Synthesis of S-([18F]fluoromethyl)-(+)-McN5652 as a potential PET radioligand for the serotonin transporter

Nuclear Medicine and Biology 28 (2001) 857– 863

Synthesis of S-([18F]fluoromethyl)-(⫹)-McN5652 as a potential PET radioligand for the serotonin transporter Jo¨rg Zessina,*, Olli Eskolab, Peter Brusta, Jo¨rgen Bergmanb, Jo¨rg Steinbacha, Pertti Lehikoinenb, Olof Solinb, Bernd Johannsena a

Institut fu¨r Bioanorganische und Radiopharmazeutische Chemie, Forschungszentrum Rossendorf, Postfach 510119, 01314 Dresden, Germany b Turku PET Centre, Radiochemistry Laboratory, Porthaninkatu 3–5, FIN-20500 Turku, Finland Received 30 December 2000; received in revised form 7 April 2001; accepted 15 May 2001

Abstract The present study describes the synthesis of the [18F]fluoromethyl analogue of (⫹)-McN5652 ([18F]FMe-McN) as a new potential tracer for the serotonin transporter. In vitro binding studies have shown that FMe-McN displays only slightly lower affinity for the serotonin transporter (Ki ⫽ 2.3 ⫾ 0.1 nM) than (⫹)-McN5652 (Ki ⫽ 0.72 ⫾ 0.2 nM). The radiofluorinated tracer [18F]FMe-McN was prepared by reaction of normethyl (⫹)-McN5652 with the fluoromethylation agent [18F]bromofluoromethane in an overall radiochemical yield of 5 ⫾ 1% (decay-corrected, related to [18F]fluoride) and with high specific radioactivity (200 –2,000 GBq/␮mol at the end of synthesis). © 2001 Elsevier Science Inc. All rights reserved. Keywords: (⫹)-McN5652; [18F]fluoromethyl analogue; [18F]bromofluoromethane; Serotonin transporter; Positron emission tomography

1. Introduction The serotonin transporter (serotonin reuptake sites) located on the presynaptic nerve terminals controls the serotonin (5-HT) concentration in the synaptic cleft. Dysfunction of the 5-HT transporter is involved in psychiatric disorders such as depression, anxiety, and suicide [2,11,13, 23]. Noninvasive investigations of the 5-HT transporter density in the living brain with positron emission tomography (PET) is an important tool for the understanding of the pathophysiology of these disorders provided that a suitable radiotracer is available. Several ligands with high affinity and specificity for the 5-HT transporter have been labeled with 11C or 18F such as fluoxetine [5,9], sertraline [10], citalopram [8], and paroxetine derivatives [19,22]. Most of these radioligands have shown in vivo a low ratio between specific binding to the 5-HT transporter and unspecific binding which prevents the use for imaging the density of the 5-HT reuptake sites in humans [3]. The cocaine derivative [11C]RTI-357 [6] and the phenyl-

* Corresponding author. Tel.: ⫹351-260-2807; fax: ⫹351-260-2915. E-mail address: [email protected] (J. Zessin).

thiobenzylamines [11C]DAPP and [11C]DASB [7] are results of the continued search for alternative radioligands. These radiotracers have shown moderate target-to-nontarget ratios in primates and humans, but the biological evaluation is not completely yet. Only (⫹)-[11C]McN5652 (Fig. 1) has recently been demonstrated to be suitable for clinical use by studies in animals and humans. This radiotracer labels in vivo the 5-HT transporter in the brain of mice with high target-to-nontarget ratios and high selectivity [18]. A satisfactory ratio between specific (midbrain) and nonspecific (cerebellum) binding of the radioligand was reached in the human brain after 115 min [20]. By this time the total amount of radioactivity in the brain is quite low due to the short half-life of 11C (20.4 min). Suehiro and co-workers [16] therefore made an effort to label the compound (⫹)-McN5652 with fluorine-18 (halflife 110 min) by preparation of the [18F]fluoroethyl analogue [18F]1 (Fig. 2). Replacement of the original S-methyl group by the S-fluoroethyl moiety led to a considerable decrease in affinity and selectivity towards the 5-HT transporter. The penetration of the [18F]fluoroethyl analogue into the brain of mice and the target-to-nontarget ratios are lower in comparison with (⫹)-[11C]McN5652. These results sug-

