Synthesis, radio-LC-MS analysis and biological evaluation of 99mTc-techmazenil

Article abstract of DOI:10.1002/jlcr.813

A ^99mTc-labelled compound with the biological characteristics of flumazenil would be useful for determination of neuronal viability after the onset of a stroke. Therefore, we have derivatized Ro-15-3890 (a flumazenil metabolite bearing a carbox

Full text of DOI:10.1002/jlcr.813

JOURNAL OF LABELLED COMPOUNDS AND RADIOPHARMACEUTICALS
J Label Compd Radiopharm 2004; 47: 199–208.
Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/jlcr.813
Research Article
Synthesis, radio-LC-MS analysis and biological
evaluation of 99mTc-techmazenil
Davy Kieffer, Bernard Cleynhens, Kristin Verbeke, Hubert Vanbilloen,
Tjibbe de Groot, Christelle Terwinghe, Alfons Verbruggen* and Guy
Bormans
Laboratory of Radiopharmaceutical Chemistry, K.U. Leuven, Herestraat 49,
B-3000 Leuven, Belgium
Summary
99mTc-labelled compound with the biological characteristics of flumazenil would
A
be useful for determination of neuronal viability after the onset of a stroke. Therefore,
we have derivatized Ro-15-3890 (a flumazenil metabolite bearing a carboxylic acid
group instead of an ethyl ester) by coupling it with a bisamino bisthiol tetraligand
bearing
a
3-hydroxypropyl side chain (3-hydroxypropyl-BAT) to enable
labelling with technetium-99m. After purification by RP-HPLC, the ligand was
deprotected and labelled in a ‘one pot’ reaction, yielding a 99mTc-BAT-propylester of
Ro-15-3890 (99mTc-techmazenil). Radio-LC-MS analysis of the isolated main
peak showed the molecular ion mass (608.0618) of the expected 99mTc-techmazenil.
The biodistribution of 99mTc-techmazenil was investigated in normal mice and
indicated that the tracer is cleared from plasma mainly by the hepatobiliary
system and shows a very low uptake in brain. In vitro binding studies
on mice brain slices indicated that techmazenil does not bind to benzodiazepine
receptors. Copyright # 2004 John Wiley & Sons, Ltd.
Key Words: flumazenil; radio-LC-MS; technetium-99m; stroke
Introduction
Therapeutic (i.e. thrombolytic) interventions in patients having suffered an
acute ischemic stroke can only be successful if they are initiated when viable
tissue still exists within the area affected by the blood flow disturbance.
Aggressive treatment aimed at reperfusion may however be harmful if morpho-
logical integrity has not been preserved. Therefore, clear distinction between
necrotic and potentially viable tissue, known as penumbra, is essential.1,2
*Correspondence to: A. Verbruggen, Laboratory of Radiopharmaceutical Chemistry, U.Z. Gasthuisberg,
Herestraat 49, B-3000 Leuven, Belgium. E-mail: alfons.verbruggen@uz.kuleuven.ac.be
Copyright # 2004 John Wiley & Sons, Ltd.
Received 29 October 2003
Revised 17 December 2003
Accepted 2 January 2004

