Asymmetric Synthesis of Spirocyclic 2-Benzopyrans for Positron Emission Tomography of σ1 Receptors in the Brain

Sharpless asymmetric dihydroxylation of styrene derivative 6 afforded chiral triols (R)-7 and (S)-7, which were cyclized with tosyl chloride in the presence of Bu2SnO to provide 2-benzopyrans (R)-4 and (S)-4 with high regioselectivity. The additional hydroxy moiety in the 4-position was exploited for the introduction of various substituents. Williamson ether synthesis and replacement of the Boc protective group with a benzyl moiety led to potent σ1 ligands with high σ1/σ2-selectivity. With exception of the ethoxy derivative 16, the (R)-configured enantiomers represent eutomers with eudismic ratios of up to 29 for the ester (R)-18. The methyl ether (R)-15 represents the most potent σ1 ligand of this series of compounds, with a Ki value of 1.2 nM and an eudismic ratio of 7. Tosylate (R)-21 was used as precursor for the radiosynthesis of [18F]-(R)-20, which was available by nucleophilic substitution with K[18F]F K222 carbonate complex. The radiochemical yield of [18F]-(R)-20 was 18%–20%, the radiochemical purity greater than 97% and the specific radioactivity 175–300 GBq/µmol. Although radiometabolites were detected in plasma, urine and liver samples, radiometabolites were not found in brain samples. After 30 min, the uptake of the radiotracer in the brain was 3.4% of injected dose per gram of tissue and could be reduced by coadministration of the σ1 antagonist haloperidol. [18F]-(R)-20 was able to label those regions of the brain, which were reported to have high density of σ1 receptors.


Figure 1. Development of fluorinated PET tracers
The fluoroalkyl substituted 2-benzofurans 2a-d (n = 1-4) were derived from the spirocyclic 2-benzofuran 1; the potent σ 1 antagonist 3 represents the lead for the enantiomerically pure 2-benzopyrans 4. Although the 2-benzopyran 3 (K i = 1.3 nM) represents a very potent σ 1 receptor antagonist [35], enantiomers of 2-benzopyran based σ 1 ligands were not yet investigated. Due to their structural similarity to the spirocyclic 3-substituted 2-benzopyrans 3 [36] and 2-benzofurans 1 and 2, 4-substituted 2-benzopyrans 4 were considered as new type of σ 1 receptor ligands. Moreover, the 2-benzopyran scaffold was not exploited for the development of a fluorinated PET tracer so far. In this communication we report the first enantioselective synthesis of 4-substituted spirocyclic 2-benzopyrans of type 4, their affinity towards σ receptors and the generation and biological evaluation of a [ 18 F]-labeled PET tracer based on this scaffold.

tert-Butyl
(R)-10 (210 mg, 0.66 mmol) was dissolved in THF (15 mL). NaH (60% dispersion in paraffin liquid, 80 mg, 2.0 mmol) was added and the mixture was stirred for 1 h at ambient temperature. Then iodoethane (0.53 mL, 6.6 mmol) was added dropwise and the mixture was stirred for 2.5 h at ambient temperature. A 1 m solution of lithium bis(trimethylsilyl)amide (4.5 mL) was added and the mixture was heated to reflux overnight. The mixture was allowed to cool to ambient temperature and stirred overnight. Then water and CH 2 Cl 2 were added. After separation of the layers, the aqueous layer was extracted with CH 2 Cl 2 (3×). The combined organic layers were dried (Na 2 SO 4 ), filtered and the solvent removed in vacuo. The crude product was purified by flash column chromatography (Ø = 2 cm, h = 15 cm, cyclohexane-ethyl acetate = 9:1, V = 10 mL) to give (R)-14 as a pale yellow oil (R f = 0.15, cyclo-hexane-ethyl acetate = 9:1), yield 66 mg (29%). C 20