0969-8051/01/$ – see front matter © 2001 Elsevier Science Inc. All rights reserved. PII: S 0 9 6 9 - 8 0 5 1 ( 0 1 ) 0 0 2 4 8 - 7

858

J. Zessin et al. / Nuclear Medicine and Biology 28 (2001) 857– 863

Fig. 1. Structure of (⫹)-McN5652 (trans-(⫹)-1,2,3,5,6,10b-hexahydro-6[4-([11C]methylthio)-phenyl]pyrrolo-[2,1-a]-isoquinoline).

gest that such a modification of the S-methyl group has a considerable effect on the affinity to the binding site. Alternatively, a [18F]fluoromethyl analogue of (⫹)McN5652 (Fig. 2) can be synthesized directly by the reaction of demethylated (⫹)-McN5625 with [18F]bromofluoromethane. This [18F]fluoromethylation reagent was introduced in the eighties by Coenen and co-workers [4]. The structure of the resulting S-[18F]fluoromethyl analogue ([18F]FMe-McN, [18F]2) is close to the original structure of (⫹)-McN5652. The affinity of this new radioligand to the 5-HT transporter therefore should be similar to that of (⫹)-McN5652. In this paper we describe the preparation of FMe-McN, its preliminary biological evaluation as well as the synthesis of [18F]FMe-McN ([18F]2).

2. Materials and methods 2.1. General All reagents and solvents were of analytical or HPLC grade and were used without further purification. (⫹)-

Fig. 2. Structure of

18

F-fluorinated McN5652 analogues.

McN5652 was prepared starting from 2-phenylpyrrolidine and 4-methylthiomandelic acid as described in an earlier report [24]. S-Demethylation of (⫹)-McN5652 was performed by treatment with sodium amide in liquid ammonia [24] to yield the (⫹)-thiol 3. This intermediate was reacted with acetyl chloride to obtain the related thioester precursor (⫹)-4 (see Fig. 3) [17]. NMR spectra were recorded on a Varian INOVA 400 spectrometer. 1H and 13C NMR spectra were obtained, using CDCl3 as a solvent and internal standard. The chemical shifts are expressed in ppm and related to tetramethylsilane. Mass spectrometric analyses were performed on a Perkin Elmer mass spectrometer PE SCIEX API 150 EX, using the positive ion mode with a turbo ion spray as an ion source. Samples were introduced via a Merck Purospher RP 18e (3␮m, 30 x 2 mm) column using 0.1% ammonium formate(aq.)/methanol (10/90, v/v) as eluent with a flowrate of 0.3 mL/min (Merck Hitachi L-6200 pump). High pressure liquid chromatography (HPLC) was performed using LaChrom systems from Merck-Hitachi. The chemical and radiochemical purity was determined with a Purospher RP18 column (Merck, 125 x 3 mm, 5 ␮m) isocratically eluted with water/acetonitrile (50/50) containing 0.1 M ammonium formate at a flow rate of 0.5 mL/min. Semipreparative HPLC purification was carried out with a Kromasil RP18 column (7 ␮m, 300 x 8 mm) isocratically eluted with water/acetonitrile (60/40) containing 0.1 M ammonium formate at a flow rate of 4 mL/min. Enantiomeric purity was determined with 1H NMR of the (⫹)-MTPA salts (R-(⫹)-␣-methoxy-␣-trifluoromethylphenylacetic acid, Mosher’s acid) as reported by Villani et al. [21] or by HPLC using a CHIROBIOTIC T column (Astec, 250 mm x 4.6 mm, 5 ␮m) isocratically eluted with methanol/acetic acid/triethylamine (1000/1/0.5) at a flow rate of 1 mL/min. Radio thin layer chromatography (radio TLC) was performed using Merck silica gel 60 F254 plates eluted with ethyl acetate/hexane/methanol (5/5/1). The compounds were detected using a radio TLC scanner [15]. 2.2. Synthesis of (⫹)-trans-1,2,3,5,6,10b-hexayhdro-6-[4(fluoromethylthio)phenyl]pyrrolo-[2,1-a]isoquinoline (FMe-McN, 2) The thioester (⫹)-4 (100 mg, 0.34 mmol) was hydrolyzed by treatment with a 1 M solution of tetrabutylammonium hydroxide (TBAH) in methanol (600 ␮L, 0.6 mmol) at room temperature. The reaction was completed within 10 min. The solution was diluted with dimethyl formamide (4 mL), placed in a sealed reaction vessel and cooled down with methanol/dry ice. Bromofluoromethane (200 mg, 1.8 mmol) was trapped in this solution at -78 °C. The reaction mixture was allowed to warm up to room temperature. After stirring at ambient temperature for 5 min, water (10 mL) was added. The solution was extracted with dichloromethane (3 x 25 mL), the combined organic layers were washed