200
D. KIEFFER ET AL.
Although X-ray computed tomography (CT), T2-weighed and diffusion
weighed MRI are excellent tools to determine the full extent of an infarction at
later time points (2–3 weeks after onset), they appear to be inadequate in the
early hours after stroke onset.3,4 At present, a complex multitracer study,
requiring quantitative measurements of blood flow and oxygen consumption
by positron emission tomography (PET), is the only reliable method to detect
irreversible damage in an early stage.5,6 Due to the short half-life of the PET
isotopes, this technique is difficult to apply in an acute setting. Therefore, a
neuronal viability marker, labelled with technetium-99m, allowing distinction
between necrotic and ischemic tissue in an acute setting would be helpful.
Because of their abundance in the cortex and their sensitivity to ischemic
damage, visualization of the distribution of g-aminobutyric acid (GABA)
receptors can be used as an indicator of neuronal integrity.
It has been suggested that radiolabelled central benzodiazepine receptor
ligands can detect neuronal damage since they mark intact cortical
neurons.1,2,7–11 Several studies using 11C-flumazenil have shown that reduction
of flumazenil binding is proportional to the actual infarction size as
determined by later MRI-scans.7–9,11 The main advantage of the carbon-11
label is preservation of the chemical structure and the biological
characteristics of flumazenil. Unfortunately the short half-life of 11C
and the need for a cyclotron and a PET-camera for production and
visualization of these derivatives limit the use of 11C-flumazenil in an acute
clinical setting.
On the other hand iodine-123 labelled iomazenil has been proposed and
used for the same purpose.12 123I has several advantages such as its half-life of
13 h and SPECT-camera visualization. However, clinical routine use of this
tracer agent is limited due to its high cost and its non-continuous availability.
A number of fluorescent derivatives of flumazenil have been described in the
literature. These molecules showed high affinity for the benzodiazepine
receptor in spite of the introduction of a bulky fluorescent group.13 This
indicates that conjugation of flumazenil with a technetium-99 m chelate (e.g. a
neutral 99mTc-bisamino bisthiol complex, 99mTc-N2S2 or 99mTc-BAT) could be
feasible without loss of affinity for the benzodiazepine receptor. Bearing in
mind the availability and relatively low cost of technetium-99m and the high
resolution visualization of 99mTc-tracers by widely available SPECT-cameras,
a technetium-99m labelled compound with the biological characteristics of
flumazenil would be an interesting alternative for in vivo visualization of the
benzodiazepine receptors in an acute setting analogous to the application of
99mTc-ECD for the visualization of epileptic foci.14
The aim of this study was the synthesis and the biological evaluation in mice
of 99mTc-techmazenil (Figure 1), a 99mTc-BAT-propyl ester of Ro-15-3890, the
acid derivative of flumazenil.
Copyright # 2004 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2004; 47: 199–208

SYNTHESIS OF 99mTc-TECHMAZENIL
201
O
OCO
C
N
N
S
N
S
OH
HO
N
+
N
CH3
F
O
Ro-15-3890
+ oxalyl chloride
O
OCO
C
N
N
N
S
N
S
O
N
CH3
F
O
2) + SnCl2
+ TcO4
1) + HCl 0.5M
-
T: 100˚C
+ NaKtartrate
T :100˚C
O
O
N
O
N
N
S
Tc
N
S
N
CH3
F
O
99mTc-techmazenil
Figure 1. Synthesis of bis-S-trityl-N-Boc-N0-(3-propyl)-BAT ester of Ro-15-
3890 and 99mTc-techmazenil (proposed structure)
Results and discussion
To enable the preparation of a conjugate of a flumazenil derivative such as Ro-
15-3890 (Figure 1) with a neutral 99mTc-chelate such as a 99mTc-BAT complex,
a hydroxypropyl group was introduced on one of the amines, yielding bis-S-
trityl-N-Boc-N0(3-hydroxypropyl)-BAT.
Several methods were explored for esterification of Ro-15-3890 with bis-S-
trityl-N-Boc-N0-(3-hydroxypropyl)-BAT. In a first method, the protected
hydroxypropyl substituted BAT was reacted with Ro-15-3890 in the
presence of methanesulfonylchloride. However, analysis using thin layer
Copyright # 2004 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2004; 47: 199–208

