- Email: [email protected]
Synthesis and evaluation of an 18F-labeled dopa prodrug as a PET tracer for studying brain dopamine metabolism
Nuclear Medicine & Biology, Vol. 23, pp. 295-301, Copyright 0 1996 Elsevier Science Inc.
ISSN 0969-8051/96/$15.00 + 0.00 SSDI 0969-8051(95)02083-7
1996
ELSEVIER
Synthesis and Evaluation of an “F-Labeled Dopa Prodrug as a PET Tracer for Studying Brain Dopamine Metabolism Kiichi Ishiwata , ’ * Masaki Shinoda, ’ ,2 Shin-Ichi Ishii, ’ Tadushi Nozaki’ and Michio Senda’ ‘POSITRON MEDICAL CENTER, TOKYO METROPOLITAN INSTlTUTE OF GERONTOLOGY, l-l TOKYO 173, JAPAN, AND *FACULTY OF HYGIENIC SCIENCE, KITASATO UNIVERSITY, SAGAMIHARA 228, JAPAN
NAKA-CHO, ITABASHI, l-15-1 KITASATO,
ABSTRACT. In the quantitative studies of presynaptic dopamine metabolism by PET with 6-[“FlfluoroL-dopa (6-[‘8F]FDOPA), metabolic analysis in the plasma is required to determine the precise input function because of susceptibility of the compound to peripheral metabolism. In this study, we prepared 6418F]fluoro-O-pivaloyl-L.-dopa (6-[ ‘sF]FPDOPA) as a prodrug of 6-[“FIFDOPA, and evaluated its potential as a PET tracer in mice. If the 6-[ “F]FPDOPA is stable peripherally and is hydrolyzed to 64 18F]FDOPA in the brain tissues, disadvantage of the 6-[ “FIFDOPA will b e overcome. Compared with the 6-[ 18F]FDOPA, the initial brain uptake of the 6-[18F]FPDOPA was lower; however, the uptake in the latter become comparable, and the uptake ratios of striatum to other reference regions were larger. Medication of mice with inhibitors of aromatic amino acid decarboxylase and catechol-O-methyl transferase greatly enhanced the striatal uptake of the two compounds. The reduced brain uptake of the compounds by L-phenylalanineloading suggested transport through the blood-brain barrier by the neutral amino acid transporter. HPLC analysis showed the presence of 6J18F]FPDOPA, 6-[r8F]FDOPA and 6-[“Flfluorodopamine in the striatum; however, 6-[ “Flfluoro-3-0-methyl+dopa was a predominate metabolite in the brain and plasma as in the case of [t’F]FDOPA. Results suggested that 6-[18F]FPDOPA had characteristics as a prodrug of 6-[‘8F]FDOPA; however, the compound was also labile to metabolic alteration in viva. NUCL MED BIOL 23;3:295-301, 1996. KEY WORDS. Dopamine
metabolism,
Prodrug,
6-[‘“Flfluoro-O-pivaloyl-L-dopa,
Positron
emission
tomogra-
phy
INTRODUCTION
sibility, this study deals with the use of a prodrug of 6-[18F]FDOPA as a PET tracer. The compound L-3-(3-hydroxy-4-pivaloyloxyphenyl)alanine (4-Opivaloyl-L-dopa) has been developed as a potential prodrug of L-dopa to improve therapeutic effect (11, 14, 25). The compound showed longer mean residence time and larger bioavailability of L-dopa concentrations than did L-dopa itself in rats. The pivaloyl ester of the compound is hydrolyzed in ho; however, the bulky pivaloyl group may interfere with the COMT reaction because the pivaloyl group migrates to equilibrium of the 4-O-isomer (53-59%) and 3-O-isomer in an aqueous solution (13). Electrophilic fluorination of 4-0-pivaloyl-L-dopa with acetyl [‘*F]hypofluorite gave predominantly 6-[‘sF]fluoro-O-pivaloylL-dopa (6-[18F]FPDOPA) (16). We expected that the 6-[“flFPDOPA was taken up by the brain and then hydrolyzed to 6-[‘“FIFDOPA. If peripheral hydrolysis of the ester-bond of the compound is substantially suppressed during a couple of hours, metabolite analysis is not required in PET studies. In this work, we characterized the properties of the compound in mice, compared with 6-[18F]FDOPA, and evaluated the potential as a PET tracer.
Currently, 6-[‘sF]fluoro-L-3,4-dihydroxyphenylalanine (6-[i*F]fluoro-L-dopa, 6-[‘8F]FDOPA) is used for studying the presynaptic dopaminergic function by positron emission tomography (PET). The 6-[18F]FDOPA is labile to metabolic alteration in rodents as well as primates (3-5,9, 20). Some metabolites such as 6-[18F]fluorodopamine ([i8F]FDA) are not taken up by the brain. However, 6-[‘*F]fluoro-L-3methyoxy-4-hydroxyphenylalanine (6-[18F]fluoro-3-0-methyl-L-dopa, [i8F]FMDOPA), which is produced by catechol-O-methyl transferase (COMT), crosses the blood-brain barrier into the brain and is present homogeneously in the brain tissues (7, 8). Therefore, metabolite analysis in the plasma is required to determine the precise input function in the quantitative assessment of dopamine metabolism. To suppress the metabolism of 6-[i8F]FDOPA, human subjects are usually given selective inhibitors of peripheral aromatic amino acid decarboxylase (AADC) (12, 24) prior to PET studies. Combined use of the COMT and AADC inhibitors to diminish [“FIFMDOPA in the plasma shows great advantage to the PET studies (18, 23). Another approach to overcome the disadvantage of the 6-[18F]FDOPA is the choice of less susceptible compounds to the metabolism. Meta-[‘sF]fluoro-L-tyrosine derivatives are candidates (6, 21, 22). As a third pos-
MATERIALS
*Author for correspondence. Accepted 31 October 1995.
The 4-0-pivaloyl-L-dopa was supplied by Banyu Pharmaceutical Co. (Tokyo), and L-methyl-N-acetyl-[P-3-methoxy-4-acetoxyphenyl)]alaninate was from Dr. K. Hatano, Iwate Medical School (Morioka, Japan). Both 3-O-methyl-D,L-dopa and 3-methoxytyramine were purchased
AND METHODS
296
K. Ishiwata
from Sigma Chemical Co. (St. Louis, MO); dopamine, D,L-dopa, homovanilic acid, and L-phenylalanine were from Wako Pure Chemical Industries (Tokyo); and 3,4-dihydroxyphenylacetic acid was from Aldrich Chemical Co. (Milwaukee, WI). Carbidopa and tolcapone (RO 40-7592) were supplied from Sankyo Co. (Tokyo) and Hoffman-La Roche (Basel, Switzerland), respectively. Other chemicals used were analytical grade and purchased from Wako Pure Chemical Industries (Tokyo).
Acetyl [“Flhypofluorite, which was produced using a previously reported method (15, 16), was bubbled into 6 mL of acetic acid containing 15-20 mg of 4-O-pivaloyl-L-dopa at room temperature (Fig. 1). After the solution was evaporated to dryness, the residue was dissolved in 2 mL of H,O. The solution was applied to separation by high-performance liquid chromatography (HPLC). A preparative reverse-phase column used was YMC-Pack ODS S-5 (20 mm i.d. x 250 mm length, YMC Co. Ltd., Kyoto). The mobile phase was a mixture of CH,OH and 50 mM NaH,PO+ (35:65, v/v), and passed at a flow rate of 10 mL/min. The elution profile was monitored with a UV detector at 271 nm and an NaI radioactivity detector. Fractions of 6-[“F]FPDOPA and 2-[‘*F]FPDOPA were collected and evaporated to dryness. The residue was dissolved in physiological saline and passed through a 0.22~p,rn membrane filter. Both 6,[“F]FDOPA and 2-[“F]FDOPA were prepared as previously reported (15, 16). The specific activity of 6-[“F]FDOPA was 23.9 - 30.0 GBq/mmol. The [‘*F]FMDOPA was synthesized using L-methyl-N-acetyl-[P-3-methoxy-4-acetoxyphenyl)]alaninate as a starting material according to the method of Adam and Jivan (1). The [“F]FDA was prepared enzymatically using AADC in rat striatum homogenate as previously described (17). Radiochemical purity of compounds was analyzed by HPLC. For 6-[“FIFPDOPA and 2-[“FIFPDOPA, a Crestpak C18S (4.6 mm i.d. x 150 mm length, Japan Spectroscopic Co., Ltd., Tokyo) was eluted with a mixture of CH,OH and 50 mM NaH,P04 (25:75, v/v) at a flow rate of 2 mL/min at 50°C. For 6-[“F]FDOPA, 2-[“F]FDOPA, and [‘8F]FMDOPA, the mobile phase was a mixture of CH,OH and 0.1% acetic acid containing 1 mM EDTA disodium and 1 mM sodium octylsulfate (1:9, w/w) at a flow rate of 2 mL/min at room temperature.
Biodistribution
Study
Either 6-[“FIFPDOPA or 6-[“F]FDOPA was injected intravenously into four groups of male ddY mice (8 to 9 weeks old), and tissue distribution of the radioactivity was measured. The first group was the control; the second group was given carbidopa (5 mg/kg) intraperitoneally 30 min before the tracer injection; the third group was given cardibopa (5 mg/kg) and tolcapone (10 mg/kg) intraperitoneally 30 min and 60-90 min, respectively, before the tracer injection. The last group, which was pretreated with carbidopa, was received with a mix-
(CH3)3CCOO-
ture of tracer and L-phenylalanine (100 mg/kg). These mice received 1.0 MBq of tracers, and then were killed by cervical dislocation at 5, 15,30, 60, and 120 min postinjection. The blood was removed by heart puncture using a heparinized syringe. The tissues were dissected and weighed. The radioactivity of the tissues was counted using an autogamma counter. The tissue uptake of radioactivity was expressed as the percent of injected dose per gram of tissue.