tert-Butyl (S)-4-ethoxy-3,4-dihydrospiro[[2]benzopyran-1,4′-piperidine]-1′-carboxylate ((S)-14)
(S)-10 (200 mg, 0.63 mmol) was dissolved in THF (5 mL). A 1 m solution of lithium bis(trimethylsilyl)amide (6.3 mL) was added and the mixture was stirred for 1 h at ambient temperature. Then iodoethane (500 μL, 6.3 mmol) was added dropwise and the mixture was stirred for 16 h at ambient temperature. NaH (60% dispersion in paraffin liquid, 250 mg, 6.3 mmol) and iodoethane (500 μL, 6.3 mmol) were added and the mixture was heated to reflux overnight. The mixture was allowed to cool to ambient temperature and stirred for 3 days. Then water was added. After separation of the layers, the aqueous layer was extracted with ethyl acetate (3×). The combined organic layers were dried (Na 2 SO 4 ), filtered and the solvent was removed in vacuo. The (R)-13 (360 mg, 1.1 mmol) was dissolved in CH 2 Cl 2 (5 mL). The solution was cooled to 0 °C. Trifluoroacetic acid (0.7 mL) was added and the mixture was stirred for 2 h at 0 °C. Then a 2 M aqueous solution of NaOH was added. After separation of the layers, the aqueous layer was extracted with CH 2 Cl 2 (3×). The combined organic layers were dried (Na 2 SO 4 ), filtered and the solvent was removed in vacuo. The residue was dissolved in CH 2 Cl 2 (5 mL). Benzaldehyde (30 μL, 0.30 mmol) and sodium triacetoxyborohydride (76 mg, 0.36 mmol) were added and the mixture was stirred for 26 h at ambient temperature. Then a 2 M aqueous solution of NaOH (3 mL) and water (3 mL) were added. After separation of the layers, the aqueous layer was extracted with CH 2 Cl 2 (3×). The combined organic layers were dried (Na 2 SO 4 ), filtered and the solvent was removed in vacuo. The crude product was purified by flash column chromatography (Ø = 0.75 cm, h = 15 cm, cyclohexane-ethyl acetate = 4:1, V = 5 mL) to give (R)-15 as a colorless oil (R f = 0.27, cyclohexane-ethyl acetate = 5:5), yield 31 mg (9%). C 21  (S)-13 (50 mg, 0.15 mmol) was dissolved in CH 2 Cl 2 (5 mL). Trifluoroacetic acid (200 μL) was added and the mixture was stirred for 4.5 h at ambient temperature. Then a 2 M aqueous solution of NaOH was added. After separation of the layers, the aqueous layer was extracted with CH 2 Cl 2 (3×). The combined organic layers were dried (Na 2 SO 4 ), filtered and the solvent was removed in vacuo. The residue was dissolved in CH 2 Cl 2 (10 mL). Benzaldehyde (70 μL, 0.69 mmol) and sodium triacetoxyborohydride (96 mg, 0.45 mmol) were added and the mixture was stirred overnight at ambient temperature. The reaction was stopped by the addition of a 2 M aqueous solution of NaOH. After separation of the layers, the aqueous layer was extracted with CH 2 Cl 2 (3×). The combined organic layers were dried (Na 2 SO 4 ), filtered and the solvent was removed in vacuo. The (R)-14 (49 mg, 0.14 mmol) was dissolved in CH 2 Cl 2 (10 mL). Trifluoroacetic acid (200 μL) was added and the mixture was stirred for 3 h at ambient temperature. Then a 2 M aqueous solution of NaOH (10 mL) was added. After separation of the layers, the aqueous layer was extracted with CH 2 Cl 2 (3×). The combined organic layers were dried (Na 2 SO 4 ), filtered and the solvent was removed in vacuo. The residue was dissolved in CH 2 Cl 2 (10 mL). Benzaldehyde (60 μL, 0.59 mmol) and (after 45 min) sodium triacetoxyborohydride (181 mg, 0.85 mmol) were added and the mixture was stirred overnight at ambient temperature. The reaction was stopped by the addition of a 2 M aqueous solution of NaOH. After separation of the layers, the aqueous layer was extracted with CH 2 Cl 2 (2×) and ethyl acetate (1×). The combined organic layers were dried (Na 2 SO 4 ), filtered and the solvent was removed in vacuo. The crude product was purified by flash column chromatography twice (1. Ø = 1. (S)-14 (63 mg, 0.18 mmol) was dissolved in CH 2 Cl 2 (5 mL). Trifluoroacetic acid (300 μL) was added and the mixture was stirred for 2 h at ambient temperature. Then a 2 M aqueous solution of NaOH was added. After separation of the layers, the aqueous layer was extracted with ethyl acetate (3×). The combined organic layers were dried (Na 2 SO 4 ), filtered and the solvent was removed in vacuo. The residue was dissolved in CH 2 Cl 2 (5 mL). Benzaldehyde (40 μL, 0.39 mmol) and after 15 min, sodium triacetoxyborohydride (120 mg, 0.57 mmol) were added and the mixture was stirred at ambient temperature for 8 h. The reaction was stopped by the addition of a 2 M aqueous solution of NaOH. After separation of the layers, the aqueous layer was extracted with CH 2 Cl 2 (3×). The combined organic layers were dried (Na 2 SO 4 ), filtered and the solvent was removed in vacuo. The crude product was purified by flash column chromatography (Ø = 1.25 cm, h = 15 cm, cyclohexane-ethyl acetate = 6:1, V = 5 mL) to give (S)-16 as a pale yellow oil (R f = 0.31, cyclohexane-ethyl acetate = 5:5), yield 28 mg (46%). C 22

