0.25 ml ethylene glycol (2.23–4.46 mmol) for 1a–1d or
0.1 ml (1.7 mmol) 2-fluoroethanol dissolved in 2 ml anhydrous THF.
Arylsulfonyl chloride (2.5 equiv. for 1a–1d, 1.25 equiv. for 2a–2d) dissolved
in 3 ml THF was added. To this cooled solution was added potassium
trimethylsilanolate (5 equiv.) in portions over 30 min. The resulting slurry was
stirred for 2 h at ice-bath temperature. The slurry was poured into 100 ml
ice-cold water and the aqueous phase extracted with dichloromethane. The
organic layer was dried over sodium sulfate, filtered, and dichloromethane
removed by rotary evaporation. Final purification of the crude product was
performed by silica HPFC.
Ethyleneglycol-1,2-benezenesulfonate (1a). Gradient HPFC (100% hexane to
50:50 v/v hexane–dichloromethane) gave a colorless oil that solidified (mp 40–
428C) upon standing (63% yield). TLC 70:30 v/v hexane–ethyl acetate
1
(Rf=0.21). H NMR (400 MHz, CDCl3, d): 4.23 (s, 4H), 7.53–7.58 (m, 4H),
7.66–7.68 (m, 2H), 7.84–7.88 (m, 4H). 13C NMR (100 MHz, CDCl3, d): 67.03,
128.13, 129.58, 134.36, 135.58. HRMS-ESI: m/z+Na calculated, 365.0124;
Copyright # 2005 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2005; 48: 735–747

[
18F]2-FLUOROETHYL ARYLSULFONATES
743
found, 365.0136. Analytically calculated for C14H14S2O6: C, 49.11; H, 4.12.
Found: C, 49.27; H, 4.26.
Ethyleneglycol-1,2-4-bromobenezenesulfonate (1b). Gradient HPFC (50:50
hexane–dichloromethane to 100% dichloromethane) gave a white solid (mp
1458C) in 32% yield. TLC 20:80 v/v hexane–ethyl acetate (Rf=0.22). 1H NMR
(400 MHz, CDCl3, d): 4.26 (s, 4H), 7.71 (apparent s, 8H). 13C NMR
(100 MHz, CDCl3, d): 67.17, 129.58, 129.79, 132.96, 134.58. HRMS-ESI:
m/z+Na calculated, 520.8334; found, 520.8346. Analytically calculated for
C14H12Br2S2O6: C, 33.62; H, 2.42; Br, 31.95. Found: C, 33.78; H, 2.50; Br,
32.08.
Ethyleneglycol-1,2-4-nitrobenezenesulfonate (1c). Multiple extractions with
dichlormethane (6 ꢂ 50 ml) gave solid material that was purified by HPFC
(30:70 hexane–dichloromethane to 100% dichloromethane). A white solid (mp
1948C; literature8 mp 1868C) was isolated in 55% yield. TLC dichloromethane
1
(Rf=0.36). H NMR (400 MHz, CD3CN, d): 4.31 (s, 4H), 8.06 (dm, 4H,
J=9.3 Hz), 8.37 (dm, 4H, J=9.3 Hz). HRMS-ESI: m/z+Na calculated,
454.9826; found, 454.9828. Analytically calculated for C14H12N2S2O10: C,
38.89; H, 2.80; N, 6.48. Found: C, 39.25; H, 2.92; N, 6.50.
Ethyleneglycol-1,2-3,4-dibromobenezenesulfonate
(1d). Gradient
HPFC
(100% hexane to 100% dichloromethane) gave white solid (mp 1688C) in
43% yield. TLC 20:80 v/v hexane–dichloromethane (Rf=0.37). H NMR
1
(400 MHz, CDCl3, d): 4.31 (s, 4H), 7.64 (dd, 2H, J=8.5, 2.0 Hz), 7.83 (d, 2H,
J=8.5 Hz), 8.10 (s, 2H). 13C NMR (100 MHz, CDCl3, d) 67.22, 126.27, 127.34,
132.34, 132.66, 134.71, 135.59. HRMS-ESI: m/z+Na calculated, 676.6545;
found 676.6513. Analytically calculated for C14H10Br4S2O6: C, 25.56; H, 1.53;
Br, 48.58. Found: C, 25.64; H, 1.51; Br, 48.20.