J. Zessin et al. / Nuclear Medicine and Biology 28 (2001) 857– 863

859

Fig. 3. Synthesis scheme of the thioester precursor 4, and FMe-McN 2.

with water (2 x 30 mL), dried with magnesium sulfate and evaporated in vacuum. The crude product was purified by flash chromatography (silica gel, petroleum ether/ethyl acetate/methanol 5/5/2.5). The purified (⫹)-FMe-McN (60 mg, 0.19 mmol) dissolved in acetonitrile (1 mL) was combined with (⫹)-O-dip-toluoyltartaric acid in acetonitrile (1.5 mL). After standing over night, off-white needles were filtered and washed with cold acetonitrile (5 mL) to give the (⫹)-FMe-McN(⫹)-O-di-p-toluoyltartrate (70 mg, 30%). 1 H-NMR (free amine): 1.78 –1.91 (m, 2H); 1.92–2.04 (m, 1H); 2.36 –2.40 (1H), 2.54 –2.61 (m, 1H); 2.86 –2.90 (m, 1H), 2.95–3.00 (m, 1H); 3.02–3.06 (m, 1H); 3.46 –3.51 (m, 1H), 4.18 (t, JHH ⫽ 5 Hz, 1H); 5.71 (d, JHF ⫽ 53 Hz, 2H); 6.89 (d, JHH ⫽ 7.7 Hz, 1H); 7.07–7.11 (m, 1H); 7.15–7.18 (m, 2H); 7.21–7.25 (m, 2H); 7.39 –7.41 (m, 2H) 13 C-NMR (free amine): 22.17; 30.24; 45.71; 54.08; 56.16; 63.81; 76.69; 88.73 (JCF ⫽ 216 Hz); 125.71; 126.15; 126.18; 129.39; 129.74; 130.72; 136.85; 139.02; 146.43

After end of bombardment (EOB) the irradiated target water was transferred to the reaction vessel containing the phase transfer catalyst Kryptofix 2.2.2. (48 ␮mol, 18 mg), potassium carbonate (36 ␮mol, 5 mg) and acetonitrile (1 mL). The solvents were evaporated at 110 °C under vacuum with helium flow. Two further portions of acetonitrile (1 mL) were added during the evaporation. The reaction vessel was allowed to cool down for one minute. Dibromomethane 5 (50 ␮L, 700 ␮mol) in acetonitrile (1 mL) was then added to the dry kryptofix/[18F]F- mixture. The reaction mixture was heated again to 110 °C and the volatile products were transferred to the preparative GC column (Porapak Q, 80 –100 mesh, 300 mm x 7.8 mm) with helium as carrier gas. The GC column was heated to 100 °C. The gaschromatographic separation of [18F]bromofluoromethane was monitored with a radioactivity detector. The [18F]bromofluoromethane fraction was collected at room temperature in a second reaction vessel containing the reagents for the synthesis of [18F]FMe-McN.

2.3. Synthesis of [18F]bromofluoromethane [18F]Fluoride was produced with a MGC-20 cyclotron (Jefremov-Institute, St. Petersburg, Russia) using the 18 O(p,n)[18F] nuclear reaction. The synthesis of [18F]bromofluoromethane 6 (see Fig. 4) was carried out in an automated synthesis apparatus [1].

Fig. 4. Synthesis of [18F]bromofluoromethane.

