Synthesis of [18F]-6-deoxy-6-fluoro-D-glucose ([ 18F]6FDG), a potential tracer of glucose transport

Article abstract of DOI:10.1002/jlcr.1003

With the goal of developing a PET radioligand for the in vivo assessment of glucose transport, 6-deoxy-6-[¹⁸F]fluoro-D-glucose ([ ¹⁸F]6FDG) was prepared in two steps from ¹⁸F^-. Starting with D-glucose, the tosyl- and mesyl-derivatives of 3,5-O-benzylidene-1,2-O-isopropylidene-α-D-glucofuranose were prepared by known methods. Reaction of either of these precursors with ¹⁸F ^- resulted in the formation of 3,5-O-benzylidene-6-deoxy-O-[ ¹⁸F]-fluoro-1,2-O-isopropylidene-α-D-glucofuranose in high yield. Subsequent hydrolysis resulted in the production of [¹⁸F]6FDG. Under optimal conditions, [¹⁸F]6FDG is produced 60-70 min after end of bombardment (EOB) in 71 ± 12% yield (decay corrected, based upon fluoride) with a radiochemical purity of ≥96%. Preliminary experiments have indicated that [¹⁸F]6FDG may be a more representative in vivo tracer for the glucose transporter than 2FDG. Copyright

Full text of DOI:10.1002/jlcr.1003

JOURNAL OF LABELLED COMPOUNDS AND RADIOPHARMACEUTICALS
J Label Compd Radiopharm 2005; 48: 845–854.
Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/jlcr.1003
Research Article
Synthesis of [18F]-6-deoxy-6-fluoro-d-glucose
([18F]6FDG), a potential tracer of glucose transport
Timothy R. Neal1,*, William C. Schumann2, Marc S. Berridge1,3 and
Bernard R. Landau2
1 3D Imaging, LLC, 7650 First Place, Building B, Suite A, Oakwood Village,
OH 44146-6713, USA
2 Department of Medicine, Case Western Reserve University School of Medicine,
Cleveland, OH 44106, USA
3 Department of Chemistry, Case Western Reserve University, Cleveland,
OH 44106, USA
Summary
With the goal of developing a PET radioligand for the in vivo assessment of glucose
transport, 6-deoxy-6-[18F]fluoro-d-glucose ([18F]6FDG) was prepared in two steps
from 18Fꢀ. Starting with d-glucose, the tosyl- and mesyl-derivatives of 3,5-O-
benzylidene-1,2-O-isopropylidene-a-d-glucofuranose were prepared by known meth-
ods. Reaction of either of these precursors with 18Fꢀ resulted in the formation of
3,5-O-benzylidene-6-deoxy-6-[18F]-fluoro-1,2-O-isopropylidene-a-d-glucofuranose in
high yield. Subsequent hydrolysis resulted in the production of [18F]6FDG. Under
optimal conditions, [18F]6FDG is produced 60–70 min after end of bombardment
(EOB) in 71 ꢁ 12% yield (decay corrected, based upon fluoride) with a radiochemical
purity of 596%. Preliminary experiments have indicated that [18F]6FDG may be a
more representative in vivo tracer for the glucose transporter than 2FDG. Copyright
# 2005 John Wiley & Sons, Ltd.
Key Words: fluorine-18; 6-deoxy-6-[18F]fluoro-d-glucose; 2FDG; glucose transport
Introduction
Glucose is the main source of energy for mammalian cells and plays a central
role in the metabolic system responsible for normal biological function. The
in vivo assessment of glucose transport is of major clinical importance because
abnormalities in glucose transport have been associated with various
*Correspondence to: Timothy R. Neal, 3D Imaging, LLC, 7650 First Place, Building B, Suite A, Oakwood
Village, OH 44146-6713, USA. E-mail: tneal@cyclo-tech.com
Contract/grant sponsor: National Institutes of Health; contract/grant number: DK 14507
Copyright # 2005 John Wiley & Sons, Ltd.
Received 6 July 2005
Revised 20 July 2005
Accepted 26 July 2005