2-fluoroethyl-benezenesulfonate (2a). Gradient HPFC (100% hexane to 50:50
hexane–ethyl acetate) gave a colorless oil in 64% yield. TLC 70:30 v/v hexane–
ethyl acetate (Rf=0.38). 1H NMR (400 MHz, CDCl3, d): 4.30 (m, 2H,
JHF=27.2 Hz), 4.59 (m, 2H, JHF=47.1 Hz), 7.58 (m, 2H), 7.68 (m, 1H), 7.95
(m, 2H). 13C NMR (100 MHz, CDCl3, d) 68.66 (JCF=21.2 Hz), 80.52
(JCF=174.1 Hz), 127.96, 129.36, 134.08, 135.64. HRMS-ESI: m/z+Na
calculated, 227.0149; found 227.0151. Analytically calculated for C8H9FO3S:
C, 47.05; H, 4.44. Found: C, 47.09; H, 4.26
2-Fluoroethyl-4-bromobenezenesulfonate ð2bÞ. Gradient HPFC (100% hexane
to 75:25 hexane–ethyl acetate) gave a colorless oil that solidified upon standing
solid (mp 498C) in 47% yield. TLC 70:30 v/v hexane–ethyl acetate (Rf=0.38).
1H NMR (400 MHz, CDCl3, d): 4.31 (m, 2H, JHF=27.2 Hz), 4.59 (m, 2H,
Copyright # 2005 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2005; 48: 735–747

744
J. L. MUSACHIO ET AL.
JHF=47.0 Hz), 7.72 (m, 2H), 7.80 (m, 12H). 13C NMR (100 MHz, CDCl3, d)
68.92 (JCF=21.2 Hz), 80.44 (JCF=174.1 Hz), 129.38, 129.43, 132.71, 134.68.
HRMS-ESI: m/z+Na calculated, 304.9254; found 304.9259. Analytically
calculated for C8H8BrFO3S: C, 33.94; H, 2.85; Br, 28.22. Found: C, 34.19; H,
2.87; Br, 28.27.
2-Fluoroethyl-4-nitrobenezenesulfonate (2c). Gradient HPFC (100% hexane
to 80:20 hexane–ethyl acetate) gave a white solid (mp 1188C) in 50% yield.
TLC 80:20 v/v hexane–ethyl acetate (Rf=0.21). H NMR (400 MHz, CDCl3,
1
d): 4.41 (m, 2H, JHF=30.0 Hz), 4.60 (m, 2H, JHF=45.0 Hz), 8.14 (d, 2H,
J=6.0 Hz), 8.41 (d, 2H, J=6.0 Hz). 13C NMR (100 MHz, CDCl3, d): 69.58
(JCF=20.5 Hz), 80.30 (JCF=174.1 Hz), 124.52, 129.32, 141.54, 150.90.
HRMS-CI: m/z+NH4 calculated, 267.0445; found 267.0460. Analytically
calculated for C8H8FNO5S: C, 38.56; H, 3.24; N, 5.62. Found: C, 38.46; H,
3.25; N, 5.49.
Synthesis of 2-fluoroethyl-3,4-dibromobenezenesulfonate (2d). Gradient HPFC
(100% hexane to 50:50 hexane–dichloromethane) gave a white solid (mp 68–
1
708C) in 57% yield. TLC 20:80 v/v hexane–dichloromethane (Rf=0.63). H
3
NMR (400 MHz, CDCl3, d): 4.35 (dt, 2H, JFH=26.0 Hz, JHH=4.0 Hz), 4.61
3
(dt, 2H, JFH=48.0 Hz, JHH=4.0 Hz), 7.71 (dd, 1H, J=8.4 Hz, 2.1 Hz), 7.83
(d, 1H, J=8.4 Hz) 8.16 (s, 1H). 13C NMR (100 MHz, CDCl3, d): 69.28 (JC-F
21.1 Hz), 80.35 (JC-F 174.1 Hz), 126.10, 127.39, 132.02, 132.73, 134.61, 135.99.
HRMS-ESI: m/z+Na calculated, 382.8359; found 382.8352. Analytically
calculated for C8H7Br2FSO3: C, 26.54; H, 1.95; Br, 44.14. Found: C, 26.62; H,
1.94; Br, 44.17.