202
D. KIEFFER ET AL.
chromatography (TLC) and mass spectrometry (MS) did not show the desired
esterified product (results not shown).
In another approach, addition of the protected BAT-propanol after
activation of desethylflumazenil (Ro-15-3890) using DCC and DMAP was
attempted. The desired product was obtained in very low yields (5%).
The best results were obtained by conversion of Ro-15-3890 to the acid
chloride using oxalyl chloride on a disposable solid phase column followed by
reaction with the protected hydroxypropyl-BAT in CH2Cl2 (20% yield).
Labelling of the conjugate with technetium-99 m was achieved after
deprotection of the thiol groups and the BOC-protected amines using a
mixture of trifluoroacetic acid, anisole and methanesulfonic acid. The
deprotected conjugate was then isolated, purified and stored for labelling.
Alternatively, a more convenient method was a one-pot two-step reaction to
deprotect and radiolabel the S,S0-bis-trityl-N-Boc protected conjugate.15 In a
first step the protected conjugate was heated under acidic conditions in a
boiling water bath to obtain the deprotected complex which was subsequently
labelled with technetium-99m in a second heating step after addition of
solutions of phosphate buffer pH 7, NaKtartrate, SnCl2 and 99mTcO4ꢀ. Using
this one-pot method, the 99mTc-techmazenil complex was obtained in a
radiochemical yield of 80%, which is comparable to the yields of the more
traditional two-pot labelling procedure.
Confirmation of the identity of 99mTc-techmazenil was obtained using liquid
chromatography in combination with mass spectrometry using a radiometric
detector (radio-LC-MS). A preparation reconstituted using an eluate with a
relatively high technetium content (Monday morning eluate, containing
0.07 ng Tc per GBq) was purified by reversed phase high performance liquid
chromatography (RP-HPLC). Radio-LC-MS analysis of the isolated main
peak showed the molecular ion mass of the expected 99mTc-techmazenil
complex (608.0618 Da, Figure 2) on the mass spectrometer channel at the time
of elution of a peak in the radiometric channel (retention time (Rt): 14 min)
with a relative error of 10 ppm.
Before proceeding to animal testing of 99mTc-techmazenil, a number of
biological parameters were determined. In the literature, a parabolic relation-
ship is suggested for the passive diffusion of neutral molecules into the central
nervous system (CNS) with an optimal passage over the blood–brain barrier
found for compounds with a logarithm of the partition coefficient in 1-
octanol/water (log Poct) of about 2. The log Poct of 99mTc-techmazenil is 1.49,
compatible with reasonable brain uptake. In addition, the logarithm of the
partition coefficient in cyclohexane/water (log Pcyh) was found to be ꢀ2.51.
From these values, D log P, defined as log P (1-octanol/water) ꢀ log P
(cyclohexane/water) and related to the overall hydrogen-bonding ability,
was calculated and found to be 4.0, which predicts a brain/blood ratio in rat
Copyright # 2004 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2004; 47: 199–208

SYNTHESIS OF 99mTc-TECHMAZENIL
203
Figure 2. Radio-LC-MS analysis of Tc-techmazenil (A) Single mass chromato-
gram (607.846–608.250): peak with Rt: 14.010. (B) Radiometric channel: peak
with identical Rt as peak in A. (C) Experimental mass spectrum (relative error:
7.4 ppm)
brain of 0.065 according to the formula: log (Cbrain/Cblood)=ꢀ0.604*D
log P+1.23.16
The biological affinity for the benzodiazepine receptor was assessed by
in vitro autoradiography on mice brain sections after incubation with the
tracer agent. Pre-incubation was the same in all the experiments, but two
different incubation times (30 and 40 min) and three different wash times (15,
30 and 60 s) were assessed. However, for all tested conditions no binding of
99mTc-techmazenil to a brain structure was found. This was unexpected in view
of the affinity for benzodiazepine receptors observed for conjugates of Ro-15-
3890, which were esterified via the same carboxyl group with bulky fluorescent
groups.13
Biodistribution studies in normal mice showed no uptake of 99mTc-
techmazenil in the brain at 10 and 60 min p.i. (respectively 0.05 and 0.03%
of injected dose (ID), Table 1). Furthermore, extensive hepatobiliary excretion
(68.1% of ID in liver and intestines at 60 min p.i.) was observed. The absence
of significant radioactivity in the stomach, even after 60 min, indicates that no
pertechnetate is reformed in vivo.
Copyright # 2004 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2004; 47: 199–208