846
T. R. NEAL ET AL.
pathological conditions such as diabetes mellitus,1,2 Alzheimer’s disease,3–5
Huntington’s disease6 and the biochemistry of various tumors.7
The development of a suitable radioligand for PET would assist in the
assessment of glucose transporters in vivo, allowing investigation of the crucial
role of these transporters in glucose homeostasis. The assessment of glucose
uptake in tissues (transport and/or phosphorylation) is possible by imaging
with 2-deoxy-2-[18F]fluoro-d-glucose (2FDG). Results with 2FDG are some-
times contradictory because alteration of the transport and phosphorylation
steps may develop independently with different pathologies, even in offsetting
directions, while direct distinction between these two steps is not possible.8,9 In
one study of the effect of insulin on glucose blood-brain barrier transport,10
and in another study to estimate rates of transport in insulin resistance and
diabetes,11 it was necessary to account for the phosphorylation of 2FDG in
addition to transport. 3-O-Methyl-d-glucose (3-OMG) is a glucose analog that
enters the cell by glucose transport but is not phosphorylated, thus allowing
transport rates to be measured without the complications of subsequent
metabolism of the tracer.12–14 This tracer has been labeled with carbon-11 and
used for in vivo studies of glucose transport in the brain and the heart.15–19
Despite the favorable characteristics of 3-OMG, its usefulness is limited by
the short half-life of carbon-11 (20 min), which probably explains the small
number of studies with this tracer. Recently, the uptake of the a- and
b-anomers of 11C-methyl-d-glucose (MG) in mouse kidney has been
reported,20 and the potential of MG as a tracer for sodium-dependent glucose
transporters (SGLTs) merits further study.
Two radiolabeled versions (123I and 125I) of 6-deoxy-6-iodo-d-glucose
(6IDG) have been suggested as potential imaging agents of glucose
transport.21,22 It was reported that [123I]6IDG was responsive to insulin in
in vitro preparations, but not in isolated rat heart,21 and that [123I]6IDG was
able to detect a defect in glucose transport in db/db diabetic mice.22 However,
uptake in brain and heart was lower than that of 3-OMG23 and the usefulness
of 6IDG as a tracer of glucose transport in selected organs in clinical
applications has been questioned.24
Quite some time ago, one of us examined the structural requirements for
active transport of glucose by intestine25–27 and for stimulation of glucose
transport in muscle by insulin.28 2-Deoxyglucose (2DG) was shown responsive
to insulin but not transported by intestine, 3-OMG and 3-deoxyglucose were
responsive to insulin and actively transported by intestine and 6-O-
methylglucose was neither responsive to insulin nor transported by intestine.
Galactose, a-methyl-d-glucose, and 6-deoxy-6-fluoro-glucose (6FDG) were
found to be transported by intestine, but 6-deoxy-6-iodo-galactose was not.
Thus, the structural requirements for insulin stimulated transport by muscle
are different than for active transport by intestine. This work was done long
Copyright # 2005 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2005; 48: 845–854

SYNTHESIS OF [18F]6FDG
847
before the facilitated glucose transporters (GLUTs) in several tissues and the
SGLTs in the brush border of intestine and the proximal tubules were
characterized.
As we developed an interest in assessment of glucose transport in vivo, we
were led to consider 6-deoxy-6-[18F]fluoro-d-glucose ([18F]6FDG) as a
potential tracer for glucose transport. 6FDG is subject to transport (vide
supra), but because it lacks the 6-hydroxyl group, once inside the cell it cannot
undergo phosphorylation at this position, which occurs as the next step in
normal glucose metabolism. 6FDG competitively inhibits fermentation of
glucose by intact yeast, but not yeast extracts.29 Kinetic studies support the
hypothesis that 6FDG and 2DG compete with glucose at the same transport
site and that 6FDG does not affect any metabolic pathway in yeast cells.30
Several 6-deoxy-6-[18F]fluoro hexose analogs have been reported as potential
PET tracers (although not necessarily of glucose transport), including
6-deoxy-6-[18F]fluoro-l-fucose,31–33 6-deoxy-6-[18F]fluoro-d-galactose34–36
and 2,6-dideoxy-2-fluoro-6-[18F]-fluoro-b-d-glucopyranosyl fluoride.37 To our
knowledge, [18F]6FDG has not previously been reported.
Results and discussion
The preparation of unlabeled 6FDG by fluoride-ion displacement of a
sulfonate group from the tosyl- and mesyl-derivatives of 3,5-O-benzylidene-
1,2-O-isopropylidene-a-d-glucofuranose (5 and 6, respectively), followed by
acidic hydrolysis of the protecting groups has been reported.38,39 We prepared
5 and 6 from d-glucose by a slight modification of the published procedure,38
as outlined in Scheme 1. Reaction of d-glucose with acetone employing
sulfuric acid as catalyst formed 1,2–5,6 di-O-isopropylidene-d-glucofuranose,
which was not purified but immediately subjected to acid hydrolysis to
produce 1,2-O-isopropylidene-d-glucofuranose ð1Þ. Reaction of 1 with tosyl
chloride or mesyl chloride resulted in the formation of 6-deoxy-1,2-O-
isopropylidene-6-(40-methylbenzene)sulfonyloxy-d-glucofuranose ð2Þ or 6-deoxy-
1,2-O-isopropylidene-6-methylsulfonyloxy-d-glucofuranose ð3Þ, respectively. Diol
protection of 2 and 3 by reaction with benzaldehyde in the presence of zinc
chloride produced 3,5-O-benzylidene-6-deoxy-1,2-O-isopropylidene-6-(40-methyl-
benzene)sulfonyloxy-a-d-glucofuranose ð4Þ and 3,5-O-benzylidene-6-deoxy-1,2-
O-isopropylidene-6-methylsulfonyloxy-a-d-glucofuranose ð5Þ, respectively.
For convenience, reaction conditions typically used for the preparation of
2FDG from 1,3,4,6-tetra-acetyl-2-trifluoromethylsulfonyl-b-d-mannopyra-
nose were examined to determine the suitability for production of
[18F]6FDG from 4 and 5. Reaction of either precursor with 18Fꢀ resulted in
the formation of the fluorinated intermediate 638,39 in high yield. Based upon a
very limited number of experiments, there was no observable difference in the
Copyright # 2005 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2005; 48: 845–854