Metabolic
Radiopharmaceuticals
CH2CH-C02H
et al.
Study
Mice pretreated with carbidopa received 10 MBq of tracers, and were killed by cervical dislocation 30 min postinjection. The blood removed by heart puncture was centrifuged to obtain plasma. The plasma, striaturn, cerebral cortex, and cerebellum were treated to prepare the samples for HPLC by a previously described method (17). Metabolites in the samples were analyzed by HPLC using a Nova-Pak Cl8 column (8 mm i.d. x 100 mm length, Waters) equipped in an RCM 8 x 10 module. The column was eluted with an ion pair solution with a gradient modifier at a constant flow rate of 2 mL/min at room temperature. The initial solution was a mixture of solution A (1% acetic acid containing 1 mM EDTA and sodium octylsulfate) and solution B (40% MeOH containing 1% acetic acid, 1 mM EDTA, and sodium octylsulfate (3/l, v/v) for the first 5 min, after which a solvent with a linearly increasing gradient of solution B (25-100%) was eluted for 40 min. During the last 15 min solution B was delivered, and then the column was reequilibrated with the initial conditions. The elution profile was detected with a UV detector and a radioactivity monitor (FLO-ONE/Beta A200, Packard), and the radioactivity in each 1.0 mL collected using a fraction collector was measured with an auto-gamma counter. Radioactivity was corrected for the half-life of 18F. A portion of the sample applied was also measured to calculate the total applied radioactivity, and the percentage of radioactivity in each peak of the total applied radioactivity was calculated. Recovery of the radioactivity was essentially quantitative.
RESULTS Radiochemistry Fluorination of 4-0-pivaloyl-L-dopa with acetyl [“flhypofluorite (Fig. 1) followed by HPLC-separation (Fig. 2) gave two major labeled components and other minor components. HCl-hydrolysis of peaks 1 and 3 gave 6-[“FIFDOPA and 2-[“F]FDOPA, respectively, which were identified by comparing their retention times with those of the authentic samples. Peaks 1 and 3 showed maximal UV absorption at 276 nm and 269 nm, corresponding to hmax values of 6-[“FIFDOPA (283 nm) and 2-[“F]FDOPA (271 nm), respectively (3). Therefore, peaks 1 and 3 were identified as 6-[“FIFPDOPA and 2-[“FIFPDOPA, respectively. Peak 4 was not identified, but probably is 2,6-di-[‘*F]fluoro-derivative (described elsewhere). The maximal radiochemical yields of the 6-[“FJFPDOPA and 2-[“FIFPDOPA were 26% and 15%, respectively,
1) AcO’*F, AcOH 2) HPLC separation
t
(CH3)3CCOO-
iJH2
/
\
CH2CH-C02H kH2
Q 18F
FIG. 1. Synthesis of ~-3*(6-[ ‘*F]fluoro-3*hydroxy-4-pivaloyloxyphenyl)alanine
([ ‘sF]FPDOPA).
“F-Labeled
297
Dopa Prodrug
..a ~
~ . . . . . . . . . . . . . . . . . “”
. ...+* p
M -<
..:! :
: i
%
.
; ::: iF
: :
i’ .
I :
j’ 1: i’
. . . . I......
..:
:,......
:..&..
:;
::
: i
I
1
I 0
L
20
40
60
L 80
100
Retention time (min) FIG. 2. A preparative HPLC chromatogram of the reaction mixture of fluorinated ~-34 3.hydroxy-4.pivaloyloxyphenyl)alanine with acetyl ] ‘sF]hypofluorite in acetic acid using a reverse-phase column. HPLC conditions are described in text. Peak 2 corresponds to 4-0.pivaloyl+dopa.
based on acetyl [“Flhypofluorite, with radiochemical and the preparation time of 100 min.
purity
of >99%
Biodistribution The tissue distribution of 6-[“FIFDOPA and 6-[‘*F]FPDOPA are summarized in Tables 1 and 2. Two tracers showed similar tissue distribution. The highest uptake was observed in the pancreas followed by the
TABLE Mice
1. Tissue
Distribution
of Radioactivity
kidney and small intestine. The pancreas uptake of 6-[‘*F]FDOPA slightly increased for the first 30 min, whereas that of 6-[“FIFPDOPA decreased with time. The brain uptake of two compounds was the lowest among the organs studied, and this decreased with time. Figure 3 shows regional brain distribution of the two tracers and the effect of pretreatment with enzyme inhibitors on the distribution. The uptake of 6-[‘*F]FDOPA in the striatum of the control mice was slightly higher than that in the cerebral cortex and cerebellum. The radioactivity levels in three regions decreased with time. On the other hand, pretreatment with carbidopa enhanced the striatal uptake in which the level of radioactivity increased for the first 30 min, then decreased. Pretreatment with both carbidopa and tolcapone resulted in the largest difference between the striatal uptake and the uptake by other regions. The striatal uptake increased for the first 60 min and remained nearly constant. In control mice, the uptake of 6-[‘*F]FPDOPA by the striatum, cortex, and cerebellum was significantly lower at 5 min (p < 0.01) than that of 6-[‘sF]FDOPA. The striatal uptake increased slightly for the first 30 min, and then decreased with time. After 30 min, radioactivity levels of the two tracers in the striatum were comparable, but the level for 6-[‘*F]FPDOPA in the cortex and cerebellum was significantly lower than that for 6-[‘*F]FDOPA (p < 0.01). Consequently, the difference between the striatal uptake and the uptake by reference regions was larger in the 6-[‘*F]FPDOPA than in 6-[“F]FDOPA. Carbidopa also enhanced the striatal uptake of the 6-[“FIFPDOPA; however, the effect was not as large as compared with the case of 6-[t*F]FDOPA. Pretreatment with both carbidopa and tolcapone resulted in the increasing striatal uptake during the entire period investigated. In Figure 4, the uptake ratios of striatum to other reference regions are represented. In the control group, uptake ratios of striatum to other tissues for 6-[‘*F]FPDOPA were greater than those for 6-[‘*F]FDOPA. In the mice treated with carbidopa alone, the ratios for the 6-[‘*F]FPDOPA increased with time, whereas the ratios for the 6-[‘*F]FDOPA remained constant after 60 min. The two enzyme inhibitors made the ratios larger for both compounds. Co-injection of L-phenylalanine significantly reduced the initial uptake (5 min postinjection) of both 6-[‘*F]FDOPA and 6-[‘*F]FPDOPA to 52 - 76% of the control in three brain regions (Table 3). Metabolism Metabolites in the plasma and brain were analyzed by HPLC. chromatogram of the striatal metabolites of 6-[‘*F]FPDOPA
After
% Injected 5min Blood Brain Heart Lung Liver Spleen Pancreas Small inte ,stine Kidney Muscle Bone a Mean + SD (n = 4).
1.74 It 0.19
0.42 kO.08 1.52 f0.16 1.55 kO.16 3.49 kO.48 1.68 f 0.24 11.19kO.63
5.30+ 1.12 9.27 3~2.98 1.10 f 0.15 1.31 kO.27
15 min
0.90+0.06 0.30 f0.04 0.73 kO.04 0.89 f0.06 2.10 f 0.68 0.83 f 0.32 12.16 f 1.78
4.38 f0.59 4.71 f 1.16 0.85 kO.05 0.81 kO.12
i.v. Injection
of 6-t “F]FDOPA
into
dose/g tissue”
30 min
60 min
0.64 f 0.12 0.25 fO.O1 0.51 f 0.06 0.70* 0.30 1.13 +0.39 0.50 f0.05 12.65 kO.16 2.75 k0.47 3.05 f0.57 0.69 f0.04 0.69 f.O.07
0.38 f 0.06 0.15 fO.O1 0.34kO.02 0.49 zk 0.13 0.61kO.32 0.31 + 0.05 7.81 3~ 1.18 1.83 kO.61 2.68 fO.12 0.44 f0.05 0.51 f 0.04
120 min
0.22 f0.03 O.lOfO.O1 0.21 * 0.01
0.46 f 0.37 0.25 +0.03 0.22 f0.04 4.28kO.77 1.12 f 0.38 2.41 kO.77 0.25 fO.O1 0.42 f0.07
A typical is shown
298
K. Ishiwata
TABLE Mice
2. Tissue
Distribution
of Radioactivity
After
% Injected
2.30f 0.13 0.23 + 0.02 1.11 f0.14 1.70 + 0.16
7.63 kO.75 2.08 f 0.12 14.83 + 2.36 4.53 310.69 19.30 f 1.88 0.58 kO.03 1.07 f 0.10
of 6-[ 18F]FPDOPA
into
dose/g tissue*
15 min
30 min
60 min
120 min
1.04 f0.27 0.22 f 0.07 0.59 f 0.08 0.92 k 0.21 4.40 k0.76 0.86 +0.09 a.57 f 1.40 2.48 k0.87 4.89 f 1.75 0.04 f0.03 0.67 k 0.28
0.66 f0.05 0.18 rkO.02 0.39 f 0.05 0.55 * 0.07 0.94 IL 0.05 0.40 IL 0.03 7.26 f 0.79 1.69 f 0.38 4.17 IL 1.17 0.36 f 0.04 0.49 f 0.10
0.26 kO.05 0.09 IL 0.01 0.16 f0.03 0.26f0.04 0.50,O.ll 0.16 f 0.04 4.86kO.76
0.10 f 0.01
5min Blood Brain Heart Lung Liver Spleen Pancreas Small intestine Kidney Muscle Bone
i.v. Injection
et al.