204
D. KIEFFER ET AL.
Table 1. Biodistribution of 99mTc-TECHMAZENIL in mice
% of ID (n=4)
% of ID/g (n=4)
10 min p.i.
mean ꢂ sd
60 min p.i.
mean ꢂ sd
10 min p.i.
mean ꢂ sd
60 min p.i.
mean ꢂ sd
Urine
0.1 ꢂ 0.1
13.5 ꢂ 1.3
45.1 ꢂ 3.9
0.7 ꢂ 0.1
0.4 ꢂ 0.1
14.2 ꢂ 4.0
0.6 ꢂ 0.4
0.0 ꢂ 0.0
0.0 ꢂ 0.0
4.9 ꢂ 0.7
5.6 ꢂ 0.8
1.3 ꢂ 0.5
18.9 ꢂ 3.2
0.1 ꢂ 0.0
0.0 ꢂ 0.0
68.1 ꢂ 3.8
0.9 ꢂ 0.2
0.0 ꢂ 0.0
0.0 ꢂ 0.0
0.9 ꢂ 0.7
}
}
Kidneys
Liver
Lungs
20.4 ꢂ 2.2
17.8 ꢂ 2.8
2.2 ꢂ 0.6
2.3 ꢂ 0.3
3.4 ꢂ 0.6
0.9 ꢂ 0.8
0.1 ꢂ 0.0
0.2 ꢂ 0.0
1.7 ꢂ 0.2
1.9 ꢂ 0.8
8.1 ꢂ 0.9
0.4 ꢂ 0.0
0.2 ꢂ 0.1
18.5 ꢂ 3.7
1.5 ꢂ 0.4
0.1 ꢂ 0.0
0.1 ꢂ 0.1
0.3 ꢂ 0.1
Heart
Intestines
Stomach
Cerebrum
Cerebellum
Blood
The low brain uptake of 99mTc-techmazenil is surprising, in view of its
molecular weight of 608 the fact that a molecular weight limit and between 400
and 700 is accepted for significant entry into the brain.16
Experimental
Materials
Ro-15-3890 (desethylflumazenil) was a gift from Hoffman-La Roche (Basel,
Switzerland). All other reagents were ACS or HPLC grade and purchased
commercially. Na99mTcO4 was obtained from a 99Mo/99mTc generator
(Ultratechnekow FM, Mallinckrodt, Petten, The Netherlands).
1H-NMR spectra were recorded using a Varian 200 MHz spectrometer
(Varian, Palo Alto, CA, USA). Chemical shifts are reported in ppm relative to
the internal standard tetramethylsilane (TMS, d ¼ 0).
Reversed phase HPLC was carried out using a system consisting of a Merck
Hitachi L-7100 separation module connected to a RP C18 column (XterraTM
RP18 5 mm 250 mm ꢁ 4.6 mm, Waters Milford, USA). The column eluate
was analysed using a radiometric detector (3-in NaI(Tl) detector connected
to a radiation analyser module, Canberra Packard, Meriden, Conneticut,
USA).
The radio-LC-MS system consisted of a Waters 2690 separation module
connected to a RP C18 column (XTerraTM MS C18 3.5 m 2.1 mm ꢁ 50 mm).
The column eluate was monitored for radioactivity using a radiometric
detector (3-in NaI(Tl) detector connected to a radiation analyser module, The
Nucleus, Oak Ridge, USA). Finally the column eluate was directed to a time-
of-flight mass spectrometer (Micromass LCT, Waters) equipped with an
orthogonal ESI probe.
Copyright # 2004 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2004; 47: 199–208