General method: syntheses of [18F]2-fluoroethyl arylsulfonates ([18F]2a22e)
To a septum-sealed 5 ml V-vial containing 5 mg of Kryptofix1 222 and 0.5 mg
of K2CO3 dissolved in 0.5 ml of acetonitrile–water (98:2 v/v) was added
remotely [18F]fluoride ion (ca 740–7400 MBq) in 0.02–0.5 ml [18O]water. The
[18F]fluoride ion was dried by heating at 1108C with nitrogen sweep gas
(200 ml/min) and addition of three 1 ml aliquots of acetonitrile over about
15 min. To the activated [18F]fluoride ion complex, was added 8.1 mmol of a
precursor (one of 1a–1e) dissolved in 1 ml acetonitrile and the reaction heated
for 5 min at 1108C before addition of 1 ml of water. The quenched reaction
was injected onto a C-18 semi-preparative column (10 mm ꢂ 250 mm) using
either 10 m Luna1 (Phenomenex, Torrance, CA) or custom-packed 5 m
Adsorbosphere UHS (Alltech Associates, Deerfield, IL, Part No. C-6000F).
Columns were eluted with isocratic mixtures of 70–95% acetonitrile/water (4–
6 ml/min flow). Labeled product (one of [18F]2a–2e from 1a–1e, respectively)
was collected in 1–2 ml of mobile phase and radioactivity measured. Dilution
Copyright # 2005 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2005; 48: 735–747

[
18F]2-FLUOROETHYL ARYLSULFONATES
745
of product with 20 ml of water was followed by SPE using 200 mg of C-18
packing in a 3 cm3 vacuum column assembly (Waters, Milford, MA). The SPE
column containing the labeled product was dried with nitrogen gas (10 ml/
min ꢂ 0.5 min) before elution of the product in 1 ml acetonitrile.
Syntheses of [18F]2b-carbomethoxy-3b-(4-chlorophenyl)-8-(2-fluoroethyl)
nor-tropane ([18F]FECNT, [18F]3) via SPE isolated [18F]2a22e (Method A)
To a 12 mm ꢂ 50 mm test tube containing 0.75 mg (2.7 mmol) of 6 dissolved in
0.2 ml acetonitrile was added labeling agent (one of [18F]2a–2e) in 1 ml
acetonitrile and the radioactivity in the tube measured. To minimize bumping
of the solution in the open tube, a helium sweep gas (10 ml/min) was
introduced via disposable 3.5 in spinal needle. The reaction was heated for
10 min at 1308C at which time the oven was turned off and the reaction cooled
for 5 min. Three ml of 40:60 acetonitrile/10 mM aqueous ammonium
hydroxide pH 10.0 (initial mobile phase) was added to the tube and the crude
[18F]3 injected onto an XTerra1 Prep RP18 column (10 m; 7.8 mm ꢂ 300 mm).
A gradient HPLC method starting with initial mobile phase and ramping up to
70% acetonitrile over 30 min at 6 ml/min (UV=229 nm) was employed. Pure
[18F]3 (tR=19.2 min), well-resolved from 6 (tR=8.2 min), and labeling agent
(tRs=6.5, 10.9, 8.2, 14.8, 8.4 min for [18F]2a–2e, respectively) was collected in
about 4 ml of HPLC buffer, and the radioactivity assayed to determine
isolated radiochemical yield based on initial labeling agent contained in the
open tube.