Materials
The guinea pig brain and rat liver for the σ 1 and σ 2 receptor binding assays were commercially available (Harlan-Winkelmann, Borchen, Germany).

Preparation of Membrane Homogenates from Guinea Pig Brain
According to [35][36][37]: five guinea pig brains were homogenized with the Potter (500-800 rpm, 10 up-and-down strokes) in 6 volumes of cold 0.32 M sucrose. The suspension was centrifuged at 1,200 × g for 10 min at 4 °C. The supernatant was separated and centrifuged at 23,500 × g for 20 min at 4 °C. The pellet was resuspended in 5-6 volumes of buffer (50 mM TRIS, pH 7.4) and centrifuged again at 23,500 × g (20 min, 4 °C). This procedure was repeated twice. The final pellet was resuspended in 5-6 volumes of buffer and frozen (−80 °C) in 1.5 mL portions containing about 1.5 mg protein/mL.

Preparation of Membrane Homogenates from Rat Liver
According to [35][36][37]: two rat livers were cut into small pieces and homogenized with the Potter (500-800 rpm, 10 up-and-down strokes) in 6 volumes of cold 0.32 M sucrose. The suspension was centrifuged at 1,200 × g for 10 min at 4 °C. The supernatant was separated and centrifuged at 31,000 × g for 20 min at 4 °C. The pellet was resuspended in 5-6 volumes of buffer (50 mM TRIS, pH 8.0) and incubated at room temperature for 30 min. After the incubation, the suspension was centrifuged again at 31,000 × g for 20 min at 4 °C. The final pellet was resuspended in 5-6 volumes of buffer and stored at −80 °C in 1.5 mL portions containing about 2 mg protein/mL.

Protein Determination
The protein concentration was determined by the method of Bradford [38], modified by Stoscheck [39]. The Bradford solution was prepared by dissolving 5 mg of Coomassie Brilliant Blue G 250 in 2.5 mL of EtOH (95%, v/v). 10 mL deionized H 2 O and 5 mL phosphoric acid (85%, m/v) were added to this solution, the mixture was stirred and filled to a total volume of 50.0 mL with deionized water. The calibration was carried out using bovine serum albumin as a standard in 9 concentrations (0.1, 0.2, 0.4, 0.6, 0.8, 1.0, 1.5, 2.0 and 4.0 mg/mL). In a 96-well standard multiplate, 10 µL of the calibration solution or 10 µL of the membrane receptor preparation were mixed with 190 µL of the Bradford solution, respectively. After 5 min, the UV absorption of the protein-dye complex at λ = 595 nm was measured with a platereader (Tecan Genios, Tecan, Crailsheim, Germany).

General Protocol for the Binding Assays
According to [35][36][37]: the test compound solutions were prepared by dissolving approximately 10 µmol (usually 2-4 mg) of test compound in DMSO so that a 10 mM stock solution was obtained.
To obtain the required test solutions for the assay, the DMSO stock solution was diluted with the respective assay buffer. The filtermats were presoaked in 0.5% aqueous polyethylenimine solution for 2 h at room temperature before use. All binding experiments were carried out in duplicates in 96-well multiplates. The concentrations given are the final concentrations in the assay. Generally, the assays were performed by addition of 50 µL of the respective assay buffer, 50 µL test compound solution in various concentrations (10 −5 , 10 −6 , 10 −7 , 10 −8 , 10 −9 and 10 −10 mol/L), 50 µL of corresponding radioligand solution and 50 µL of the respective receptor preparation into each well of the multiplate (total volume 200 µL). The receptor preparation was always added last. During the incubation, the multiplates were shaken at a speed of 500-600 rpm at the specified temperature. Unless otherwise noted, the assays were terminated after 120 min by rapid filtration using the harvester. During the filtration each well was washed five times with 300 µL of water. Subsequently, the filtermats were dried at 95 °C. The solid scintillator was melted on the dried filtermats at a temperature of 95 °C for 5 min. After solidifying of the scintillator at room temperature, the trapped radioactivity in the filtermats was measured with the scintillation analyzer. Each position on the filtermat corresponding to one well of the multiplate was measured for 5 min with the [ 3 H]-counting protocol. The overall counting efficiency was 20%. The IC 50 -values were calculated with the program GraphPad Prism ® 3.0 (GraphPad Software, San Diego, CA, USA) by non-linear regression analysis. Subsequently, the IC 50 values were transformed into K i -values using the equation of Cheng and Prusoff [40]. The K i -values are given as mean value ± SEM from three independent experiments.