SYNTHESIS OF 99mTc-TECHMAZENIL
205
For the in vitro autoradiography studies a phosphor imaging system was
used (CycloneTM, Canberra Packard).
S,S0-bis-trityl-N-Boc-N0-(3-hydroxypropyl)-1,2-ethanedicysteamine (S,S0-bis-
trityl-N-Boc-N0-(3-hydroxypropyl)BAT)
S,S0-bis-trityl-N-Boc-BAT, (3.53 g; 4.6 mmol) was synthesized according to the
described method17 and was dissolved in a mixture of 20 ml acetonitrile and
1.9 ml diisopropylethylamine. 3-Bromo-1-propanol (1.4 g; 10 mmol) was
added and the mixture was refluxed overnight. After solvent and organic
base evaporation, the residue was dissolved in 20 ml CH2Cl2. After several
consecutive washing steps using respectively water, 0.1 M citric acid, 1 M
sodium bicarbonate and saturated saline solution, the organic layer was dried
using magnesium sulphate and evaporated.
Purification was performed with column chromatography using a solvent
system of hexane–ether–triethanolamine (50–45–5) to yield 35% of a
colourless, viscous oil.
1H-NMR (CDCl3) d 1.38 (s, 9 H, C(CH3)3), 1.48 (m, 2 H, CH2CH2CH2OH),
2.2–2.4 (m, 10H, CH2NHCH2+2 ꢁ SCH2CH2N, NCH2CH2CH2OH), 2.93
(m, 4 H, CH2NBocCH2), 3.61 (m, 2H, CH2OH), 7.2–7.4 (m, 30 H, ArH).
MS ESI+: m/z (M+H)+ found 823 (calcd, C52H58N2O3S2, 823).
Bis-S-trityl-N-Boc-N0-(3-propyl)-BAT ester of Ro-15-3890
Ro-15-3890 (27.5 mg; 0.1 mmol), was dissolved in 5 ml 0.5 M phosphate buffer
pH 8. This solution was applied on a disposable PRP-column (Chromafix HR-
.
P 700 mg, Macherey-Nagel, Duren, Germany) after which it was rinsed with
water (10 ml). The column was dried for 1 h with a flow of N2 gas (1 bar). A
vial containing oxalyl chloride (100 ml, 1 mmol) was inserted between the N2
source and the PRP column. Again, a N2 flow of 1 bar was applied for 1 h after
which the column was eluted with 5 ml of CH2Cl2. The solvent, along with the
excess of oxalyl chloride, was evaporated. S,S0-bis-trityl-N-Boc-N0-(3-hydro-
xypropyl)BAT (82.3 mg; 0.1 mmol), dissolved in a mixture of 5 ml of CH2Cl2
and 15 ml of pyridine was added to the residue. The mixture was stirred for 2
days after which it was purified using RP-HPLC, using as the mobile phase a
mixture of acetonitrile, tetrahydrofuran and 0.1 M ammonium acetate
(76.5:8.5:15), eluted in an isocratic way at a flow of 1 ml/min. The appropriate
fraction was evaporated under reduced pressure and the residue was led over a
Sep-Pak C18 high capacity cartridge (Alltech, Deerfield, Illinois, USA) that
was then rinsed with 50 ml water. Afterwards, the product was eluted from this
column using 50 ml pure MeOH (3 times) and a slightly yellow oil was
obtained after evaporation (yield: 15%).
MS ESI+: m/z (M+H)+ found 1080 (calcd, C65H66FN5O5S2, 1080).
Copyright # 2004 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2004; 47: 199–208