Syntheses and formulation of [18F]FECNT ([18F]3) via direct collection of
[18F]2d (Method B)
Typical example: Starting with 4.44 GBq (120 mCi) of [18F]fluoride ion
1.97 GBq (53 mCi) of pure [18F]2d was prepared from 1d in 40 min, as
described above. Purified [18F]2d (90:10 v/v MeCN–water, 4 ml/min;
tR=5.3 min, k0=1.1; 1d tR=7.7 min, k0=2.1 on UHS Adsorbosphere) was
collected (ca 1.1 ml) directly into a 12 mm ꢂ 50 mm test tube containing
0.73 mg (2.6 mmol) of 6 dissolved in 0.1 ml acetonitrile. The 18F-fluoroethyla-
tion and HPLC purification were conducted as described in Method A to give
1.21 GBq (32.8 mCi) of [18F]3 that was isolated on a Waters 200 mg C-18 SPE
column and eluted in 1 ml of USP grade ethanol. Addition of 9 ml of normal
saline followed by sterile filtration through a Millipore Millex MP filter gave
1.10 GBq (29.7 mCi) of [18F]3 in a synthesis time of 97 min (46% decay-
corrected yield). Specific radioactivity, chemical purity, and radiochemical
purity of [18F]3 were assessed by analytical HPLC using a 4.6 mm ꢂ 250 mm
10 m C-18 Luna1 column. Mobile phase consisted of 50:50 v/v acetonitrile–
10 mM ammonium formate at 2 ml/min (absorbance detector l=229 nm).
Analytical HPLC analysis of a 0.1 ml aliquot of [18F]3 confirmed the
Copyright # 2005 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2005; 48: 735–747

746
J. L. MUSACHIO ET AL.
radioligand (tR=7.7 min) to be of high radiochemical and chemical purity
(>90%). Comparison of the UV carrier peak associated with [18F]3 to that of
a calibration curve generated with authentic 3 enabled calculation of specific
radioactivity that was 252 GBq/mmol (6810 mCi/mmol) at end of synthesis.
Syntheses of [18F]6-iodo-2-[40-N-(2-fluoroethyl)methylamino]phenyl-imida-
zo[1,2-a]pyridine, ([18F]FEM-IMPY, [18F]4), via SPE isolated [18F] 2b22e
Radiolabeled fluoroethylating agents [18F]2b–2e were prepared and isolated as
described above. Method A was followed for the radioalkylation reaction
using 0.97 mg (2.8 mmol) of HM-IMPY. Semi-preparative HPLC employed a
C-18 Luna1 column using gradient conditions (40:60 v/v acetonitrile–10 mM
ammonium hydroxide pH 9.5 to 90% acetonitrile over 30 min) at 6 ml/min
(absorbance detector l=254). Pure [18F]4 (tR=15.6 min), resolved from 7
(tR=11.8 min) and labeling agent (tRs=11.8, 9.4, 14.8 and 10.2 min for
[18F]2b–2e, respectively), was collected in about 4 ml of HPLC buffer. The
radioactivity was measured to determine isolated radiochemical yield of [18F]4,
based on starting labeling agent contained in the open tube.
Syntheses of [18F]2-fluoroethyl-4-nitrophenyl ether ([18F]5) via SPE isolated
[18F]2b22e
[18F]5 was prepared from [18F]2a–2e in a similar manner to [18F]4. Specifically,
1.2 mg (8.6 mmol) of 4-nitrophenol, 8, was dissolved in 0.2 ml DMF and then
8 ml of 1 M tetrabutylammonium hydroxide in methanol was added 30 s before
the addition of the 18F-fluoroethylating agent. Heating in an open vessel for
10 min at 1308C, as described above, was followed by HPLC purification.