860

J. Zessin et al. / Nuclear Medicine and Biology 28 (2001) 857– 863

2.4. Synthesis of (⫹)-trans-1,2,3,5,6,10b-hexayhdro-6[4-(18F]fluoromethylthio)phenyl]-pyrrolo-[2,1-a]isoquinoline ([18F]FMe-McN, [18F]2) The thioester precursor (⫹)-4 (0.5 mg, 1.5 ␮mol) was treated with TBAH (0.3 M in methanol, 60 ␮L, 18 ␮mol). The hydrolysis was completed after 10 min at room temperature. The reaction mixture was then diluted with dimethyl formamide (200 ␮L). [18F]Bromofluoromethane was trapped in this solution at room temperature. After trapping of [18F]bromofluoromethane, the reaction vessel was heated at 40 °C for 1 min to perform the [18F]fluoromethylationreaction. The reaction mixture was diluted with water/acetonitrile (50/50, 400 ␮L). [18F]FMe-McN was separated by semipreparative HPLC using the system described above. The retention time of [18F]FMe-McN was 16.9 min. The fraction containing [18F]FMe-McN was diluted with water (15 mL). The resulting mixture was pressed through a Chromafix RP18 cartridge (200 mg, preconditioned with 5 mL ethanol and 10 mL water). The cartridge was washed with water (5 mL) and the product was eluted with absolute ethanol (1 mL). 2.5. Radioligand binding assays Radioligand binding assays were performed on brain membranes from porcine brain. Samples of the caudate nucleus were obtained from a local slaughterhouse. For membrane preparation the brain tissue was homogenized in 10 volumes of ice-cold buffer (Tris-HCl buffer, pH 7.4, containing 120 mM NaCl, 5 mM KCl) with an Ultra-Turrax T25. The homogenate was centrifuged at 20,000 g for 10 min. The resulting pellet was resuspended with the UltraTurrax and centrifuged again at 20,000 g for 10 min. After repeating the same procedure the pellet was resuspended in 10 volumes of buffer and stored at -20 °C until use in the binding studies. All samples were re-homogenized before use. Radioligand binding assays to the 5-HT transporter were performed using [3H]paroxetine (17.1 Ci/mmol, Du Pont, Dreieich, Germany) as radioligand. Brain membrane preparations were incubated in a total volume of 5.0 mL for 60 min at 20 °C in buffer (50 mM TRIS-HCl, 100 mM NaCl, 5 mM KCl, pH 7.4) containing the unlabeled ligands. Nonspecific binding was determined in the presence of 10 ␮M clomipramine (Research Biochemicals International, Natick, USA). The binding assays were terminated by rapid filtration through GF/B glass fiber filters (Whatman, Maidstone, England) on a 30 place Cell Harvester (Brandel, Gaithersburg, MD, USA). The filters were presoaked in 0.3% polyethylene imine. Following four washes with 4 mL portions of ice-cold buffer (see above), the filter paper containing the membrane-bound [3H]paroxetine was transferred into 10 mL of liquid scintillation cocktail (UltimaGold, Packard), and the radioactivity on each filter was counted in a Canberra-Packard liquid scintillation counter

(TRI-CARB 2100TR). Aliquots of the incubation fluid were also measured. Corrections were made for the binding of [3H]paroxetine to the filters. Increasing concentrations (between 0.01 nM and 1 ␮M) of the unlabeled ligands were competed against a fixed concentration of the radiolabelled ligand (total binding ⬃ 1500 dpm). From the competition curves IC50 values were calculated using the program Fig. P (Biosoft). From these data Ki values were calculated using the following equation: Ki ⫽ C50/(1 ⫹ [L]/Kd), where [L] is the free tracer concentration. 2.6. Stability of FMe-McN and [18F]FMe-McN The structure of the product resulting from the decomposition of FMe-McN was determined by mass spectroscopy. For that purpose, (⫹)-FMe-McN (1 mg) was dissolved in a methanol/water mixture (20/80) containing 0.01 M ammonium formate. This solution was kept at room temperature. Samples of this solution were analyzed with the LC-MS system after 0, 15, and 38 min. The stability of [18F]FMe-McN was investigated in ethanol, propylene glycol, sodium bicarbonate solution (0.11 M), and mixtures of ethanol or propylene glycol and sodium bicarbonate solution (0.11 M). The ethanolic solution of [18F]FMe-McN resulted from solid phase extraction after synthesis and purification of [18F]FMe-McN. Parts of this ethanolic solution were diluted with various volumes of sodium bicarbonate solution. The propylene glycol solution of [18F]FMe-McN was prepared by evaporating the ethanolic solution almost to dryness. The residue was dissolved in propylene glycol and diluted with the appropriate amount of a sodium bicarbonate solution. The solution of [18F]FMe-McN in sodium bicarbonate solution (0.11 M) was prepared in a similar manner. The decomposition was monitored by radio TLC (silica gel 60254 plates with hexane/ethyl acetate/methanol 5:5:1 as mobile phase).