Protocol of the σ 2 Receptor Binding Assay
According to [35][36][37]: the assays were performed with the radioligand

General
For Solid Phase Extraction (SPE), Sep-Pak ® C18 cartridges Plus, Plus short and Plus light (Waters, Eschborn, Germany) as well as Chromabond HR-X ® cartridges (Machery-Nagel, Düren, Germany) were tried and C18 cartridges Plus were applied routinely.

General
The experimental protocols were approved by the local ethics committee and conducted according to the national and EU regulations for animal research. Female CD1 mice (10-12 weeks old, 33.8 ± 6.2 g) were obtained from the Medizinisch-Experimentelles Zentrum, Universität Leipzig and housed under a 12 h/12 h light/dark cycle with free access to food and water for at least 24 h before experiments. Animals were sacrificed by CO 2 asphyxiation after anaesthesia with O 2 /CO 2 mixture.

In Vivo Metabolism of [ 18 F]-(R)-20
The metabolism of [ 18 F]-(R)-20 was investigated at 30 min after injection of the radiotracer (166.4 ± 65.4 MBq, dissolved in ca. 150 µL saline) into the left or right vena caudalis lateralis. Metabolites were investigated in plasma, urine, brain and liver samples. Urine samples were analyzed directly. Plasma samples were obtained by centrifugation of EDTA blood (12,000 rpm, 4 °C, 10 min) obtained by heart puncture. Brain and liver samples were acquired by homogenization of the organs in ice-cold 50 mM TRIS-HCl buffer (pH 7.4) using a PotterS ® Homogeizer (B. Braun). The tissues were treated in a borosilicate glass cylinder by 10 strokes of a PTFE plunge at a speed of 800-1000 min −1 .
The precipitation of proteins in plasma, brain, and liver samples was performed using twofold extraction of aliquots with ice-cold MeCN (1:4 v/v; for plasma 1:7 v/v), centrifugation of precipitates, and gentle concentration of the combined supernatants (~60°C, argon flow). In addition, precipitation experiments were supplemented by an alternative precipitation method using aqueous MeOH (MeOH/H 2 O 9:1). The percentages of the parent radiotracer and radiometabolites were analysed by radio-HPLC and radio-TLC. The extraction efficiency was controlled using a calibrated γ-counter (Wallac WIZARD, Perkin Elmer).

Ex Vivo Autoradiography Studies
The tracer distribution in the brain under control and blocking conditions was determined by ex vivo autoradiography studies. [ 18 F]-(R)-20 was administered via right vena caudalis lateralis without (31.0 MBq) or with (31.7 MBq) co-application 1 mg/kg haloperidol (Tocris, Bristol, UK) [44]. Animals were sacrificed at 30 min p.i. Blood was collected by heart puncture, the brain quickly removed and transferred on ice, and all samples were weighed. Radioactivity in brain and plasma samples was counted using an automated γ-counter and expressed as percentage of injected dose per gram (%ID/g). For autoradiography, the brain hemispheres were frozen immediately after isolation in 2-methylbutane (Carl Roth, Karlsruhe, Germany) at −30 °C (>2 min). Serial sagittal slices of 12 µm thickness were cut on a cryostat (Microm, Walldorf, Germany) from approx. midline, +500 µm, +1,000 µm and +1,500 µm. Slices were exposed overnight on an imaging plate (SR 2025, Fuji, Tokyo, Japan), scanned afterwards with a high-resolution phosphorimager (HD-CR 35 Bio, raytest). 2D densitometry of the whole brain was performed with AIDA software (raytest).