1.45 + 0.43 2.41 f 0.37
0.19 rto.05 0.31 f 0.03
0.04* 0.00 O.lOfO.O1 0.10 f 0.04 0.16 f 0.04 0.09 f 0.01
2.39 k0.45 0.62 f 0.19 2.36f0.77 0.12 f 0.01 0.25 f 0.02
* Mean k SD (n = 4).
in Fig. 5. Radioactive peaks 2, 4, 5, and 7 were 6-[18F]FDOPA, [‘sF]FMDOPA, [“FIFDA, and 6-[‘*F]FPDOPA by comparing retention times of the authentic samples. Peaks 3 and 6 were tentatively identified to be [‘*F]FHVA and [‘8F]Fm&&omparing their re-
tention times with those of L-dopa and its related metabolites (Fig. 5). A similar analytical profile (except for 6,[“F]FPDOPA) was observed in the metabolites of 6-[“FIFDOPA. Percentages of metabolites in the plasma and brain 30 min after the injection are summarized in Table 4.
A control
carbidopa and tolcapone
carbidopa
1
Time after Injection (min)
control * -
carbidopa and tolcapone
carbidopa
slriatum corbex cerebellum
120
T
0
60
120
Time after injection (min)
0
60
120
FIG. 3. Effects of carbidopa and tolcapone on the regionalbrain distribution of radioactivity after i.v. injection of [‘8F]-FDOPA (A) and _ I’8FlJ?PDOPA (B) into mice. Three groups of mice were used: (left) control mice; (center) mice pretreated with carbidopa; (right) mice pretreated with carbidopa and tolcapone. Symbols: circle, striatum; tri* cerebral cortex; and angle, square, cerebellum. Mean *SD
r*F-Labeled
*4 3
.z d
299
Dopa Prodrug
I
control
carbidopa
I
2
1
-+
-F
i
1 U--o--
striatumlcortex striatumkerebellum
- carbihpa and tolcapone
0
60
0
FIG. 4. Effects of carbidopa and tolcapone on the uptake ratios of striatum to cerebral cortex and of striatum to cerebellum after i.v. injection of [‘sF]FDOPA (A) and [ ‘sF]FPDOPA (B) into mice. Three groups of mice were used: (left) control mice; (center)mice pretreated with carbidopa; (right) mice pretreated with carbidopa and tolcapone. Symbols: circle, ratios of striatum to cerebral cortex; and solid circle, ratios of striatum to cerebellum. Mean *SD (n = 4).
0 0
120
60
120
60
120
Time after Injection (min)
B 4
1
1
control
1carbiahpa and tolcapone
carbidopa
3
.; d
2
I
0 I 0
I
1 60
I
I 120
--+---CI I 0
suiatumkortex striatumkerebellum I I I 60 120 0 Time after Injection (min)
I
I 60
,
1 120
I
DISCUSSION
In the case of 6.[‘sF]FPDOPA, an unchanged form was only 2-S% of total ‘*F in all samples, and other metabolites were found with similar proportions as the metabolites of 6-[lRF]FDOPA. A major metabolite of the two compounds was [18F]FMDOPA in all samples. A fifth of total radioactivity in the striatum was detected in [“FIFDA. The [18F]FDA was negligibly found in other brain tissues and in the plasma.
Although 6-[r8F]FDOPA is a useful tracer evaluating presynaptic dopaminergic function with PET, it is susceptible peripherally to metab o 1’IC c h ange. Metabolite analysis is required to measure the precise input function for the quantitative assessment. Suppression of the me-
TABLE 3. Effect of L-phenylalanine Loading on Tissue Distribution min After i.v. Injection of 6.[‘sF]FDOPA and 6-[“F]FPDOPA into Mice % Injected
Blood Striatum Cerebral cortex Cerebellum
3.32 0.74 0.57 0.67
* Mean k SD (n = 4-5). Student’s t-tests were carried
sp < 0.05.
6.[ “F]FPDOPA
+ L*phenylalanine
f 0.41 f 0.18 AZ0.13 + 0.14
out between
3.55 0.56 0.30 0.41
5
dose/g tissue*
6-[ “FIFDOPA Control
of Radioactivity CarbidopaHTreated
f f f f
0.38 0.15 0.05$ 0.06+
the control
Control 1.90 0.23 0.24 0.32
+ + + f
0.22 0.08 0.06 0.01
and the L-phenylalanine-loading
+ L-phenylalanine 2.76 0.12 0.15 0.20
group.
f f f f
0.373 0.02t 0.02t 0.03t
tp < 0.01.
300
K. Ishiwata
0
30
60
Retention time (min)
FIG. 5. An HPLC chromatogram of the ‘sF-labeled metabolites in the striatum 30 min after i.v. injection of [‘sF]FPDOPA into mice pretreated with carbidopa. HPLC conditions are described in text. Radioactivity in each 1 mL fraction was measured with an auto-gammer counter. Retention times of Ledopa and related metabolites are indicated in parentheses: 3,4-dihydroxyphenylacetic acid, DOPAC; homovanik acid, HVA; 3-O-methyldopa, MDOPA; dopamine, DA; and MTYR, 3emethoxytyramine.
tabolism by oral administration of enzyme inhibitors such as carbidopa and nitecapone (12, 18, 24), is usually preferred. Although HPLC analysis is mostly reliable for the metabolite analysis of 6-[“FIFDOPA in the plasma, it is time-consuming and complicated for routine PET studies. Therefore, simple methods have been proposed in lieu of HPLC (2, 19). However, medication or metabolites analysis would not be necessary if one could use a 6-[18F]FDOPA analog that is stable to peripheral metabolism. We considered the use of an “F-labeled dopa prodrug for the PET study of dopamine metabolism and evaluated the potential of the drug. The present work represents several characteristics of 6-[‘*F]FPDOPA as a prodrug of 6-[‘*F]FDOPA. In the mice without medication, the striatal uptake of 6-[18F]FPDOPA was lower but more selective, compared with 6-[i8F]FDOPA. Pretreatment of the mice with carbidopa,
TABLE 4. MetaboIites of 6-[“F]FDOPA sues 30 min After i.v. Injection
an inhibitor of AADC, enhanced the striatal uptake of both compounds. Tolcapone (RO 40-7592) has been shown to be a potent peripherally and centrally acting reversible inhibitor of COMT (27). Combined use of this inhibitor and carbidopa resulted in an increased striatal uptake of two tracers as well as increased uptake ratios of striaturn to reference regions as described previously in the study of 6-[‘aF]FDOPA (10). On the other hand, these effects of the enzyme inhibitors indicate that the 6-[18F]FPDOPA is also susceptible to the metabolic alteration, as discussed below, and that the bulky pivaloyl group in the equilibrium of the 4-O-isomer and 3-O-isomer (13) did not necessarily interfere with the COMT reaction in mice. Tissue distribution of 2-[‘8F]FPDOPA showed no selective uptake by the striatum as in the case of 2-[i8F]FDOPA (data not shown). This result also supports the notion that the pivaloyl group has a minor effect on the inhibition of the COMT. Reduced brain uptake of 6-[‘*F]FPDOPA by loading of L-phenylalanine demonstrates that the compound crosses the blood-brain barrier through the large neutral amino acid transport system as 6-[18F]FDOPA. Although the compound may be hydrolyzed by peripheral esterase (26), presence of 6-[i8F]FPDOPA in the brain tissues supports the transport of 6-[18F]FPDOPA itself into the brain. Although the pivaloyl ester makes 6-[“F]FDOPA more lipophilic, the brain uptake decreased, suggesting that the bulky group on the catechol O-ester interfered with the transport. A preliminary metabolite study confirmed the presence of 6-[‘8F]FPDOPA, 6-[“FIFDOPA, and 6-[“FIFDA in the striatum. However, a major metabolite in the plasma and brain tissues was 6-[“FIFMDOPA. These results are explained in two ways. The first is that 6-[“FIFPDOPA was taken up as a prodrug by the brain and hydrolyzed to 6-[‘8F]FDOPA. Second, the 6-[“FIFPDOPA was hydrolyzed peripherally to 6,[“FIFDOPA, which was also taken up by the brain. Clearly, the biological half-life of 6-[18F]FPDOPA as a prodrug was shorter than 30 min. Therefore, in view of kinetic analysis of the dopamine metabolism both 6-[18F]FPDOPA and 6-[‘8F]FDOPA should be considered as an input function. In the rodents, the degradation of [i8F]FDOPA analogs is extremely rapid as compared with primates (3). Therefore, further studies in primates by PET will evaluate the potential of 6-[“F]FPDOPA as a PET tracer. In conclusion, 6-[“FIFPDOPA showed characteristics as a prodrug of 6-[‘*F]FDOPA. Compared to 6-[‘*F]FDOPA, more selective uptake of the 6-[‘*F]FPDOPA by the striatum was found in the mice with and
and 6.[‘sF]FPDOPA
Metabolites 6.FDOPA 6-[18F]FDOPA Plasma Striatum Cerebral cortex Cerebellum 6-[“FIFPDOPA Plasma Striatum Cerebral cortex Cerebellum * Mean f SD (n = 3-4).
14.2 f 6.1
2.3 f 0.9 8.0f4.1 6.4+ 3.7 4.8 f3.2 2.1+ 1.2
5.5 * 1.9 4.8f3.4
in Plasma and Brain Tis-
(%)* FDA
FMTYR
FHVA
FMDOPA
3.2 f 7.8k 6.8 f 5.1 f
57.5 f 6.2 48.9f 12.8 63.5 f4.0 68.8 f 5.8
0.8 f0.3 18.3 k8.4 3.9 kO.9 1.9k1.8
2.6f2.6 6.9 f0.9 1.8k2.0
52.2 f9.6 58.0 5 6.8 68.1 I!Z8.5 70.1f9.6
1.9 kO.3 13.9 k4.2 1.8_+ 1.3 2.2 f 1.8
0.3 f 0.4 8.7 k0.9 5.8_+ 1.7 4.1 f 0.4
1.7 1.9 2.5 1.0
4.0+3.3 6.0+ 1.2 6.3 k4.4 4.6kl.l
et al.