3. Results and discussion 3.1. Chemistry The synthesis of FMe-McN 2, the fluoromethyl analogue of (⫹)-McN5652, is depicted in Fig. 3. The starting compound (⫹)-McN5652 was prepared from the commercially available N-vinyl-2-pyrrolidinone and 4-methylthiobenzaldehyde according to [24]. In order to prepare the precursor 4 for the synthesis of nonradioactive FMe-McN, (⫹)McN5652 was S-demethylated by treatment with sodium amide in liquid ammonia as described in an earlier report [24]. The thiol function of compound 3 was protected as thioester by reaction with acetyl chloride to yield the more stable thioester 4 [17]. This thioester precursor 4 was hydrolyzed by treatment with TBAH. The intermediate thiolate 3 was reacted with

J. Zessin et al. / Nuclear Medicine and Biology 28 (2001) 857– 863 Table 1 Inhibition of [3H]paroxetine binding to membranes from porcine caudate nucleus by FMe-McN, (⫹)-McN5652 and other inhibitors Inhibitor

Ki (nM)

Paroxetine Clomipramine (⫹)-McN5652 Fluoxetine FMe-McN Imipramine Citalopram Venlafaxin Protriptyline Desipramine Nisoxetine Serotonin

0.19 ⫾ 0.05 0.19 ⫾ 0.04 0.72 ⫾ 0.20 0.89 ⫾ 0.12 2.30 ⫾ 0.05 2.90 ⫾ 0.49 6.61 ⫾ 0.64 35.0 ⫾ 5.0 35.9 ⫾ 7.1 36.1 ⫾ 3.1 57.9 ⫾ 14.0 592 ⫾ 98

bromofluoromethane to afford (⫹)-FMe-McN 2 in sufficient yields. The structure of the product was confirmed by mass spectrometric and NMR spectroscopic analyses. The 1 H NMR spectrum of the product is characterized by a doublet for the fluoromethyl moiety at 5.7 ppm with a H-F coupling constant of 53.1 Hz. For the fluoromethyl group, the 13C NMR spectrum shows a doublet at 88.8 ppm with a C-F coupling constant of 216 Hz. 3.2. In vitro binding studies FMe-McN was used for in vitro binding studies to determine its affinity to the 5-HT transporter in comparison with (⫹)-McN5652 and other ligands. The 5-HT transporter assay used tissue homogenates prepared from porcine caudate nucleus and [3H]paroxetine as radioligand. The results are shown in Table 1. The replacement of hydrogen by fluorine at the S-methyl group of McN5652 slightly reduces the affinity to the 5-HT transporter. However, it is still better by a factor 3 than the affinity of citalopram, the most selective 5-HT uptake inhibitor known so far. We therefore expect [18F]FMe-McN to be suitable for in vivo imaging of the 5-HT transporter. 3.3. Radiochemistry [18F]Bromofluoromethane 6 was synthesized in a remote-controlled apparatus from dibromomethane 5 by nucleophilic substitution of bromine with [18F]fluoride (Fig. 4) as previously reported [1]. After drying of the [18F]fluoride/ kryptofix 2.2.2. complex the reaction with dibromomethane was carried out in acetonitrile at 110 °C while purging with helium. [18F]Bromofluoromethane was separated from unreacted dibromomethane, acetonitrile, and side-products by gas chromatography using Porapak Q as stationary phase. The retention time of the product 6 was about 10 min. The trapping of 6 was usually completed within one minute. The [18F]fluoromethylation agent 6 was obtained with a decay-