Synthesis
The synthesis of enantiomerically pure 2-benzopyrans of type 4 started with 2-bromostyrene (5). Bromine lithium exchange with n-BuLi at −78 °C led to an aryllithium intermediate, which was treated with 1-Boc-piperidin-4-one to yield the styrene derivative 6. The Boc-protected piperidone was chosen because the handling (work-up, purification, isolation) of the resulting carbamate was much easier than the handling of alternative benzyl substituted tertiary amines (e.g., 12). Removal of the excess of 1-Boc-piperidin-4-one was performed by LiBH 4 reduction of the keto group after complete reaction of the ketone with the lithiated vinylbenzene derivative. Flash chromatographic separation of the resulting 1-Boc-piperidin-4-ol resulted in 75% yield of the tertiary alcohol 6.
The Sharpless Asymmetric Dihydroxylation of terminal alkene 6 with AD-mix-β employing the standard protocol [45][46][47], led only to low yields of diol (R)-7 (Scheme 1). Increasing of the amount of chiral ligand (DHQD) 2 PHAL, oxidant (K 2 OsO 4 ) and cooxidant (Na 3 [Fe(CN) 6 ] and a longer reaction time did not lead to reproducible high yields of diol (R)-7. It was assumed that the reason for the low yields of (R)-7 was the low solubility of alkene 6 in a 1:1 tert-butanol-water solvent mixture. Despite the addition of different cosolvents (e.g., tert-butyl methyl ether, THF) to the solvent mixture, the yields were not improved. However, the yield was considerably raised when increasing the amount of solvent mixture from 10 mL to 60 mL/mmol alkene. The (R)-configured triol (R)-7 was obtained in a reproducible yield of 82%. Analogously the (S)-configured enantiomer (S)-7 was accessible in 74% yield, when using AD-mix-α for the dihydroxylation step.
Reaction of triol (R)-7 with tosyl chloride, NEt 3 and DMAP provided the 2-benzofuran (S)-8 in 50% yield. It is assumed that tosyl chloride reacts predominantly with the primary OH-moiety of triol (R)-7. In the presence of base the resulting primary tosylate forms an oxirane, which is opened by the tertiary alcohol giving the five-membered 2-benzofuran. Finally the primary alcohol reacts with a second equivalent tosyl chloride to form the tosylate (S)-8. Reaction of 2-(2-bromophenyl)oxirane (R)-9 with n-BuLi and subsequently with 1-Boc-piperidin-4-one led to an analogous 2-benzofuran ((S)-10) upon regioselective opening of the oxirane ring by the intermediate lithium alcoholate [35]. However, addition of catalytic amounts of Bu 2 SnO during the tosylation of triol (R)-7 afforded selectively the 2-benzopyran (R)-11, which was isolated in 62% yield. The addition of Bu 2 SnO was crucial for the synthesis of the 2-benzopyran scaffold. Bu 2 SnO is described as additive for the selective tosylation of the primary OH-moiety of diols by shielding the other OH-moiety [48]. In case of triol (R)-7, shielding of the secondary OH-moiety by Bu 2 SnO leads to selective activation (i.e., tosylation) of the primary OH-moiety for the nucleophilic substitution.
The enantiomeric purity of (R)-11 and (S)-11, which was prepared analogously, was analyzed by chiral HPLC using Daicel Chiralpak IB column, resulting in 85% ee for (R)-11 and 77.2% ee for (S)-11. The moderate enantiomeric excess is explained by the high solvent amount used for the Sharpless Asymmetric Dihydroxylation leading to a lower concentration of the chiral alkaloid ligand due to dilution effect. The lower concentration of the chiral catalyst may lead to an increased amount of uncatalyzed dihydroxylation of alkene 6. This effect has already been described in the literature [33].
For the introduction of the desired benzyl group at the piperidine ring the Boc protective group of (R)-11 was cleaved off with trifluoroacetic acid (TFA). Without further purification, the resulting secondary amine was reductively alkylated with benzaldehyde and NaBH(OAc) 3 to afford the benzyl-substituted alcohol (R)-12. After synthesis of the enantiomer (S)-12, the enantiomeric purity of both enantiomers was determined by chiral HPLC using a Daicel Chiralpak AD-H column, which resulted in 92.2% ee for (R)-12 and 76.2% ee for (S)-12.
The structure of the 2-benzopyran (R)-12 was unambiguously identified by comparison of its NMR spectra with those of the hydroxymethyl substituted 2-benzofuran (S)-10, which was obtained from the oxirane (R)-9 as reported previously by halogen/metal exchange, addition to piperidinone and exchange of the Boc-protective group with a benzyl group [33]. The 1 H-NMR spectra of the 2-benzopyran (R)-12 and the 2-benzofuran (S)-10 show three doublets of doublets for the ArCH(OR)CH 2 OR substructure. However the chemical shift of the dd for the methine proton of the 2-benzopyran (R)-12 is around 0.8 ppm high-field shifted (4.51 ppm) compared to the dd for the methine proton of the five-membered 2-benzofuran (S)-10 (5.29 ppm). After assigning the 13 C-NMR signals on the basis of the gHSQC (= gradient heteronuclear single quantum coherence) NMR spectrum, the identity of the 2-benzopyran substructure was proved by 2D gHMBC (= gradient heteronuclear multiple bond correlation) NMR spectroscopy. In this NMR experiment, couplings between protons and carbon atoms over 2-3 bonds are detected. In the 2D gHMBC NMR spectrum of the 2-benzopyran (R)-12 ( Figure 2) a coupling between the OCH 2 signals and the quaternary spiro-C-atom is observed indicating a distance of 2-3 bonds. On the contrary a corresponding crosspeak for the 2-benzofuran (S)-10 is not seen, since four bonds separate the corresponding protons and C-atom.