206
D. KIEFFER ET AL.
99mTc-BAT-propyl ester of Ro-15-3890 (99mTc-techmazenil, Figure 1)
1 Mg of protected 3-propyl-BAT ester of Ro-15-3890 was dissolved in 1 ml
ethanol and this solution was added to a labelling vial containing 0.5 ml HCl
0.1 M. The solution was heated for 10 min in a boiling water bath. The mixture
was neutralized using 1 ml of phosphate buffer 0.5 M pH 7 before addition of
NaKtartrate (1 mg), SnCl2 (100 mg in 25 ml HCl 0.1 M) and generator eluate
(370 MBq 99mTc in the form of pertechnetate in 1 ml). The vial was heated
again for 10 min in a boiling water bath.
The labelling reaction mixture was analysed and purified using RP-HPLC
on an XterraTM RP18 column (5 mm, 4.6 mm ꢁ 250 mm) which was eluted at a
flow rate of 1 ml/min using a mixture of CH3CN and 0.1 M NH4OAc (65:35).
Radio-LC-MS analysis
To enable accurate mass calculations, the column eluate was mixed with a lock
mass solution (flumazenil 0.01 mg/ml) infused at a flow rate of 5 ml/min. The
mass difference between the theoretical mass and the measured accurate mass
of 99mTc-techmazenil was divided by the theoretical mass, yielding the relative
error expressed as ppm. The HPLC column was eluted at a flow rate of 300 ml/
min with linear gradient mixtures of CH3CN and 0.1 M NH4OAc (t=0 min
0% CH3CN; t=20 min 90% CH3CN V/V).
Determination of the partition coefficients: log Poct, log Pcyh and D log P
To a test tube containing 2 ml phosphate buffer (0.025 M, pH 7.4) and 2 ml 1-
octanol, 25 ml of HPLC purified 99mTc-techmazenil solution was added. The
test tube was shaken for 5 min, followed by centrifugation for 10 min at
6000 rpm (Centrifuge 4226, ALC, Milano, Italy). Finally 50 mg of the octanol
phase and 500 mg of the water phase was withdrawn and the radioactivity of
both samples was measured using a gamma counter equipped with a 3-in.
NaI(Tl) scintillation detector (1480 Wizard 300, Wallac, Finland). The results
were used to calculate the log Poct. This experiment was performed in
triplicate.
A similar procedure was followed for the determination of log Pcyh using
cyclohexane instead of 1-octanol. The D log P value was calculated by
subtraction of log Pcyh from log Poct
.
In vitro autoradiography
For determination of the binding affinity of 99mTc-techmazenil for the
benzodiazepine receptor, a male NMRI mouse was sacrificed by decapitation;
the brain was removed and quick-frozen in isopentane cooled over dry ice to
ꢀ408C. Twenty-micrometer sections were cut at ꢀ208C and thaw-mounted
Copyright # 2004 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2004; 47: 199–208

SYNTHESIS OF 99mTc-TECHMAZENIL
207
onto gelatine-coated slides, desiccated in a stream of air and frozen at ꢀ808C
until processing.
Tissue sections were pre-incubated for 10 min at room temperature in Tris–
HCl buffer (50 mM, pH 7.4) to remove any endogenous ligand. Sections were
then incubated at room temperature with 300 ml of 99mTc-techmazenil solution
(3.7 kBq/ml) for 30 or 40 min. Three different intervals of rising in distilled
water were assessed (15, 30 and 60 s) after which the sections were exposed to a
high performance storage phosphor screen (Canberra, Packard). After 24 h,
the screen was analysed using a phosphor imaging system.
Biodistribution in mice
RP-HPLC purified 99mTc-techmazenil (15 kBq/100 ml) was injected in 8 normal
male NMRI mice (body mass 30–40 g) via a tail vein. Four of them were
sacrificed at 10 min p.i. and the other four at 60 min p.i. and the organs and
body parts were dissected and weighed. The activity in the dissected organs
and body parts was measured using a gamma counter. Results are expressed as
percentage of injected dose (% of ID) and percentage of injected dose per
gram (% of ID/g). For calculation of total blood radioactivity, blood mass
was assumed to be 7% of the body mass.18
Conclusion
The formation of the ester of S,S0-bis-trityl-N-Boc-N0-(3-hydroxypropyl)BAT
with desethylflumazenil (Ro-15-3890) was accomplished by means of solid
phase activation of Ro-15-3890 using oxalyl chloride. The traditional two-pot
two-step labelling method was simplified to a one-pot two-step labelling
method with similar high yields of 99mTc-techmazenil. The identity of the
expected Tc-complex was confirmed using radio-LC-MS.
Although physicochemical parameters of this Tc-complex (log Poct, log Pcyh
and D log P) were found to be compatible with brain uptake, in vivo
biodistribution studies in mice indicated absence of brain uptake. This is
probably due to the rather large size of techmazenil. Unexpectedly, in vitro
binding studies with mice brain slices showed no affinity of 99mTc-techmazenil
for the benzodiazepine receptors.
In view of these negative results, 99mTc-techmazenil is not useful for the
in vivo visualization of cerebral benzodiazepine receptors.
Acknowledgements
The authors wish to thank Mr P. Weber (Hoffman-La Roche) for the kind
donation of Ro-15-3890.
References
1. Heiss WD, Graf R, Wienhard K. Cerebrovasc Dis 2001; 11: 73–81.
Copyright # 2004 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2004; 47: 199–208