6.FI’DOPA
0.1 fO.l
4.5 f 7.5 2.2 k3.0 2.4 f 1.8 2.6f 1.2
“F-Labeled
301
Dopa Prodrug
without pre-medication. However, peripheral metabolism of the compound still has disadvantages in that metabolite analysis in the plasma is required for quantitative assessment of dopamine metabolism by PET, as is the case of 6-[‘*F]FDOPA.
This work was supported by a grant-in-aid for scientific research from the Ministry of Education, Science and Culture, Japan, and a grant from the Kudo Science Foundation. The authors thank Dr. K. Tomimoto, Banyu Pharmaceutical Co., Tokyo, for supplying PDOPA and valuable discussion; Dr. K. Hatano, Iwate Medical School, Morioka, Japan, for supplying L-methyl-N-acetyl-[P-3-methoxy-4-acetoxyphenyl)]alaninate, Sankyo Co. Ltd., Tokyo; and Hoffman-La Roche, Basel, Switzerland, for supplying carbidopa and tolcapone (RO 40-7592), respectively, as well as the staff of the Positron Medical Center, TMIG, for their cooperation.
13
14
15.
16.
17
References 1. Adam M. J. and Jivan S. (1992) Synthesis and separation of 3-O-methyl-2and 6-[‘sF]-fluorodopa. J. Label. Compds. Radiophurm. 31, 39-43. 2. Chan G. L.-Y., Hewitt K. A., Pate B. D., Schofield P., Adam M. J. and Ruth T. J. (1992) Routine determination of [‘*F]-L-6-fluorodopa and its metabolites in blood plasma is essential for accurate positron emission tomography studies. Life Sci. 50, 309-318. 3. Gumming P., Haiisser M., Martin W. R. W., Grierson J., Adam M., Ruth T. J. and McGeer E. D. (1988) Kinetics of m vitro decarboxylation and the m viva metabolism of 2-18F- and 6-‘*F-fluorodopa in the hooded rat. Biothem. Pharmacol. 37, 247-250. 4. Gumming P., Boyes B. E., Martin W. R. W., Adam M., Grierson J., Ruth T. J. and McGeer E. D. (1987) The metabolism of [18F]6-fluoro-L-3.4-dihydroxyphenylalanine in the hooded rat. J. Neurochem. 48, 601608. 5. Gumming P., LQer G. C., Kuwabara H. and Gjedde A. (1993) Pharmacokinetics of plasma 6-[‘8F]fluoro-L-3.4-dihydroxyphenylalanine ([18F]FDOPA) in humans. J. Cereb. Blood Flow Metab. 13, 668-675. 6. Dejesus 0. T., Sunderland J. J., Nickles J. R., Mukherjee J. and Appelman E. H. (1990) Synthesis of radiofluorinated analogs of m-tyrosine as potential L-dopa tracers via direct reaction with acetylhypofluorite. Appl. R&at. Isot. 41, 433437. 7. Doudet D. J., McLellam C. A., Carson R., Adams H. R., Miyake H., Aigner T. G., Finn R. T. and Cohen R. M. (1991) Distribution and kinetics of 3-0-methyl-6-[‘8F]fluoro-L-DOPA in the rhesus monkey brain. J. Cereb. Blood Flow Metab. 11, 726-734. 8. Flrnau G., Sood S., Chirakal R., Nahmias C. and Garnett E. S. (1987) Cerebral metabolite of 6-[“Flfluoro-L-3,4-dihydroxyphenylalanine in the pnmate. J. Neurochem. 48, 1077-1082. 9. Flmau G., Sood S., Chlrakal R., Nahmias C. and Garnett E. S. (1988) Metabolite of 6-[“Flfluoro-L-dopa in human blood. J. Nucl. Med. 29, 363369. IO. Giinther I., Psylla M., Reddy G. N., Antonini A., Hnberli M., Beer H.-F. and Leenders K. L. (1994) Comparison of bioavailability of (NCA) 6-[18fifluoro-L-dopa with different enzyme inhibitors. J. Label. Compd.s. Radiopharrn. 35, 480482 (Abstract). 11. Hisaka A., Kasamatsu S., Takenaga N. and Ohtawa M. (1990) Absorption of a novel prodrug of L-dopa, L-3-(3shydroxy-4-pivaloyloxyphenyl)alanine (NB-355). Drug Merab. Dispos. 18, 621-625. 12. Hoffman J. M., Melega W. P., Hawk T. C., Grafton S. C., Luxen A., Ma-
18.
19.
20.
21.
22. 23.
24.
25.
26. 27.
honey D. K., Barrio J. R., Huang S.-C., Mazziotta J. C. and Phelps M. E. (1992) The effects of carbidopa administration on 6-[‘*F]fluoro-L-DOPA kinetics in positron emission tomography. J. Nucl. Med. 33, 1472-1477. Ihara M, Nakajima S., Hisaka A., Tsuchiya Y., Sakuma Y., Suzuki H., Kitani K. and Yano M. (1990) Hydrolysis and acyl migration of a catechol monoester of L-dopa: L-3-(3-hydroxy-4-pivaloyloxyphenyl)alanine. J. Phm. Sci. 79, 703-708. Ihara M., Tsuchiya Y., Sawasaki Y., Hisaka A., Takehana H., Tomimoto K. and Yano M. (1989) A new potential prodrug to improve the duration of L-dopa: L-3-(3.hydroxy-4-plvaloyloxyphenyl)alanine. J Phurm. Sci. 78, 525-529. Ishiwata K., lshii S. and Senda M. (1993a) HPLC using physiological saline for the quality control of radiopharmaceuticals used in PET studies. Appl. R&z. Isot. 44, 1119-1124. Ishiwata K., Ishii S., Senda M., Tsuchlya Y. and Tomimoto K. (199310) Electrophilic synthesis of 6-[18F]fl uoro-L-dopa: Use of 4-0-pivaloyl-L-dopa as a suitable precursor for routine production. ApeI. R&at. Isot. 44, 755759. Ishiwata K., Kubota K., Kubota R., Iwata R., Takahashi T. and Ido T. (1991) Selective 2-[18F]fluorodopa uptake for melanogenesis in murine metastatic melanomas. J. Nucl. Med. 32, 95-101. Laihinen A., Rinne J. O., Rinne U. K., Haaparanta M., Routsalainen U., Bergman J. and Solin 0. (1992) [‘*F]-6-Fluorodopa PET scanning in Parkinson’s disease after selective COMT inhibitlon with nitecapone (OR462). Neurology 42, 199-203. McLellan C. A., Doudet D. J., Brucke T., Aigner T. G. and Cohen R. M. (1991) New rapid analysis method demonstrates differences in 6-[‘sF]fluoro-L-dopa plasma input curves with and without carbidopa and in hemiMPTP lesioned monkeys. Appl. R&z. Isot. 42, 847-854. Melega W. P., Grafton S. T., Huang S.-C., Satyamurthy N., Phelps M. E. and Barrio J. R. (1991) ~-6-[‘~F]fl uoro-DOPA metabolism in monkeys and h “mans: Biochemical parameters for the formation of tracer kinetic models with positron emission tomography. J. Cereb. Blood Flow Metnb. 11, 890897. Murari D., Dejesus 0. T., Sunderland J. J. and Nickles R. J. (1992) [‘8F]Fluoro-~-fluoromethylene-m-tyrosine analogs, potential PET agents for presynaptic dopamine terminals: Synthesis and spectroscopic characterization. Appl. Radiar. Isot. 43, 969-977. Namavari M., Satyamurthy N., Phelps M. E. and Barrio J. R. (1993) Synthesis of 6-[18F] and 4-[‘8F]fluoro-L-m-tyrosine via regioselective radiofluorodestannylation. Appl. R&at. Isot. 44, 527-536. Sawle G. V., Bum D. J., Morrish P. K., Lammertsma A. A., Snow B. J,, Luthra S., Osman S. and Brooks D. J. (1994) The effect of entecapone (OR-61 1) on brain [‘8F]-6-L-fluorodopa metabolism: Implications for levedopa therapy of Parkinson’s disease. Neurology 44, 1292-1287. Tedroff J., Aquilonius S.-M., Laihinen A., Rinne U., Hartvig P., Andersson J., Lundqvist H., Haaparanta M., Solin O., Antoni G., Gee A. D., Ulin J. and LBngstrlim B. (1990) Striatal kinetics of [“Cl-(+)-nomifensine and 6-[18F]fluoro-L-dopa in Parkinson’s disease measured with positron emission tomography. Acta Neural. SC&. 81, 24-30. Tye S. J., Rupniak N. M. J., Naruse T., Miyaji M. and Iversen S. D. (1989) NB-355: A novel prodrug for L-dopa with reduced risk for peak-dose dyskinesias in MPTP-treated squirrel monkeys. Clin. Neuropharmacol 12, 393403. Zur M. and Nemerson Y. (1978) The esterase activity of coagulation factor VII. Evidence for intrinsic activitv of the zvmoeen. I v -J. Biol. Chem. 253. 2203-2209. Ztircher G., Keller H. H., Kettler R., Borgulya J., Bonetti E. P., Eigenman R. and Da Parada M. (1990) RO 40-7592, a novel, verv _ notent and orallv active inhibitor of catechol-O-methyltransferase: A pharmacological study in rats. Adu. Neural. 53, 497-503.