848
T. R. NEAL ET AL.
HO
1) Acetone,
H2SO4
O
3) RSO2Cl
HO
HO
D-Glucose
2) H+
O
O
1
R
R
O
O
S
S
O
O
O
O
O
4) C6H5CHO
ZnCl2
O
O
HO
O
O
O
HO
O
O
2 (R = p-tolyl)
3 (R = CH3)
4 (R = p-tolyl)
5 (R = CH3)
Scheme 1. Formation of the tosyl- and mesyl-derivatives of 3,5-O-benzylidene-
1,2-O-isopropylidene-a-d-glucofuranose from d-glucose
yield of 6 from the tosylate 4 or the mesylate 5; subsequent experiments
employed 5 as precursor. Because of the relatively high yield from these initial
experiments, the only further optimization of the synthetic procedure we
undertook was to determine the minimum necessary amount of precursor in
the reaction. When 5 mg precursor ðn ¼ 3Þ was used, 6 was obtained in
26 ꢁ 6% chemical yield (range 17–29%); increasing the amount of precursor
to 7.5–10 mg resulted in the production of 6 in 75 ꢁ 12% yield (decay
corrected, based upon fluoride, n ¼ 4, range 58–86%).
Hydrolysis of 6 for 10 min in 1 N HCl at 1108C resulted in the formation of
[18F]6FDG. The hydrolysis yield was separately determined for only two
experiments; yields of 69 and 95% were observed. It is likely that the 69%
represents a lower limit of expected hydrolysis yield. The overall yield obtained
from several subsequent experiments (vide infra) suggests the hydrolysis
typically proceeds in high yield. Purification was achieved by passing the
reaction mixture sequentially over small columns packed with C18 reversed-
phase silica and neutral alumina. Product purity was determined by TLC and
HPLC analysis. The [18F]6FDG produced by this sequence was compared to
an authentic sample of unlabeled 6FDG (Sigma Chemical). Analysis by TLC
(co-spotting with labeled product) and HPLC found the authentic 6FDG to be
identical to the labeled material. The conversion of 5 to [18F]6FDG is shown in
Scheme 2.
Copyright # 2005 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2005; 48: 845–854