208
D. KIEFFER ET AL.
.
2. Heiss WD, Graf R, Lottgen J, Ohta K, Fujita T, Wagner R, Grond M, Wienhard
K. J Cerebr Blood F Met 1997; 17: 388–400.
3. Kummer Rv, Meyding-Lamade U, Forsting M, Rosin L, Rieke K, Hacke W,
Sartor K. Am J Neuroradiol 1994; 15: 9–15.
4. Warach S, Gaa J, Siewert B, Wielopolski P, Edelman RR. Ann Neurol 1995; 37:
231–241.
5. Baron JC, Rougemont D, Soussaline F, Bustany P, Crouzel C, Bousser MG,
Comar D. J Cerebr Blood F Met 1984; 4: 140–149.
6. Baron JC. Cerebrovasc Dis 1999; 9: 193–201.
7. Heiss WD, Grond M, Thiel A, Ghaemi M, Sobesky J, Rudolf J, Bauer B,
Wienhard K. Stroke 1998; 29: 454–461.
8. Heiss WD, Kracht L, Grond M, Rudolf J, Bauer B, Wienhard K, Pawlik G.
Stroke 2000; 31: 366–369.
9. Heiss WD, Kracht LW, Thiel A, Grond M, Pawlik G. Brain 2001; 124: 20–29.
.
10. Heiss WD, Graf R, Fujita T, Ohta K, Bauer B, Lottgen J, Wienhard K. Stroke
1997; 28: 2045–2051.
.
11. Thiel A, Lottgen J, Grond M, Pietrzyk, Heiss WD. J Comput Assist Tomo 2001;
25: 446–451.
12. Nakagawara J, Sperling B, Lassen NA. Stroke 1997; 28: 124–132.
13. Janssen MJ, Hulst R, Kellogg RM, Hendriks MM, Ensing K, De Zeeuw A.
Pharmazie 2000; 55: 42–48.
14. Oliveira AJ, da Costa JC, Hilario LN, Anselmi OE, Palmini A. Epilepsia 1999; 40:
693–702.
15. Cleynhens BJ, Bormans GM, Vanbilloen HP, Vanderghinste DV, Kieffer DM, de
Groot TJ, Verbruggen AM. Tetrahedron Lett 2003; 44: 2597–2600.
16. Young RC, Mitchell RC, Brown TH, Ganellin CR, Griffiths R, Jones M, Rana
KK, Saunders D, Smith IR, Sore NE, Wilks TJ. J Med Chem 1998; 31: 656–671.
17. Bormans G, Cleynhens B, de Groot TJ, Mortelmans L, Moretti JL, Verbruggen A.
J Label Compd Radiopharm 2003; 46: 575–585.
18. Fritzberg AR, Whitney W, Kuni C, Klingensmith W. Int J Nucl Med Biol 1982; 9:
79–82.
Copyright # 2004 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2004; 47: 199–208