ISSN 0969-8051/96/$15.00 + 0.00 SSDI 0969-8051(95)02083-7
1996
ELSEVIER
Synthesis and Evaluation of an “F-Labeled Dopa Prodrug as a PET Tracer for Studying Brain Dopamine Metabolism Kiichi Ishiwata , ’ * Masaki Shinoda, ’ ,2 Shin-Ichi Ishii, ’ Tadushi Nozaki’ and Michio Senda’ ‘POSITRON MEDICAL CENTER, TOKYO METROPOLITAN INSTlTUTE OF GERONTOLOGY, l-l TOKYO 173, JAPAN, AND *FACULTY OF HYGIENIC SCIENCE, KITASATO UNIVERSITY, SAGAMIHARA 228, JAPAN
NAKA-CHO, ITABASHI, l-15-1 KITASATO,
ABSTRACT. In the quantitative studies of presynaptic dopamine metabolism by PET with 6-[“FlfluoroL-dopa (6-[‘8F]FDOPA), metabolic analysis in the plasma is required to determine the precise input function because of susceptibility of the compound to peripheral metabolism. In this study, we prepared 6418F]fluoro-O-pivaloyl-L.-dopa (6-[ ‘sF]FPDOPA) as a prodrug of 6-[“FIFDOPA, and evaluated its potential as a PET tracer in mice. If the 6-[ “F]FPDOPA is stable peripherally and is hydrolyzed to 64 18F]FDOPA in the brain tissues, disadvantage of the 6-[ “FIFDOPA will b e overcome. Compared with the 6-[ 18F]FDOPA, the initial brain uptake of the 6-[18F]FPDOPA was lower; however, the uptake in the latter become comparable, and the uptake ratios of striatum to other reference regions were larger. Medication of mice with inhibitors of aromatic amino acid decarboxylase and catechol-O-methyl transferase greatly enhanced the striatal uptake of the two compounds. The reduced brain uptake of the compounds by L-phenylalanineloading suggested transport through the blood-brain barrier by the neutral amino acid transporter. HPLC analysis showed the presence of 6J18F]FPDOPA, 6-[r8F]FDOPA and 6-[“Flfluorodopamine in the striatum; however, 6-[ “Flfluoro-3-0-methyl+dopa was a predominate metabolite in the brain and plasma as in the case of [t’F]FDOPA. Results suggested that 6-[18F]FPDOPA had characteristics as a prodrug of 6-[‘8F]FDOPA; however, the compound was also labile to metabolic alteration in viva. NUCL MED BIOL 23;3:295-301, 1996. KEY WORDS. Dopamine
metabolism,
Prodrug,
6-[‘“Flfluoro-O-pivaloyl-L-dopa,
Positron
emission
tomogra-
phy
INTRODUCTION
sibility, this study deals with the use of a prodrug of 6-[18F]FDOPA as a PET tracer. The compound L-3-(3-hydroxy-4-pivaloyloxyphenyl)alanine (4-Opivaloyl-L-dopa) has been developed as a potential prodrug of L-dopa to improve therapeutic effect (11, 14, 25). The compound showed longer mean residence time and larger bioavailability of L-dopa concentrations than did L-dopa itself in rats. The pivaloyl ester of the compound is hydrolyzed in ho; however, the bulky pivaloyl group may interfere with the COMT reaction because the pivaloyl group migrates to equilibrium of the 4-O-isomer (53-59%) and 3-O-isomer in an aqueous solution (13). Electrophilic fluorination of 4-0-pivaloyl-L-dopa with acetyl [‘*F]hypofluorite gave predominantly 6-[‘sF]fluoro-O-pivaloylL-dopa (6-[18F]FPDOPA) (16). We expected that the 6-[“flFPDOPA was taken up by the brain and then hydrolyzed to 6-[‘“FIFDOPA. If peripheral hydrolysis of the ester-bond of the compound is substantially suppressed during a couple of hours, metabolite analysis is not required in PET studies. In this work, we characterized the properties of the compound in mice, compared with 6-[18F]FDOPA, and evaluated the potential as a PET tracer.
Currently, 6-[‘sF]fluoro-L-3,4-dihydroxyphenylalanine (6-[i*F]fluoro-L-dopa, 6-[‘8F]FDOPA) is used for studying the presynaptic dopaminergic function by positron emission tomography (PET). The 6-[18F]FDOPA is labile to metabolic alteration in rodents as well as primates (3-5,9, 20). Some metabolites such as 6-[18F]fluorodopamine ([i8F]FDA) are not taken up by the brain. However, 6-[‘*F]fluoro-L-3methyoxy-4-hydroxyphenylalanine (6-[18F]fluoro-3-0-methyl-L-dopa, [i8F]FMDOPA), which is produced by catechol-O-methyl transferase (COMT), crosses the blood-brain barrier into the brain and is present homogeneously in the brain tissues (7, 8). Therefore, metabolite analysis in the plasma is required to determine the precise input function in the quantitative assessment of dopamine metabolism. To suppress the metabolism of 6-[i8F]FDOPA, human subjects are usually given selective inhibitors of peripheral aromatic amino acid decarboxylase (AADC) (12, 24) prior to PET studies. Combined use of the COMT and AADC inhibitors to diminish [“FIFMDOPA in the plasma shows great advantage to the PET studies (18, 23). Another approach to overcome the disadvantage of the 6-[18F]FDOPA is the choice of less susceptible compounds to the metabolism. Meta-[‘sF]fluoro-L-tyrosine derivatives are candidates (6, 21, 22). As a third pos-
MATERIALS
*Author for correspondence. Accepted 31 October 1995.
The 4-0-pivaloyl-L-dopa was supplied by Banyu Pharmaceutical Co. (Tokyo), and L-methyl-N-acetyl-[P-3-methoxy-4-acetoxyphenyl)]alaninate was from Dr. K. Hatano, Iwate Medical School (Morioka, Japan). Both 3-O-methyl-D,L-dopa and 3-methoxytyramine were purchased
AND METHODS
296
K. Ishiwata
from Sigma Chemical Co. (St. Louis, MO); dopamine, D,L-dopa, homovanilic acid, and L-phenylalanine were from Wako Pure Chemical Industries (Tokyo); and 3,4-dihydroxyphenylacetic acid was from Aldrich Chemical Co. (Milwaukee, WI). Carbidopa and tolcapone (RO 40-7592) were supplied from Sankyo Co. (Tokyo) and Hoffman-La Roche (Basel, Switzerland), respectively. Other chemicals used were analytical grade and purchased from Wako Pure Chemical Industries (Tokyo).
Acetyl [“Flhypofluorite, which was produced using a previously reported method (15, 16), was bubbled into 6 mL of acetic acid containing 15-20 mg of 4-O-pivaloyl-L-dopa at room temperature (Fig. 1). After the solution was evaporated to dryness, the residue was dissolved in 2 mL of H,O. The solution was applied to separation by high-performance liquid chromatography (HPLC). A preparative reverse-phase column used was YMC-Pack ODS S-5 (20 mm i.d. x 250 mm length, YMC Co. Ltd., Kyoto). The mobile phase was a mixture of CH,OH and 50 mM NaH,PO+ (35:65, v/v), and passed at a flow rate of 10 mL/min. The elution profile was monitored with a UV detector at 271 nm and an NaI radioactivity detector. Fractions of 6-[“F]FPDOPA and 2-[‘*F]FPDOPA were collected and evaporated to dryness. The residue was dissolved in physiological saline and passed through a 0.22~p,rn membrane filter. Both 6,[“F]FDOPA and 2-[“F]FDOPA were prepared as previously reported (15, 16). The specific activity of 6-[“F]FDOPA was 23.9 - 30.0 GBq/mmol. The [‘*F]FMDOPA was synthesized using L-methyl-N-acetyl-[P-3-methoxy-4-acetoxyphenyl)]alaninate as a starting material according to the method of Adam and Jivan (1). The [“F]FDA was prepared enzymatically using AADC in rat striatum homogenate as previously described (17). Radiochemical purity of compounds was analyzed by HPLC. For 6-[“FIFPDOPA and 2-[“FIFPDOPA, a Crestpak C18S (4.6 mm i.d. x 150 mm length, Japan Spectroscopic Co., Ltd., Tokyo) was eluted with a mixture of CH,OH and 50 mM NaH,P04 (25:75, v/v) at a flow rate of 2 mL/min at 50°C. For 6-[“F]FDOPA, 2-[“F]FDOPA, and [‘8F]FMDOPA, the mobile phase was a mixture of CH,OH and 0.1% acetic acid containing 1 mM EDTA disodium and 1 mM sodium octylsulfate (1:9, w/w) at a flow rate of 2 mL/min at room temperature.
Biodistribution
Study
Either 6-[“FIFPDOPA or 6-[“F]FDOPA was injected intravenously into four groups of male ddY mice (8 to 9 weeks old), and tissue distribution of the radioactivity was measured. The first group was the control; the second group was given carbidopa (5 mg/kg) intraperitoneally 30 min before the tracer injection; the third group was given cardibopa (5 mg/kg) and tolcapone (10 mg/kg) intraperitoneally 30 min and 60-90 min, respectively, before the tracer injection. The last group, which was pretreated with carbidopa, was received with a mix-
(CH3)3CCOO-
ture of tracer and L-phenylalanine (100 mg/kg). These mice received 1.0 MBq of tracers, and then were killed by cervical dislocation at 5, 15,30, 60, and 120 min postinjection. The blood was removed by heart puncture using a heparinized syringe. The tissues were dissected and weighed. The radioactivity of the tissues was counted using an autogamma counter. The tissue uptake of radioactivity was expressed as the percent of injected dose per gram of tissue.