861

corrected radiochemical yield of 25% (from [18F]fluoride) and a radiochemical purity ⬎99%. Demethylated (⫹)-McN5652 3 was reacted with [18F]bromofluoromethane to yield [18F]FMe-McN. The product mixture containing [18F]FMe-McN, unreacted [18F]bromofluoromethane and polar side products was purified by reversed phase HPLC. The desired product [18F]2 was eluted at about 16.9 min. The radiochemical purity of [18F]FMe-McN was ⬎95%. Chiral HPLC analysis of the product showed optically pure (⫹)-[18F]FMe-McN (enantiomeric excess (ee) ⬎ 98%). The decay-corrected radiochemical yield of [18F]FMe-McN, calculated from end of bombardment and based on starting [18F]fluoride, was 5 ⫾ 1% (n ⫽ 12). The total synthesis time was approximately 65 min. The specific radioactivity was 200 –2,000 GBq/␮mol (5.4 –54 Ci/␮mol) at the end of the synthesis. The decay-corrected radiochemical yield (related to [18F]fluoride) of the [18F]fluoromethylation step varied from 5–9%. The radiochemical yield was not dependent on the reaction temperature in the range from 25 to 60 °C. Extension of the reaction time to up to 10 min did not significantly change the radiochemical yield of [18F]FMe-McN. The incomplete conversion of [18F]bromofluoromethane is an indication of the lower reactivity of this agent in comparison with [11C]methyl iodide. Under similar reaction conditions [11C]methyl iodide completely reacted with thiol 3 to yield [11C]McN5652. 3.4. Stability of FMe-McN and [18F]FMe-McN Special consideration was given to the stability of FMeMcN and [18F]FMe-McN since hydrolytic instabilities have been reported for several [18F]fluoromethyl compounds [4,14]. Therefore we studied the stability of FMe-McN and [18F]FMe-McN in various solvents. The rate of hydrolysis of FMe-McN depends largely on the pH value. FMe-McN was decomposed completely at pH 1 (hydrochloric acid) within 5 min. About 50% of FMeMcN was hydrolyzed after 60 min at pH 4 (acetic acid). Only 1.5% of FMe-McN undergoes solvolysis at pH 8 (sodium bicarbonate) after 60 min. Solvolysis of nonradioactive FMe-McN in a water/methanol mixture at pH 7 yielded a product with a molecular mass of 281 g/mol (detected by LC-MS). This result indicated that the sulfur-fluoromethyl bond was cleaved. Normethyl McN5652 3 was formed as product of the decomposition reaction. The stability of [18F]FMe-McN was analyzed in various solvents and solvent mixtures (Fig. 5, Table 2). In ethanol or propylene glycol no decomposition of [18F]FMe-McN was observed within 200 min. In a sodium bicarbonate solution (0.11 M) containing 50% ethanol, a share of 2% [18F]FMcN was converted within 120 min. The rate of decomposition increased with shares of ethanol lower than 50%. In mixtures of propylene glycol and sodium bicarbonate solution, [18F]FMe-McN displayed a slightly lower stabil-

862

J. Zessin et al. / Nuclear Medicine and Biology 28 (2001) 857– 863

Fig. 5. Stability of [18F]FMe-McN in different solvents and solvent mixtures 1) in a 0.11 M sodium bicarbonate solution.

ity. A proportion of 5% [18F]FMe-McN was converted after 60 min in a mixture of propylene glycol and 0.11 M sodium bicarbonate solution (50/50). Solutions with this amount of propylene glycol are acceptable for intravenous injection. The observations suggest that [18F]FMe-McN dissolved in this solvent mixture possesses a satisfactory stability for processing the injection solution. Preliminary ex vivo autoradiographic studies in rats have demonstrated that [18F]FMe-McN has a sufficient stability for this kind of investigation by use of an injection solution consisting of propylene glycol and 0.11 M sodium bicarbonate solution (50/50). Under these conditions, [18F]FMeMcN showed accumulation in brain regions with a known high density of the 5-HT transporter such as raphe nuclei, substantia nigra, thalamus and amygdala. In bones (skull) a Table 2 Chemical stability of [18F]FMe-McN in various solvents and solvent mixtures Solvent propylene glycol ethanol/0.11 M NaHCO3 propylene glycol/0.11 M NaHCO3 ethanol/0.11 M NaHCO3 propylene glycol/0.11 M NaHCO3 ethanol/0.11 M NaHCO3 0.11 M NaHCO3

Solvent composition

Time

Share of [18F]FMe-McN

— 50:50 50:50

200min 120min 120min

96% 95% 86%

25:75 25:75

30min 30min

80% 76%

10:90 —

30min 10min

55% 55%

low uptake of the radiotracer was observed. A maximum ratio of about 8 between specific (raphe nuclei) and nonspecific binding (cerebellum) was reached after 150 min [12].