SYNTHESIS OF [18F]6FDG
849
18F
H3C
O
O
S
18F
O
O
O
O
HO
HO
H+
O
18F-
O
O
O
O
O
OH
OH
O
O
[
18F]6FDG
5
6
Scheme 2. Conversion of 5 to [18F]6FDG
Under optimal conditions, [18F]6FDG was obtained in two steps from 18Fꢀ
in 71 ꢁ 12% chemical yield (n ¼ 3, range 58–80%). In one experiment starting
with 1.37 GBq (37 mCi) of 18Fꢀ, 0.733 GBq (19.8 mCi) of [18F]6FDG (54%
radiochemical yield, 80% chemical yield, based upon fluoride) was obtained
64 min after EOB with a radiochemical purity of 98%. Several production runs
have been carried out remotely in a hot cell yielding 4.81–16.3 GBq (130–
440 mCi) of [18F]6FDG from 18.5 to 42.9 GBq (500–1160 mCi) of 18Fꢀ
approximately 60–70 min after EOB in 596% radiochemical purity. By
neutralization of the acidic hydrolysis reaction mixture with sodium hydroxide
and sodium bicarbonate, dilution with sterile water for injection (USP) and
sterile filtration through a 0.22 mm filter, the final product is obtained as a
sterile, pyrogen-free and slightly hypertonic bicarbonate buffered saline
solution, ready for injection. The final solution contains about 26 mmol of
glucose from hydrolysis of the excess precursor. Under the standard reaction
conditions, the glucose concentration in the final solution is approximately
8 mM (compared to 5.5 mM for the normal physiological concentration of
glucose). The low level of glucose present is unlikely to have any consequence
on experiments conducted with the 6FDG, especially considering that the
solution is typically diluted several fold before administration.
Preliminary experiments to determine 6FDG uptake were carried out under
basal conditions using two cell lines, Clone 9 cells40,41 and 3T3 L1
adipocytes.42 Uptake of 6FDG in Clone 9 cells was stimulated by azide
addition; in 3T3 L1 adipocytes uptake was stimulated by insulin addition. In
each cell line, uptake of 6FDG was found to be inhibited by cytochalasin B.
The uptake of 6FDG was the same as uptake of 3-OMG (Dr F. Ismail-Beigi,
personal communication). Glucose uptake in Clone 9 cells is by GLUT 1
transporters and in 3T3 L1 adipocytes GLUT 1 and GLUT 4 transporters;
cytochalasin B is a specific inhibitor of the glucose transporters.
In other preliminary experiments in rats scanned in a small animal PET
scanner, [18F]6FDG was shown to lack the bladder uptake that is so
characteristic of 2FDG images to such an extent that it was not possible to
define the bladder on the images (Dr RF Muzic Jr, personal communication).
Copyright # 2005 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2005; 48: 845–854

850
T. R. NEAL ET AL.
This implies that, in contrast to 2FDG, [18F]6FDG is a good substrate for the
renal glucose transporter. In rats treated with phlorizin, which blocks the
transporter and has no effect upon a 2FDG image, the [18F]6FDG uptake in
the bladder was similar to that of 2FDG.
These preliminary experiments, which will be completed and reported in
detail elsewhere, indicate that [18F]6FDG may be an effective substrate for
glucose transport and would be a useful tracer for the assessment of alterations
in glucose transporters.
Experimental
Reagents and solvents were obtained from Sigma-Aldrich Chemical Co. and
Fisher Scientific and used without further purification unless otherwise noted.
Acetonitrile was freshly distilled from calcium hydride prior to use. Thin layer
chromatography (TLC) was carried out on E. Merck silica gel 60 F254
analytical plates. Radiation detection was performed with a Bioscan model
AR-2000 thin layer plate reader using WinScan Version 3.07 software, and
mass detection by UV light or in some cases acid charring. High performance
liquid chromatography (HPLC) was performed on a Hewlett Packard 1090
system with a photodiode array detector, a Beckman model 170 radiation
detector and in some cases a Knauer model 198 refractive index detector.
Elution conditions for TLC and HPLC varied and are specified individually
for each compound below. 3,5-O-benzylidene-6-deoxy-1,2-O-isopropylidene-
6-(40-methylbenzene)sulfonyloxy-a-d-glucofuranose and 3,5-O-benzylidene-6-
deoxy-1,2-O-isopropylidene-6-methylsulfonyloxy-a-d-glucofuranose (4 and 5,
respectively) were prepared from d-glucose as previously reported,38 with the
exception that the reaction of glucose and acetone was catalyzed by sulfuric
acid rather than ZnCl2 (this alternate procedure was mentioned in the
discussion section of the publication).
3; 5-O-benzylidene-6-deoxy-6-[18F]-fluoro-1,2-O-isopropylidene-a-d-glucofura-
nose (6)
18Fꢀ (obtained by 17 MeV proton bombardment of [18O]H2O43) was passed
through anion exchange resin and [18O]H2O was recovered. 18Fꢀ was eluted
with aqueous potassium carbonate. An aliquot of this solution containing the
desired quantity of radioactivity was transferred to a reaction vial containing
ca. 12 mg (31.9 mmol) 4,7,13,16,21,24-hexaoxa-1,10-diazabicyclo[8,8,8]hexaco-
sane (Kryptofix 2.2.2, Kryptofix). If necessary, aqueous potassium carbonate
solution was added in order to achieve the desired Kryptofix/K2CO3 mass
ratio (5–6). The water was evaporated under reduced pressure at 1108C and
dried by coevaporation with acetonitrile. A solution of 5–10 mg 3,5-O-
benzylidene-6-deoxy-1,2-O-isopropylidene-6-(40-methylbenzene)sulfonyloxy-a-
d-glucofuranose (4, 10.8–21.6 mmol) or 3,5-O-benzylidene-6-deoxy-1,2-O-
Copyright # 2005 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2005; 48: 845–854