Metabolic
Radiopharmaceuticals
CH2CH-C02H
et al.
Study
Mice pretreated with carbidopa received 10 MBq of tracers, and were killed by cervical dislocation 30 min postinjection. The blood removed by heart puncture was centrifuged to obtain plasma. The plasma, striaturn, cerebral cortex, and cerebellum were treated to prepare the samples for HPLC by a previously described method (17). Metabolites in the samples were analyzed by HPLC using a Nova-Pak Cl8 column (8 mm i.d. x 100 mm length, Waters) equipped in an RCM 8 x 10 module. The column was eluted with an ion pair solution with a gradient modifier at a constant flow rate of 2 mL/min at room temperature. The initial solution was a mixture of solution A (1% acetic acid containing 1 mM EDTA and sodium octylsulfate) and solution B (40% MeOH containing 1% acetic acid, 1 mM EDTA, and sodium octylsulfate (3/l, v/v) for the first 5 min, after which a solvent with a linearly increasing gradient of solution B (25-100%) was eluted for 40 min. During the last 15 min solution B was delivered, and then the column was reequilibrated with the initial conditions. The elution profile was detected with a UV detector and a radioactivity monitor (FLO-ONE/Beta A200, Packard), and the radioactivity in each 1.0 mL collected using a fraction collector was measured with an auto-gamma counter. Radioactivity was corrected for the half-life of 18F. A portion of the sample applied was also measured to calculate the total applied radioactivity, and the percentage of radioactivity in each peak of the total applied radioactivity was calculated. Recovery of the radioactivity was essentially quantitative.
RESULTS Radiochemistry Fluorination of 4-0-pivaloyl-L-dopa with acetyl [“flhypofluorite (Fig. 1) followed by HPLC-separation (Fig. 2) gave two major labeled components and other minor components. HCl-hydrolysis of peaks 1 and 3 gave 6-[“FIFDOPA and 2-[“F]FDOPA, respectively, which were identified by comparing their retention times with those of the authentic samples. Peaks 1 and 3 showed maximal UV absorption at 276 nm and 269 nm, corresponding to hmax values of 6-[“FIFDOPA (283 nm) and 2-[“F]FDOPA (271 nm), respectively (3). Therefore, peaks 1 and 3 were identified as 6-[“FIFPDOPA and 2-[“FIFPDOPA, respectively. Peak 4 was not identified, but probably is 2,6-di-[‘*F]fluoro-derivative (described elsewhere). The maximal radiochemical yields of the 6-[“FJFPDOPA and 2-[“FIFPDOPA were 26% and 15%, respectively,
1) AcO’*F, AcOH 2) HPLC separation
t
(CH3)3CCOO-
iJH2
/
\
CH2CH-C02H kH2
Q 18F
FIG. 1. Synthesis of ~-3*(6-[ ‘*F]fluoro-3*hydroxy-4-pivaloyloxyphenyl)alanine
([ ‘sF]FPDOPA).
“F-Labeled
297
Dopa Prodrug
..a ~
~ . . . . . . . . . . . . . . . . . “”
. ...+* p
M -<
..:! :
: i
%
.
; ::: iF
: :
i’ .
I :
j’ 1: i’
. . . . I......
..:
:,......
:..&..
:;
::
: i
I
1
I 0
L
20
40
60
L 80
100
Retention time (min) FIG. 2. A preparative HPLC chromatogram of the reaction mixture of fluorinated ~-34 3.hydroxy-4.pivaloyloxyphenyl)alanine with acetyl ] ‘sF]hypofluorite in acetic acid using a reverse-phase column. HPLC conditions are described in text. Peak 2 corresponds to 4-0.pivaloyl+dopa.
based on acetyl [“Flhypofluorite, with radiochemical and the preparation time of 100 min.
purity
of >99%
Biodistribution The tissue distribution of 6-[“FIFDOPA and 6-[‘*F]FPDOPA are summarized in Tables 1 and 2. Two tracers showed similar tissue distribution. The highest uptake was observed in the pancreas followed by the
TABLE Mice
1. Tissue
Distribution
of Radioactivity
kidney and small intestine. The pancreas uptake of 6-[‘*F]FDOPA slightly increased for the first 30 min, whereas that of 6-[“FIFPDOPA decreased with time. The brain uptake of two compounds was the lowest among the organs studied, and this decreased with time. Figure 3 shows regional brain distribution of the two tracers and the effect of pretreatment with enzyme inhibitors on the distribution. The uptake of 6-[‘*F]FDOPA in the striatum of the control mice was slightly higher than that in the cerebral cortex and cerebellum. The radioactivity levels in three regions decreased with time. On the other hand, pretreatment with carbidopa enhanced the striatal uptake in which the level of radioactivity increased for the first 30 min, then decreased. Pretreatment with both carbidopa and tolcapone resulted in the largest difference between the striatal uptake and the uptake by other regions. The striatal uptake increased for the first 60 min and remained nearly constant. In control mice, the uptake of 6-[‘*F]FPDOPA by the striatum, cortex, and cerebellum was significantly lower at 5 min (p < 0.01) than that of 6-[‘sF]FDOPA. The striatal uptake increased slightly for the first 30 min, and then decreased with time. After 30 min, radioactivity levels of the two tracers in the striatum were comparable, but the level for 6-[‘*F]FPDOPA in the cortex and cerebellum was significantly lower than that for 6-[‘*F]FDOPA (p < 0.01). Consequently, the difference between the striatal uptake and the uptake by reference regions was larger in the 6-[‘*F]FPDOPA than in 6-[“F]FDOPA. Carbidopa also enhanced the striatal uptake of the 6-[“FIFPDOPA; however, the effect was not as large as compared with the case of 6-[t*F]FDOPA. Pretreatment with both carbidopa and tolcapone resulted in the increasing striatal uptake during the entire period investigated. In Figure 4, the uptake ratios of striatum to other reference regions are represented. In the control group, uptake ratios of striatum to other tissues for 6-[‘*F]FPDOPA were greater than those for 6-[‘*F]FDOPA. In the mice treated with carbidopa alone, the ratios for the 6-[‘*F]FPDOPA increased with time, whereas the ratios for the 6-[‘*F]FDOPA remained constant after 60 min. The two enzyme inhibitors made the ratios larger for both compounds. Co-injection of L-phenylalanine significantly reduced the initial uptake (5 min postinjection) of both 6-[‘*F]FDOPA and 6-[‘*F]FPDOPA to 52 - 76% of the control in three brain regions (Table 3). Metabolism Metabolites in the plasma and brain were analyzed by HPLC. chromatogram of the striatal metabolites of 6-[‘*F]FPDOPA
After
% Injected 5min Blood Brain Heart Lung Liver Spleen Pancreas Small inte ,stine Kidney Muscle Bone a Mean + SD (n = 4).
1.74 It 0.19
0.42 kO.08 1.52 f0.16 1.55 kO.16 3.49 kO.48 1.68 f 0.24 11.19kO.63
5.30+ 1.12 9.27 3~2.98 1.10 f 0.15 1.31 kO.27
15 min
0.90+0.06 0.30 f0.04 0.73 kO.04 0.89 f0.06 2.10 f 0.68 0.83 f 0.32 12.16 f 1.78
4.38 f0.59 4.71 f 1.16 0.85 kO.05 0.81 kO.12
i.v. Injection
of 6-t “F]FDOPA
into
dose/g tissue”
30 min
60 min
0.64 f 0.12 0.25 fO.O1 0.51 f 0.06 0.70* 0.30 1.13 +0.39 0.50 f0.05 12.65 kO.16 2.75 k0.47 3.05 f0.57 0.69 f0.04 0.69 f.O.07
0.38 f 0.06 0.15 fO.O1 0.34kO.02 0.49 zk 0.13 0.61kO.32 0.31 + 0.05 7.81 3~ 1.18 1.83 kO.61 2.68 fO.12 0.44 f0.05 0.51 f 0.04
120 min
0.22 f0.03 O.lOfO.O1 0.21 * 0.01
0.46 f 0.37 0.25 +0.03 0.22 f0.04 4.28kO.77 1.12 f 0.38 2.41 kO.77 0.25 fO.O1 0.42 f0.07
A typical is shown
298
K. Ishiwata
TABLE Mice
2. Tissue
Distribution
of Radioactivity
After
% Injected
2.30f 0.13 0.23 + 0.02 1.11 f0.14 1.70 + 0.16
7.63 kO.75 2.08 f 0.12 14.83 + 2.36 4.53 310.69 19.30 f 1.88 0.58 kO.03 1.07 f 0.10
of 6-[ 18F]FPDOPA
into
dose/g tissue*
15 min
30 min
60 min
120 min
1.04 f0.27 0.22 f 0.07 0.59 f 0.08 0.92 k 0.21 4.40 k0.76 0.86 +0.09 a.57 f 1.40 2.48 k0.87 4.89 f 1.75 0.04 f0.03 0.67 k 0.28
0.66 f0.05 0.18 rkO.02 0.39 f 0.05 0.55 * 0.07 0.94 IL 0.05 0.40 IL 0.03 7.26 f 0.79 1.69 f 0.38 4.17 IL 1.17 0.36 f 0.04 0.49 f 0.10
0.26 kO.05 0.09 IL 0.01 0.16 f0.03 0.26f0.04 0.50,O.ll 0.16 f 0.04 4.86kO.76
0.10 f 0.01
5min Blood Brain Heart Lung Liver Spleen Pancreas Small intestine Kidney Muscle Bone
i.v. Injection
et al.