When [18F]2d was used for production of [18F]5, an UHS Adsorbosphere
column running a gradient of 40% acetonitrile–water to 80% acetonitrile–
water over 30 min (4.5 ml/min; UV 254 nm) was employed ([18F]5
tR=17.3 min; [18F]2d tR=24.7 min; 8 tR=6.4 min). When [18F]2a–2c and
[18F]2e were used for production of [18F]5, a methanol gradient was employed
on the UHS Adorsbosphere column (35:65 v/v methanol–water to 80:20
methanol–water over 30 min, 3.5 ml/min; [18F]5 tR=27.5 min; 8 tR=15.3 min;
[18F]2a–2c, [18F]2e tRs=20.8, 29.0, 23.1 and 25.6 min, respectively).
Assessment of volatility of [18F]2a22e
Radiolabeling agents were synthesized as described above and isolated via SPE
in 1 ml of acetonitrile. The acetonitrile solution containing [18F]2a–2e was
transferred to a 12 mm ꢂ 50 mm test tube containing 0.2 ml acetonitrile
(precursor solvent blank) and the radioactivity assayed in a dose calibrator.
This open tube was heated at 1308C for 10 min with helium sweep gas (10 ml/
Copyright # 2005 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2005; 48: 735–747

[
18F]2-FLUOROETHYL ARYLSULFONATES
747
min) during which time the acetonitrile was evaporated. Residual radioactivity
left in the tube was measured to gauge the volatility of the [18F]2a–2e.
Acknowledgements
The authors wish to thank the following individuals in our branch:
Dr Umesha Shetty for assistance with NMR and Drs Lisheng Cai and
Shui-Yu Lu for provision of precursors and reference compounds (4, 7, 5).
References
1. Goodman MM, Kilts CD, Keil R, Shi B, Martarello L, Xing D, Votaw J, Ely TD,
Lambert P, Owens MJ, Camp VM, Malveaux E, Hoffman JM. Nucl Med Biol
2000; 27: 1–12.
2. Cai L, Chin FT, Pike VW, Toyama H, Liow J-S, Zoghbi SS, Modell K, Briard E,
Shetty HU, Sinclair K, Donohue S, Tipre D, Kung M-P, Dagostin C, Widdowson
DA, Green MA, Gao W, Herman MM, Ichise M, Innis RB. J Med Chem 2004;
47: 2208–2218.
3. Bauman A, Piel M, Schirrmacher R, Rosch F. Tetrahedron Lett 2003; 44:
¨
9165–9167.
4. Block D, Coenen H, Stocklin G. J Label Compd Radiopharm 1987; 24: 1029–1042.
¨
5. Wester H, Herz M, Weber W, Heiss P, Senekowitsch-Schmidtke R, Schwaiger M,
Stocklin G. J Nucl Med 1999; 40: 205–212.
¨
6. Block D, Coenen H, Stocklin G. J Label Compd Radiopharm 1988; 25: 201–216.
¨
7. [18F]2-Fluoroethyl brosylate has recently been reported in an abstract see Plisson
C, Jarkas N, McConathy J, Voll R, Votaw J, Williams L, Goodman M. J Nucl
Med 2004; 45S: 107P.
8. Schnekenburger J. Archiv der Pharmazie 1969; 302: 822–827.
9. Bjurling P, Reineck R, Westerberg G, Gee A, Sutcliffe J, Langstrom B. Presented
¨ ¨
at Proceedings of the VIth Workshop on Targetry and Target Chemistry,
TRIUMF, Vancouver, Canada, 1995; 282–284.
10. Lu S-Y, Chin FT, McCarron JA, Pike VW. J Label Compd Radiopharm 2004; 47:
289–297.
11. Zhang M-R, Maeda J, Ogawa M, Noguchi J, Ito T, Yoshida Y, Okauchi T,
Obayashi S, Suhara T, Suzuki K. J Med Chem 2004; 47: 2228–2235.
12. Wilson AA, DaSilva JN, Houle S. Appl Radiat Isot 1995; 46: 765–770.
13. Lemaire C, Plenevaux A, Aerts J, Del Fiore G, Brihaye C, Le Bars D, Comar D,
Luxen A. J Label Compd Radiopharm 1999; 42: 63–75.
Copyright # 2005 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2005; 48: 735–747