SYNTHESIS OF [18F]6FDG
851
isopropylidene-6-methylsulfonyloxy-a-d-glucofuranose (5, 12.9–25.9 mmol)
dissolved in 100 ml anhydrous acetonitrile was added and the solution was
heated for 10 min at 1108C. The reaction mixture was concentrated with gentle
heating under a stream of Ar gas. The residue was mixed with 300 ml of ethyl
acetate, followed by 900 ml of hexanes. The solution was passed over a small
(ca. 150 mg) column of silica gel. The process was repeated with another 300 ml
of ethyl acetate and 900 ml of hexanes. The solution was analyzed by TLC
(developed twice with 4:1 hexanes:ethyl acetate; Rf ¼ 0:48).
Hydrolysis of 3,5-O-benzylidene-6-deoxy-6-[18F]-fluoro-1,2-O-isopropylidene-
a-d-glucofuranose ð6Þ
The solution of 6 described above was concentrated under a stream of Ar. To
the dried residue was added 250 ml of 1 N HCl and the mixture was heated at
1108C for 10 min. After cooling, 125 ml of 2 N NaOH and 250 ml of 1 N
NaHCO3 were sequentially added and this solution was passed through a
small column containing ꢂ200 mg each of C18 reversed-phase silica and
neutral alumina separated by a plug of glass wool. The reaction vial and small
column were rinsed with 2 ml of water. The solution was analyzed by TLC (9:1
acetonitrile:water; Rf ¼ 0:45) and HPLC (4.6 ꢃ 250 mm Alltech Econosphere
NH2 5 m column, eluting with 90:10:0.01 acetonitrile:H2O:40% aqueous
CH3NH2 at 1 ml/min; RT ¼ 8 min). For comparison purposes, authentic
6FDG was analyzed by TLC (co-spotted with labeled product, detected by
acid charring) and HPLC (refractive index detector) and found to have the
same Rf and RT as the hydrolysis product.
6-deoxy-6-[18F]fluoro-d-glucose ([18F]6FDG)
An aliquot containing 1.37 GBq (37 mCi) of 18Fꢀ was added to a vial
containing 12.4 mg (32.9 mmol) Kryptofix and 1.9 mg (13.7 mmol) K2CO3. The
water was evaporated under reduced pressure at 1108C and dried by
coevaporation with acetonitrile. A solution of 10.2 mg (26.4 mmol) 3,5-O-
benzylidene-6-deoxy-1,2-O-isopropylidene-6-methylsulfonyloxy-a-d-glucofur-
anose (6) dissolved in 200 ml anhydrous acetonitrile was added and the solution
was heated for 10 min at 1108C. The reaction mixture was cooled, mixed with
750 ml of ethyl acetate and passed over a small (ꢂ150 mg) column of silica gel.
The silica column was eluted with a second 750 ml aliquot of ethyl acetate. The
solution was concentrated to dryness under vacuum at 808C. To the residue
was added 500 ml of 1 N HCl and this mixture was heated 10 min at 1108C. The
reaction mixture was cooled and 250 ml of 2 N NaOH and 500 ml of 1 N
NaHCO3 were sequentially added. This solution was passed over a small
column containing ꢂ200 mg each of C18 reversed-phase silica and neutral
alumina (separated with a glass wool plug) through a sterile filter into a
Copyright # 2005 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2005; 48: 845–854