1.45 + 0.43 2.41 f 0.37
0.19 rto.05 0.31 f 0.03
0.04* 0.00 O.lOfO.O1 0.10 f 0.04 0.16 f 0.04 0.09 f 0.01
2.39 k0.45 0.62 f 0.19 2.36f0.77 0.12 f 0.01 0.25 f 0.02
* Mean k SD (n = 4).
in Fig. 5. Radioactive peaks 2, 4, 5, and 7 were 6-[18F]FDOPA, [‘sF]FMDOPA, [“FIFDA, and 6-[‘*F]FPDOPA by comparing retention times of the authentic samples. Peaks 3 and 6 were tentatively identified to be [‘*F]FHVA and [‘8F]Fm&&omparing their re-
tention times with those of L-dopa and its related metabolites (Fig. 5). A similar analytical profile (except for 6,[“F]FPDOPA) was observed in the metabolites of 6-[“FIFDOPA. Percentages of metabolites in the plasma and brain 30 min after the injection are summarized in Table 4.
A control
carbidopa and tolcapone
carbidopa
1
Time after Injection (min)
control * -
carbidopa and tolcapone
carbidopa
slriatum corbex cerebellum
120
T
0
60
120
Time after injection (min)
0
60
120
FIG. 3. Effects of carbidopa and tolcapone on the regionalbrain distribution of radioactivity after i.v. injection of [‘8F]-FDOPA (A) and _ I’8FlJ?PDOPA (B) into mice. Three groups of mice were used: (left) control mice; (center) mice pretreated with carbidopa; (right) mice pretreated with carbidopa and tolcapone. Symbols: circle, striatum; tri* cerebral cortex; and angle, square, cerebellum. Mean *SD
r*F-Labeled
*4 3
.z d
299
Dopa Prodrug
I
control
carbidopa
I
2
1
-+
-F
i
1 U--o--
striatumlcortex striatumkerebellum
- carbihpa and tolcapone
0
60
0
FIG. 4. Effects of carbidopa and tolcapone on the uptake ratios of striatum to cerebral cortex and of striatum to cerebellum after i.v. injection of [‘sF]FDOPA (A) and [ ‘sF]FPDOPA (B) into mice. Three groups of mice were used: (left) control mice; (center)mice pretreated with carbidopa; (right) mice pretreated with carbidopa and tolcapone. Symbols: circle, ratios of striatum to cerebral cortex; and solid circle, ratios of striatum to cerebellum. Mean *SD (n = 4).
0 0
120
60
120
60
120
Time after Injection (min)
B 4
1
1
control
1carbiahpa and tolcapone
carbidopa
3
.; d
2
I
0 I 0
I
1 60
I
I 120
--+---CI I 0
suiatumkortex striatumkerebellum I I I 60 120 0 Time after Injection (min)
I
I 60
,
1 120
I
DISCUSSION
In the case of 6.[‘sF]FPDOPA, an unchanged form was only 2-S% of total ‘*F in all samples, and other metabolites were found with similar proportions as the metabolites of 6-[lRF]FDOPA. A major metabolite of the two compounds was [18F]FMDOPA in all samples. A fifth of total radioactivity in the striatum was detected in [“FIFDA. The [18F]FDA was negligibly found in other brain tissues and in the plasma.
Although 6-[r8F]FDOPA is a useful tracer evaluating presynaptic dopaminergic function with PET, it is susceptible peripherally to metab o 1’IC c h ange. Metabolite analysis is required to measure the precise input function for the quantitative assessment. Suppression of the me-
TABLE 3. Effect of L-phenylalanine Loading on Tissue Distribution min After i.v. Injection of 6.[‘sF]FDOPA and 6-[“F]FPDOPA into Mice % Injected
Blood Striatum Cerebral cortex Cerebellum
3.32 0.74 0.57 0.67
* Mean k SD (n = 4-5). Student’s t-tests were carried
sp < 0.05.
6.[ “F]FPDOPA
+ L*phenylalanine
f 0.41 f 0.18 AZ0.13 + 0.14
out between
3.55 0.56 0.30 0.41
5
dose/g tissue*
6-[ “FIFDOPA Control
of Radioactivity CarbidopaHTreated
f f f f
0.38 0.15 0.05$ 0.06+
the control
Control 1.90 0.23 0.24 0.32
+ + + f
0.22 0.08 0.06 0.01
and the L-phenylalanine-loading
+ L-phenylalanine 2.76 0.12 0.15 0.20
group.
f f f f
0.373 0.02t 0.02t 0.03t
tp < 0.01.
300
K. Ishiwata
0
30
60
Retention time (min)
FIG. 5. An HPLC chromatogram of the ‘sF-labeled metabolites in the striatum 30 min after i.v. injection of [‘sF]FPDOPA into mice pretreated with carbidopa. HPLC conditions are described in text. Radioactivity in each 1 mL fraction was measured with an auto-gammer counter. Retention times of Ledopa and related metabolites are indicated in parentheses: 3,4-dihydroxyphenylacetic acid, DOPAC; homovanik acid, HVA; 3-O-methyldopa, MDOPA; dopamine, DA; and MTYR, 3emethoxytyramine.
tabolism by oral administration of enzyme inhibitors such as carbidopa and nitecapone (12, 18, 24), is usually preferred. Although HPLC analysis is mostly reliable for the metabolite analysis of 6-[“FIFDOPA in the plasma, it is time-consuming and complicated for routine PET studies. Therefore, simple methods have been proposed in lieu of HPLC (2, 19). However, medication or metabolites analysis would not be necessary if one could use a 6-[18F]FDOPA analog that is stable to peripheral metabolism. We considered the use of an “F-labeled dopa prodrug for the PET study of dopamine metabolism and evaluated the potential of the drug. The present work represents several characteristics of 6-[‘*F]FPDOPA as a prodrug of 6-[‘*F]FDOPA. In the mice without medication, the striatal uptake of 6-[18F]FPDOPA was lower but more selective, compared with 6-[i8F]FDOPA. Pretreatment of the mice with carbidopa,
TABLE 4. MetaboIites of 6-[“F]FDOPA sues 30 min After i.v. Injection
an inhibitor of AADC, enhanced the striatal uptake of both compounds. Tolcapone (RO 40-7592) has been shown to be a potent peripherally and centrally acting reversible inhibitor of COMT (27). Combined use of this inhibitor and carbidopa resulted in an increased striatal uptake of two tracers as well as increased uptake ratios of striaturn to reference regions as described previously in the study of 6-[‘aF]FDOPA (10). On the other hand, these effects of the enzyme inhibitors indicate that the 6-[18F]FPDOPA is also susceptible to the metabolic alteration, as discussed below, and that the bulky pivaloyl group in the equilibrium of the 4-O-isomer and 3-O-isomer (13) did not necessarily interfere with the COMT reaction in mice. Tissue distribution of 2-[‘8F]FPDOPA showed no selective uptake by the striatum as in the case of 2-[i8F]FDOPA (data not shown). This result also supports the notion that the pivaloyl group has a minor effect on the inhibition of the COMT. Reduced brain uptake of 6-[‘*F]FPDOPA by loading of L-phenylalanine demonstrates that the compound crosses the blood-brain barrier through the large neutral amino acid transport system as 6-[18F]FDOPA. Although the compound may be hydrolyzed by peripheral esterase (26), presence of 6-[i8F]FPDOPA in the brain tissues supports the transport of 6-[18F]FPDOPA itself into the brain. Although the pivaloyl ester makes 6-[“F]FDOPA more lipophilic, the brain uptake decreased, suggesting that the bulky group on the catechol O-ester interfered with the transport. A preliminary metabolite study confirmed the presence of 6-[‘8F]FPDOPA, 6-[“FIFDOPA, and 6-[“FIFDA in the striatum. However, a major metabolite in the plasma and brain tissues was 6-[“FIFMDOPA. These results are explained in two ways. The first is that 6-[“FIFPDOPA was taken up as a prodrug by the brain and hydrolyzed to 6-[‘8F]FDOPA. Second, the 6-[“FIFPDOPA was hydrolyzed peripherally to 6,[“FIFDOPA, which was also taken up by the brain. Clearly, the biological half-life of 6-[18F]FPDOPA as a prodrug was shorter than 30 min. Therefore, in view of kinetic analysis of the dopamine metabolism both 6-[18F]FPDOPA and 6-[‘8F]FDOPA should be considered as an input function. In the rodents, the degradation of [i8F]FDOPA analogs is extremely rapid as compared with primates (3). Therefore, further studies in primates by PET will evaluate the potential of 6-[“F]FPDOPA as a PET tracer. In conclusion, 6-[“FIFPDOPA showed characteristics as a prodrug of 6-[‘*F]FDOPA. Compared to 6-[‘*F]FDOPA, more selective uptake of the 6-[‘*F]FPDOPA by the striatum was found in the mice with and
and 6.[‘sF]FPDOPA
Metabolites 6.FDOPA 6-[18F]FDOPA Plasma Striatum Cerebral cortex Cerebellum 6-[“FIFPDOPA Plasma Striatum Cerebral cortex Cerebellum * Mean f SD (n = 3-4).
14.2 f 6.1
2.3 f 0.9 8.0f4.1 6.4+ 3.7 4.8 f3.2 2.1+ 1.2
5.5 * 1.9 4.8f3.4
in Plasma and Brain Tis-
(%)* FDA
FMTYR
FHVA
FMDOPA
3.2 f 7.8k 6.8 f 5.1 f
57.5 f 6.2 48.9f 12.8 63.5 f4.0 68.8 f 5.8
0.8 f0.3 18.3 k8.4 3.9 kO.9 1.9k1.8
2.6f2.6 6.9 f0.9 1.8k2.0
52.2 f9.6 58.0 5 6.8 68.1 I!Z8.5 70.1f9.6
1.9 kO.3 13.9 k4.2 1.8_+ 1.3 2.2 f 1.8
0.3 f 0.4 8.7 k0.9 5.8_+ 1.7 4.1 f 0.4
1.7 1.9 2.5 1.0
4.0+3.3 6.0+ 1.2 6.3 k4.4 4.6kl.l
et al.