Receptor Binding Studies
σ 1 and σ 2 receptor affinities were measured in competition experiments with radioligands. The σ 1 receptor binding assay was carried out using a receptor preparation from guinea pig brain and [ 3 H]-(+)pentazocine as a high-affinity and selective radioligand. The σ 2 receptor affinity was conducted with a receptor preparation from rat liver and [ 3 H]di-o-toylguanidine ([ 3 H]DTG) was used as radioligand. Since DTG is not selective for the σ 2 subtype over the σ 1 subtype, σ 1 receptors binding sites were masked by addition of non-labeled (+)-pentazocine [35][36][37].
The σ 1 and σ 2 affinities of the spirocyclic 2-benzopyrans are listed in Table 1. Most of the representatives of this new class of compounds show high σ 1 affinity with K i -values in the low nanomolar range. All 2-benzopyrans are selective towards the σ 2 receptor subtype with high selectivity factors. With exception of the ethyl ether 16 (eudismic ratio 1), the (R) enantiomers represent the eutomers. The eudismic ratio varies from 1. 2 (compounds 19, 20) indicating low enantioselective receptor binding up to 29 (esters 18) revealing very high enantioselective receptor interaction. As previously reported for other spirocyclic σ 1 receptor ligands [28], the hydroxy moiety of (R)-11 acting as H-bond donor is unfavorable in terms of high σ 1 receptor affinity (K i = 5.2 nM). Methylation of the OH group led to increased σ 1 affinity. The methyl ether (R)-15 (K i = 1.2 nM) represents the most potent σ 1 receptor ligand of this series of spirocyclic 2-benzopyrans. Larger substituents like an ethyl ((R)-16) or an ethoxycarbonylethyl group ((R)-18) reduced the σ 1 affinity slightly, whereas a substituent with a polar OH group in the side chain resulted in 10-fold reduced σ 1 affinity ((R)-19: K i = 55 nM).
The fluoroethyl derivatives 20 were synthesized having in mind fluorinated PET tracers for labeling of σ 1 receptors in the brain. Due to its similar size the fluorine atom is considered as bioisosteric replacement of a proton, but due its high electronegativity it is also regarded as a bioisostere of an OH moiety. As summarized in Table 1 the ethyl derivative (R)-16 and the fluoroethyl derivative (R)-20 show very similar σ 1 receptor affinity proving the H/F bioisosterism. On the contrary the fluoroethyl derivative (R)-20 is 10-fold more active than the hydroxyethyl derivative (R)-19 indicating that in this compound class the hydroxy moiety and the fluorine atom cannot be exchanged bioisosterically by each other. The very high σ 1 affinity of the (R)-configured fluoroethyl derivative (R)-20 (K i = 4.7 nM), which is slightly higher than the σ 1 affinity of the (S)-configured enantiomer (S)-20, rendered this compound a promising candidate for molecular imaging of σ 1 receptors after labeling with [ 18 F]fluorine.  The one-step introduction of 18 F was performed by a S N 2 substitution of the precursor (R)-21 with [ 18 F]fluoride using the K[ 18 F]F-K222-carbonate complex, prepared from a 1:1 mixture of K 2 CO 3 and Kryptofix K222. Using this complex, the precursor (R)-21 was readily transformed into the 18 F-labeled radiotracer by heating in acetonitrile at 82 °C for 20 min. According to radio-TLC and radio-HPLC analyses, only a few radioactive by-products were formed. Thus, the crude reaction mixture was diluted with water to 4 mL and directly applied to semi-preparative HPLC. The radiotracer eluted at ca. 32 min and was completely free from radioactive and non-radioactive impurities. (Figure 3) Interestingly, a considerable part of 18 F activity, mainly from highly polar components remained in the stainless steel loop. Using a PEEK loop in the HPLC device resulted in about 80%-90% elution of [ 18 F]-(R)-20. Combined isolated fractions were diluted with water (50 mL), adsorbed on a Sep-Pak C18 Plus cartridge and desorbed with pure MeOH in small portions. Adsorption of the total activity