852
T. R. NEAL ET AL.
suitable collection vial. The column was eluted with an additional 2 ml of
sterile water to yield 0.733 GBq (19.8 mCi) of 6-deoxy-6-[18F]fluoro-d-glucose
as a sterile, pyrogen-free solution (by USP sterility and pyrogen testing
procedures) 64 min after EOB (54% radiochemical yield, 80% chemical yield,
based upon fluoride, 98% radiochemical purity). The solution was analyzed by
TLC and HPLC as described above.
Conclusion
6-deoxy-6-[18F]fluoro-d-glucose ([18F]6FDG) was prepared in two steps from
18Fꢀ. Starting with d-glucose, the tosyl- and mesyl-derivatives of 3,5-O-
benzylidene-1,2-O-isopropylidene-a-d-glucofuranose were prepared by known
methods. Reaction of either of these precursors with 18Fꢀ resulted in the
formation of 3,5-O-benzylidene-6-deoxy-6-[18F]-fluoro-1,2-O-isopropylidene-
a-d-glucofuranose in high yield. Subsequent hydrolysis resulted in the
production of [18F]6FDG. Under optimal condition, [18F]6FDG is produced
in 71 ꢁ 12% radiochemical yield (decay corrected, based upon fluoride) with a
radiochemical purity of 596% 60–70 min after EOB. Preliminary evaluation
of 6FDG uptake by cell preparations in vitro and [18F]6FDG in PET scans of
rats indicates that 6FDG may be a good substrate for the glucose transporters
and would be useful in assessment of variations in glucose transport
independent of glucose phosphorylation.
Acknowledgements
This work was supported in part by Grant DK14507 from the National
Institutes of Health. Our thanks go to Bonnie Bennington for conducting the
USP sterility and pyrogen testing procedures, to D. Rubin, N. Rachdaoui, and
F. Ismail-Beigi for the cellular uptake experiments and to Chandra Spring-
Robinson, Cristian A. Salinas and Raymond F. Muzic Jr for the animal PET
studies.
References
1. Bonadonna RC, Del Prato S, Saccomani MP, Bonora E, Gulli G, Ferrannini E,
Bier D, Cobelli C, DeFronzo RA. J Clin Invest 1993; 92: 486–494.
2. Garvey W-T, Maianu L, Zhu J-H, Brechtel-Hook G, Wallace P, Baron A-D.
J Clin Invest 1998; 101: 2377–2386.
3. Jagust WJ, Seab JP, Huesman RH, Valk PE, Mathis CA, Reed BR, Coxon PG,
Budinger TF. Cereb Blood Flow Metab 1991; 11: 323–330.
4. Kalaria RN, Harik SI. J Neurochem 1989; 53: 1083–1088.
5. Simpson IA, Chundu KR, Davies-Hill T, Honer WG, Davies P. Ann Neurol 1994;
35: 546–551.
Copyright # 2005 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2005; 48: 845–854