6.FI’DOPA
0.1 fO.l
4.5 f 7.5 2.2 k3.0 2.4 f 1.8 2.6f 1.2
“F-Labeled
301
Dopa Prodrug
without pre-medication. However, peripheral metabolism of the compound still has disadvantages in that metabolite analysis in the plasma is required for quantitative assessment of dopamine metabolism by PET, as is the case of 6-[‘*F]FDOPA.
This work was supported by a grant-in-aid for scientific research from the Ministry of Education, Science and Culture, Japan, and a grant from the Kudo Science Foundation. The authors thank Dr. K. Tomimoto, Banyu Pharmaceutical Co., Tokyo, for supplying PDOPA and valuable discussion; Dr. K. Hatano, Iwate Medical School, Morioka, Japan, for supplying L-methyl-N-acetyl-[P-3-methoxy-4-acetoxyphenyl)]alaninate, Sankyo Co. Ltd., Tokyo; and Hoffman-La Roche, Basel, Switzerland, for supplying carbidopa and tolcapone (RO 40-7592), respectively, as well as the staff of the Positron Medical Center, TMIG, for their cooperation.
13
14
15.
16.
17
References 1. Adam M. J. and Jivan S. (1992) Synthesis and separation of 3-O-methyl-2and 6-[‘sF]-fluorodopa. J. Label. Compds. Radiophurm. 31, 39-43. 2. Chan G. L.-Y., Hewitt K. A., Pate B. D., Schofield P., Adam M. J. and Ruth T. J. (1992) Routine determination of [‘*F]-L-6-fluorodopa and its metabolites in blood plasma is essential for accurate positron emission tomography studies. Life Sci. 50, 309-318. 3. Gumming P., Haiisser M., Martin W. R. W., Grierson J., Adam M., Ruth T. J. and McGeer E. D. (1988) Kinetics of m vitro decarboxylation and the m viva metabolism of 2-18F- and 6-‘*F-fluorodopa in the hooded rat. Biothem. Pharmacol. 37, 247-250. 4. Gumming P., Boyes B. E., Martin W. R. W., Adam M., Grierson J., Ruth T. J. and McGeer E. D. (1987) The metabolism of [18F]6-fluoro-L-3.4-dihydroxyphenylalanine in the hooded rat. J. Neurochem. 48, 601608. 5. Gumming P., LQer G. C., Kuwabara H. and Gjedde A. (1993) Pharmacokinetics of plasma 6-[‘8F]fluoro-L-3.4-dihydroxyphenylalanine ([18F]FDOPA) in humans. J. Cereb. Blood Flow Metab. 13, 668-675. 6. Dejesus 0. T., Sunderland J. J., Nickles J. R., Mukherjee J. and Appelman E. H. (1990) Synthesis of radiofluorinated analogs of m-tyrosine as potential L-dopa tracers via direct reaction with acetylhypofluorite. Appl. R&at. Isot. 41, 433437. 7. Doudet D. J., McLellam C. A., Carson R., Adams H. R., Miyake H., Aigner T. G., Finn R. T. and Cohen R. M. (1991) Distribution and kinetics of 3-0-methyl-6-[‘8F]fluoro-L-DOPA in the rhesus monkey brain. J. Cereb. Blood Flow Metab. 11, 726-734. 8. Flrnau G., Sood S., Chirakal R., Nahmias C. and Garnett E. S. (1987) Cerebral metabolite of 6-[“Flfluoro-L-3,4-dihydroxyphenylalanine in the pnmate. J. Neurochem. 48, 1077-1082. 9. Flmau G., Sood S., Chlrakal R., Nahmias C. and Garnett E. S. (1988) Metabolite of 6-[“Flfluoro-L-dopa in human blood. J. Nucl. Med. 29, 363369. IO. Giinther I., Psylla M., Reddy G. N., Antonini A., Hnberli M., Beer H.-F. and Leenders K. L. (1994) Comparison of bioavailability of (NCA) 6-[18fifluoro-L-dopa with different enzyme inhibitors. J. Label. Compd.s. Radiopharrn. 35, 480482 (Abstract). 11. Hisaka A., Kasamatsu S., Takenaga N. and Ohtawa M. (1990) Absorption of a novel prodrug of L-dopa, L-3-(3shydroxy-4-pivaloyloxyphenyl)alanine (NB-355). Drug Merab. Dispos. 18, 621-625. 12. Hoffman J. M., Melega W. P., Hawk T. C., Grafton S. C., Luxen A., Ma-
18.
19.
20.
21.
22. 23.
24.
25.
26. 27.
honey D. K., Barrio J. R., Huang S.-C., Mazziotta J. C. and Phelps M. E. (1992) The effects of carbidopa administration on 6-[‘*F]fluoro-L-DOPA kinetics in positron emission tomography. J. Nucl. Med. 33, 1472-1477. Ihara M, Nakajima S., Hisaka A., Tsuchiya Y., Sakuma Y., Suzuki H., Kitani K. and Yano M. (1990) Hydrolysis and acyl migration of a catechol monoester of L-dopa: L-3-(3-hydroxy-4-pivaloyloxyphenyl)alanine. J. Phm. Sci. 79, 703-708. Ihara M., Tsuchiya Y., Sawasaki Y., Hisaka A., Takehana H., Tomimoto K. and Yano M. (1989) A new potential prodrug to improve the duration of L-dopa: L-3-(3.hydroxy-4-plvaloyloxyphenyl)alanine. J Phurm. Sci. 78, 525-529. Ishiwata K., lshii S. and Senda M. (1993a) HPLC using physiological saline for the quality control of radiopharmaceuticals used in PET studies. Appl. R&z. Isot. 44, 1119-1124. Ishiwata K., Ishii S., Senda M., Tsuchlya Y. and Tomimoto K. (199310) Electrophilic synthesis of 6-[18F]fl uoro-L-dopa: Use of 4-0-pivaloyl-L-dopa as a suitable precursor for routine production. ApeI. R&at. Isot. 44, 755759. Ishiwata K., Kubota K., Kubota R., Iwata R., Takahashi T. and Ido T. (1991) Selective 2-[18F]fluorodopa uptake for melanogenesis in murine metastatic melanomas. J. Nucl. Med. 32, 95-101. Laihinen A., Rinne J. O., Rinne U. K., Haaparanta M., Routsalainen U., Bergman J. and Solin 0. (1992) [‘*F]-6-Fluorodopa PET scanning in Parkinson’s disease after selective COMT inhibitlon with nitecapone (OR462). Neurology 42, 199-203. McLellan C. A., Doudet D. J., Brucke T., Aigner T. G. and Cohen R. M. (1991) New rapid analysis method demonstrates differences in 6-[‘sF]fluoro-L-dopa plasma input curves with and without carbidopa and in hemiMPTP lesioned monkeys. Appl. R&z. Isot. 42, 847-854. Melega W. P., Grafton S. T., Huang S.-C., Satyamurthy N., Phelps M. E. and Barrio J. R. (1991) ~-6-[‘~F]fl uoro-DOPA metabolism in monkeys and h “mans: Biochemical parameters for the formation of tracer kinetic models with positron emission tomography. J. Cereb. Blood Flow Metnb. 11, 890897. Murari D., Dejesus 0. T., Sunderland J. J. and Nickles R. J. (1992) [‘8F]Fluoro-~-fluoromethylene-m-tyrosine analogs, potential PET agents for presynaptic dopamine terminals: Synthesis and spectroscopic characterization. Appl. Radiar. Isot. 43, 969-977. Namavari M., Satyamurthy N., Phelps M. E. and Barrio J. R. (1993) Synthesis of 6-[18F] and 4-[‘8F]fluoro-L-m-tyrosine via regioselective radiofluorodestannylation. Appl. R&at. Isot. 44, 527-536. Sawle G. V., Bum D. J., Morrish P. K., Lammertsma A. A., Snow B. J,, Luthra S., Osman S. and Brooks D. J. (1994) The effect of entecapone (OR-61 1) on brain [‘8F]-6-L-fluorodopa metabolism: Implications for levedopa therapy of Parkinson’s disease. Neurology 44, 1292-1287. Tedroff J., Aquilonius S.-M., Laihinen A., Rinne U., Hartvig P., Andersson J., Lundqvist H., Haaparanta M., Solin O., Antoni G., Gee A. D., Ulin J. and LBngstrlim B. (1990) Striatal kinetics of [“Cl-(+)-nomifensine and 6-[18F]fluoro-L-dopa in Parkinson’s disease measured with positron emission tomography. Acta Neural. SC&. 81, 24-30. Tye S. J., Rupniak N. M. J., Naruse T., Miyaji M. and Iversen S. D. (1989) NB-355: A novel prodrug for L-dopa with reduced risk for peak-dose dyskinesias in MPTP-treated squirrel monkeys. Clin. Neuropharmacol 12, 393403. Zur M. and Nemerson Y. (1978) The esterase activity of coagulation factor VII. Evidence for intrinsic activitv of the zvmoeen. I v -J. Biol. Chem. 253. 2203-2209. Ztircher G., Keller H. H., Kettler R., Borgulya J., Bonetti E. P., Eigenman R. and Da Parada M. (1990) RO 40-7592, a novel, verv _ notent and orallv active inhibitor of catechol-O-methyltransferase: A pharmacological study in rats. Adu. Neural. 53, 497-503.