4. Conclusion The [18F]fluoromethyl analogue of (⫹)-McN5652 was synthesized as a potential radiotracer for imaging the serotonin transporter. Satisfactory radiochemical yields were achieved by [18F]fluoromethylation of demethylated (⫹)McN5652 with [18F]bromofluoromethane. The radiotracer [18F]FMe-McN is stable for 1 hour in a 0.11 M sodium bicarbonate solution containing 50% propylene glycol.

Acknowledgments We would like to thank the cyclotron staff of the Turku PET Center for providing [18F]fluoride. Thanks are also due to R. Herrlich, H. Kasper, and S. Lehnert for technical assistance.

References [1] J. Bergman, O. Eskola, P. Lehikoinen, O. Solin, Automated synthesis and purification of [18F]bromofluoromethane at high specific radioactivity, Appl. Radiat. Isot. 54 (2001) 927–933.

J. Zessin et al. / Nuclear Medicine and Biology 28 (2001) 857– 863 [2] B. Bondy, A. Erfurth, S. Jonge, M. Kruger H. Meyer, Possible association of the short allele of the serotonin transporter promoter gene polymorphism (5-HTTLPR) with violent suicide, Mol. Psychiatry. 5 (2) (2000) 193–195. [3] P. Brust, U. Scheffel, Z. Szabo, Radioligands for the study of the 5-HT transporter in vivo, Idrugs. 2 (1999) 129 –145. [4] H.H. Coenen, M. Colosimo, M. Schu¨ ller, G. Sto¨ cklin, Preparation of n.c.a. [18F]CH2BrF via aminopolyether supported nucleophilic substitution, J. Labelled Compd. Radiopharm. 23 (1986) 587–595. [5] M.K. Das, J. Mukherjee, Radiosynthesis of [F-18]Fluoxetine as a potential radiotracer for serotonin reuptake sites, Appl. Radiat. Isot. 44 (1993) 835– 842. [6] J. Helfenbein, J. Sandell, C. Halldin, S. Chalon, P. Emond, Y. Okubo, Y.-H. Chou, Y. Frangin, L. Douziech, L. Gareau, C.-G. Swahn, J.-C. Besnard, L. Farde, D. Guilloteau, PET examination of three potent cocaine derivatives as specific radioligands for the serotonin transporter, Nucl. Med. Biol. 26 (1999) 491– 499. [7] S. Houle, N. Ginovart, D. Hussey, J.H. Meyer, A.A. Wilson, Imaging the serotonin transporter with positron emission tomography: initial human studies with [11C]DAPP and [11C]DASB, Eur. J. Nucl. Med. 27 (2000) 1719 –1722. [8] S.P. Hume, C. Pascali, V.W. Pike, D.R. Turton, R.G. Ahier, R. Myers, D.M. Bateman, J.E. Cremer, L.G. Manjil, R. Dolan, Citalopram: labelling with carbon-11 and evaluation in rat as a potential radioligand for in vivo studies of 5-HT re-uptake sites. Int. J. Rad. Appl. Instrum. B 18(3) (1991) 339 –351. [9] M.R. Kilbourn, M.S. Haka, G. Mulholland, D.M. Jewett, D.E. Kuhl, Synthesis of radiolabelled inhibitors of presynaptic monamine uptake systems: fluorine-18 GBR-13119 DA, carbon-11 nisoxetine NE and carbon-11 fluoxetine 5-HT, J. Labelled Compd. Radiopharm. 26 (1989) 412– 414. [10] M.C. Lasne, V.W. Pike, D.R. Turton, The radiosynthesis of [N-methyl-11C]-sertraline, Int. J. Rad. Appl. Instrum. [A] 40(2) (1989) 147– 151. [11] M. Leboyer, P. Quintin, P. Manivet, O. Varoquaux, J.F. Allilaire, J.M. Launay, Decreased serotonin transporter binding in unaffected relatives of manic depressive patients. Biol. Psychiatry. 46(12) (1999) 1703–1706. [12] P. Marjama¨ ki, J. Zessin, O. Eskola, J. Bergman, T. Gronroos, M. Haaparanta-Solin, P. Lehikoinen, J. Steinbach, O. Solin, Biological evaluation in rat of [S-fluoromethyl-18F]fluoro-McN5652, a novel PET-radiotracer for serotonin uptake site, Eur. J. Nucl. Med. 27 (2000) 1222. [13] M.A. Menza, B. Palermo, R. DiPaola, J.I. Sage, M.H. Ricketts, Depression and anxiety in Parkinson’s disease: possible effect of genetic variation in the serotonin transporter. J. Geriatr. Psychiatry. Neurol. 12(2) (1999) 49 –52.