SYNTHESIS OF [18F]6FDG
853
6. Gamberino WC, Brennan WA. J Neurochem 1994; 63: 1392–1397.
7. Hawkins RA, Hoh C, Dahlborn M, Choi Y, Glaspy J, Tse N, Slamon D, Chen B,
Messa C, Maddahi J, Phelps ME. J Nucl Med 1991; 32: 1555–1558.
8. Wahl R-L. J Nucl Med 1996; 37: 1038–1041.
9. Garlick P-B, Medina R-A, Southworth R, Marsden P-K. Eur J Nucl Med 1999;
26: 1353–1358.
10. Hasselbalch SG, Knudsen GM, Videbaek C, Pinborg LH, Schmidt JF, Holm S,
Paulson OB. Diabetes 1999; 48: 1915–1921.
11. Williams KV, Bertoldo A, Kinahan P, Cobelli C, Kellet DE. Diabetes 2003; 52:
1619–1626.
12. Barnett JEG, Holman GD, Munday KA. Biochem J 1973; 131: 211–221.
13. Csaky T-Z, Wilson J-E. Biochim Biophys Acta 1956; 22: 185–186.
14. Kloster G, Muller-Platz C, Laufer P. J Label Compd Radiopharm 1981; 18: 855–863.
¨
15. Vyska K, Freundlieb C, Hock A, Becker V, Schmid A, Feinendegen LE, Kloster
¨
G, Stocklin G, Heiss WD. In Radioaktive isotope in klinik und forschung, Hofer R,
¨ ¨
Bergman H, Ergmann H (eds). Gasteiner Internationals: Vienna, 1982; 129–142.
16. Vyska K, Magloire JR, Freundlieb C, Hock A, Becker V, Schmid A, Feinendegen
¨
LE, Kloster G, Stocklin G, Schuier FJ, Thal HU. Eur J Nucl Med 1985; 11:
¨
97–106.
17. Gjedde A, Wienhard K, Heiss W-D, Kloster G, Diemer NH, Herholz K, Pawlik
G. J Cereb Blood Flow Metab 1985; 5: 163–178.
18. Feinendegen L-E, Herzog H, Wieler H, Patton D-D, Schmid A. J Nucl Med 1986;
27: 1867–1877.
19. Brooks DJ, Veaney RJP, Lammertsma AA, Herhold S, Turton DR, Luthra SK,
Frackowiak RS, Thomas DGT, Marshall J, Jones T. J Cereb Blood Flow Metab
1986; 6: 230–239.
20. Bormans GM, Van Oosterwyck GV, de Groot TJ, Veyhl M, Mortelmans L,
Verbruggen AM, Koepsell H. J Nucl Med 2003; 44: 1075–1081.
21. Henry C, Koumanov F, Ghezzi C, Morin C, Mathieu J-P, Vidal M, de Leiris J,
Comet M, Fagret D. Nucl Med Biol 1997; 24: 527–534.
22. Perret P, Ghezzi C, Mathieu JP, Morin C, Fagret D. Diabetes Metab Res Rev
2003; 19: 306–312.
23. Gatley SJ. J Nucl Med 2003; 44: 1082–1086.
24. Malaisse WJ, Malaisse-Lage F, Ladriere L, Schoutens A. Int J Mol Med 2000; 6:
65–68.
25. Wilson TH, Landau BR. Am J Physiol 1960; 198: 99–102.
26. Landau BR, Bernstein L, Wilson TH. Am J Physiol 1962; 203: 99–102.
27. White LW, Landau BR. J Clin Invest 1965; 44: 1200–1213.
28. Landau BR, Ship AG, Levine H. Am J Physiol 1958; 193: 461–465.
29. Blakley ER, Boyer PD. Biochim Biophys Acta 1955; 16: 576–582.
30. Blakley ER, Myoda T, Spencer JFT. Can J Biochem 1966; 44: 927–935.
31. Ishiwata K, Tomura M, Ido T, Iwata R, Sato K, Hatazawa J, Kameyama M,
Imahori Y. J Nucl Med 1990; 31: 1997–2003.
32. Ishiwata K, Ido T, Tomura M, Iwata R, Takahashi T, Kameyama M. Acta Radiol
Suppl 1991; 376: 144–145.
Copyright # 2005 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2005; 48: 845–854

854
T. R. NEAL ET AL.
33. Ishiwata K, Takahashi T, Iwata R, Tomura M, Tada M, Itoh J, Kameyama M,
Ido T. Nucl Med Biol 1992; 19: 611–618.
34. Christman DR, Orhanovic Z, Wolf AP. J Label Compd Radiopharm 1977; 13: 283.
35. Christman DR, Orhanovic Z, Shreeve WW, Wolf AP. J Label Compd
Radiopharm 1977; 13: 555–559.
36. Ishiwata K, Tomura M, Ido T, Iwata R, Itoh J, Kameyama M. Nucl Med Biol
1989; 16: 775–781.
37. Wong AW, Adam MJ, Withers SG. J Label Compd Radiopharm 2001; 44:
385–394.
38. Blakley ER. Biochem Preparations 1960; 7: 39–44.
39. Nieschalk J, O’Hagan D. J Fluorine Chem 1998; 91: 159–163.
40. Shetty M, Loeb JN, Vikstrom K, Ismail-Beigi F. J Biol Chem 1993; 268:
17225–17232.
41. Hamrahian AH, Zhang JZ, Elkhairi FS, Prasad R, Ismail-Beigi F. Arch Biochem
Biophys 1999; 368: 375–379.
42. Rachdaoui N, Nagy LE. Am J Physiol 2003; 285: E545–E551.
43. Berridge MS, Kjellstrom R. Appl Radiat Isot 1999; 50: 699–705.
Copyright # 2005 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2005; 48: 845–854