Metabolic Stability of [ 18 F]-(R)-20 in Mice
The existence of radiometabolites after injection of the radiotracer [ 18 F]-(R)-20 in mice was analyzed by radio-HPLC and radio-TLC analyses. In Figure 4 HPLC chromatograms of brain, plasma, liver and urine samples are presented in a combined manner. In brain samples (n = 4) the fraction of the non-metabolized radiotracer accounted for 91%-95% (radio-TLC) and 89%-92% (radio-HPLC) with good reproducibility. Only one highly polar (hydrophilic) radiometabolite (M1) was detected. Its retention time (t R 3.6 min) was similar to but not identical with the retention time of [ 18 F]fluoride.
Analysis of the plasma samples (n = 3) at 30 min p.i. revealed fast biotransformation (15% and 12% of [ 18 F]-(R)-20 remained unchanged as determined by acetonitrile and methanol extraction). Results from radio-HPLC and radio-TLC agreed well. The recovery of total radioactivity was 75% for acetonitrile extraction and up to 90% for methanol extraction. The main radiometabolite in plasma samples was the very polar metabolite M1 (80%-90% of total radioactivity at 30 min p.i.). Additionally, two small peaks for metabolites M2 (t R ~24.7 min) and M3 t R ~26.0 min) were detected with an intensity of lower than 3%.
In urine samples (n = 3), a large amount of radiometabolites and a very low amount of the parent radiotracer [ 18 F]-(R)-20 (non-metabolized radiotracer accounting for 2%-20% of total radioactivity) was observed.
The results obtained by analysis of liver homogenates are based on a single experiment and should be treated with caution. About 50% to 56% of the parent radiotracer [ 18 F]-(R)-20 (determined after acetonitrile and methanol extraction, respectively) remained unchanged after 30 min. The recovery of radioactivity was 45% for acetonitrile and 89% for methanol extraction. The same radiometabolite profile as in plasma and urine samples was found.

Conclusions
In this manuscript the asymmetric synthesis of spirocyclic 2-benzopyrans in enantiomerically pure form is described for the first time. The key step of the synthesis is the asymmetric dihydroxylation according to Sharpless, which allows the preparation of 4-substituted spirocyclic 2-benzopyrans. These compounds represent a new type of potent and subtype selective σ 1 receptor ligands. Some of the compounds show an eudismic ratio up to 29 (ester 18), indicating an enantioselective interaction of the σ 1 receptor with these ligands. The very potent fluorinated fluoroethoxy derivative (R)-20 was developed as fluorinated PET-tracer. The radiosynthesis based on a one-step nucleophilic substitution of tosylate (R)-21 provided the PET tracer [ 18 F]-(R)-20 in 18%-20% radiochemical yield. Whereas radiometabolites of [ 18 F]-(R)-20 were not found in the brain, plasma, liver and urine samples showed a large amount of radiometabolites. Obviously the 2-benzopyran derivative [ 18 F]-(R)-20 with a fluoroethoxy side chain is faster metabolized than the corresponding benzofuran-based radiotracers [ 18 F]2b and [ 18 F]2c with fluoroethyl or fluoropropyl side chains. In ex vivo autoradiography experiments brain regions with high σ 1 receptor expression are labeled selectively. Altogether, [ 18 F]-(R)-20 represents a very good alternative fluorinated PET tracer for imaging of σ 1 receptors in the brain.