863

[14] A. Petric, J.R. Barrio, M. Namavari, S.-C. Huang, N. Sathyamurthy, Synthesis of 3␤-(4-[18F]fluoromethylphenyl)- and 3␤-(2-[18F]fluoromethylphenyl)tropane-2␤-carboxylic acid methyl esters: New ligands for mapping brain dopamine transporter with positron emission tomography, Nucl. Med. Biol. 26 (1999) 529 –535. [15] O. Solin, Counting of Positron- Emitting Radionuclides on Thin Layer Chromatograms, Int. J. Appl. Radiat. Isot. 34 (1983) 1653– 1654. [16] M. Suehiro, J.H. Greenberg, C.Y. Shiue, C. Gonzalez, B. Dembowski, M. Reivich, Radiosynthesis and biodistribution of the S-[18F]fluoroethyl analog of McN5652, Nucl. Med. Biol. 23 (1996) 407– 412. [17] M. Suehiro, J.L. Musachio, R.F. Dannals, W.B. Mathews, H.T. Ravert, U. Scheffel, H.N. Wagner, Jr, An improved method for the synthesis of radiolabeled McN5652 via thioester precursors, Nucl. Med. Biol. 22 (1995) 543–545. [18] M. Suehiro, U. Scheffel, R.F. Dannals, H.T. Ravert, G.A. Ricaurte, H.N. Wagner, Jr, A PET radiotracer for studying serotonin uptake sites: carbon-11-McN-5652Z. J. Nucl. Med. 34(1) (1993) 120 –127. [19] M. Suehiro, A.A. Wilson, U. Scheffel, R.F. Dannals, H.T. Ravert, H.N. Wagner Jr, Radiosynthesis and evaluation of N-(3-[18F]fluoropropyl)paroxetine as a radiotracer for in vivo labeling of serotonin uptake sites by PET. J. Rad. Appl. Instrum. B. 18(7) (1991) 791–796. [20] Z. Szabo, P.F. Kao, U. Scheffel, M. Suehiro, W.B. Mathews, H.T. Ravert, J.L. Musachio, S. Marenco, S.E. Kim, G.A. Ricaurte, D.F. Wong, Wagner Jr., R.F. Dannals, Positron emission tomography imaging of serotonin transporters in the human brain using [11C](⫹)McN5652, Synapse 20 (1995) 37– 43. [21] F.J. Villani, M.J. Costanzo, R.R. Inners, M.S. Mutteler, D.E. McClure, Determination of enantiomeric purity of tertiary amines by 1H NMR of ␣-methoxy-␣-(trifluoromethyl)phenylacetic acid complexes, J. Org. Chem. 51 (1986) 3715–3718. [22] V.L. Villemagne, R.F. Dannals, P.M. Sa´ nchez-Roa, U. Scheffel, H.T. Ravert, A.A. Wilson, T.K. Natarajan, J.J. Frost, H.N. Wagner Jr, PET imaging of central serotonin uptake sites in the baboon brain with [C-11]-N-Me-Paroxetine, J. Nucl. Med. 30 (1989) 883. [23] M. Willeit, N. Praschak-Rieder, A. Neumeister, W. Pirker, S. Asenbaum, O. Vitouch, J. Tauscher, E. Hilger, J. Stastny, T. Brucke, S. Kasper, [123I]-beta-CIT SPECT imaging shows reduced brain serotonin transporter availability in drug-free depressed patients with seasonal affective disorder. Biol. Psychiatry. 47(6) (2000) 482– 489. [24] J. Zessin, P. Gucker, S.M. Ametamey, J. Steinbach, P. Brust P, F.X. Vollenweider, B. Johannsen, P.A. Schubiger, Efficient synthesis of enantiomerically pure thioester precursors of [11C]McN-5652 from racemic McN-5652, J. Labelled Compd. Radiopharm. 42 (1999) 1301–1312.