Bioconjugate Chem. 2009, 20, 1139–1145
1139
17-[4-(2-[18F]Fluoroethyl)-1H-1,2,3-triazol-1-yl]-6-thia-heptadecanoic Acid: A Potential Radiotracer for the Evaluation of Myocardial Fatty Acid Metabolism Dong Hyun Kim, Yearn Seong Choe,* Joon Young Choi, Yong Choi, Kyung-Han Lee, and Byung-Tae Kim Department of Nuclear Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, 50 Ilwon-dong, Kangnam-ku, Seoul 135-710, Korea. Received October 30, 2008; Revised Manuscript Received May 8, 2009
In this study, we synthesized 17-[4-(2-[18F]fluoroethyl)-1H-1,2,3-triazol-1-yl]-6-thia-heptadecanoic acid ([18F]1), a PET radiotracer for the evaluation of fatty acid metabolism. [18F]1 was synthesized in 20-26% decay-corrected radiochemical yields from 17-azido 6-thia-heptadecanoic acid (9) and 4-[18F]fluoro-1-butyne using click chemistry. The tissue distribution of [18F]1 in mice showed high radioactivity accumulation in heart (3.70%ID/g at 1 min, 3.28%ID/g at 10 min, and 3.01%ID/g at 60 min postinjection), a prolonged myocardial elimination half-life (>60 min), and a maximal heart-to-blood uptake ratio at 5 min postinjection (5.55). Pretreatment with etomoxir, a carnitine palmitoyl transferase (CPT) I inhibitor, reduced myocardial radioactivity uptake at 30 min postinjection by 53%, suggesting that [18F]1 was transported into the mitochondria. Analyses of heart tissue samples showed that most of the radioactivity was present in a tissue pellet (62-63%) after homogenization in CHCl3-CH3OH followed by extraction with 40% urea and 5% H2SO4, which was mostly precipitated with addition of 50% trichloroacetic acid (TCA). These results suggest that [18F]1 undergoes metabolic trapping via β-oxidation in myocardium and, thus, suggest that it has potential use as a PET radiotracer for the evaluation of myocardial fatty acid metabolism.
INTRODUCTION Free fatty acids are myocardial energy substrates under aerobic conditions (1). In the fasting state, fatty acids undergo β-oxidation, whereas in the nonfasting state, fatty acid oxidation decreases and glucose is utilized as an energy source (2-4). Therefore, radiolabeled fatty acid analogues can be used to evaluate fatty acid metabolism in myocardium using positron emission tomography (PET) or single photon emission computed tomography (SPECT). Radioiodinated fatty acid analogues have been previously investigated in this context. 15-(p-[123I]Iodophenyl)pentadecanoic acid ([123I]IPPA) was synthesized to overcome some of the drawbacks of [123I]iodoalkyl fatty acid analogues, such as in vivo deiodination and a short myocardial clearance halftime (5-7). Fatty acid analogues labeled with positron emitting radionuclides have also been developed. 1-[11C]Palmitate was found to show biexponential clearance, composed of rapid washout resulting from β-oxidation and slow washout due to intracellular lipid pool turnover, and ω-[11C]palmitate showed greater heart uptake than 1-[11C]palmitate due to its reduced metabolism in rats (8-10). To prolong myocardial retention of fatty acid analogues, two different strategies such as methyl group substitution or the insertion of a sulfur atom at a fatty acid chain have been used. As expected, methyl group substitution at C3 of IPPA retarded its myocardial metabolism, and 15-(p-[123I]iodophenyl)-3-methylpentadecanoic acid ([123I]BMIPP) and 15-(p-[123I]iodophenyl)-3,3-dimethylpentadecanoic acid ([123I]DMIPP) were found to have prolonged myocardial retention as compared with that of [123I]IPPA (11). [123I]BMIPP showed more promise than [123I]DMIPP in terms of diagnosing various myocardial disease due to its lower liver uptake (12). A similar effect was observed in methyl * Corresponding author. Yearn Seong Choe, Tel: +82-2-3410-2623, Fax: +82-2-3410-2639, email:
[email protected].
branched ω-[18F]fluoroheptadecanoic acids at C3 (3-[18F]MFHA) or C5 (5-[18F]MFHA), and these showed higher and prolonged myocardial uptake than unsubstituted ω-[18F]fluoropalmitic acid, except at initial time point (1 min) (13). As an alternative strategy, 4- and 6-thiafatty acid analogues were developed and shown to prolong its myocardial retention (14-16), e.g., 14-[18F]fluoro-6-thia-heptadecanoic acid ([18F]FTHA) was found to have excellent biological properties, e.g., high heart uptake in mice and prolonged myocardial retention (15). It has been proposed that [18F]FTHA undergoes β-oxidation and is then metabolically trapped in myocardium. However, none of these metabolites have been identified. Click chemistry represents triazole formation via the Cu(I)catalyzed 1,3-cycloaddition of azides and terminal alkynes (17) and has been shown to be useful for synthesis of new compounds due to its simple reaction conditions, stereospecificity, wide scope of application, and so forth. Therefore, click chemistry has been used to synthesize various radiotracers, including small organic molecules and peptides (18-23). Hence, in the present study, we applied click chemistry to synthesis of 17-[4-(2-[18F]fluoroethyl)-1H-1,2,3-triazol-1-yl]-6thia-heptadecanoic acid ([18F]1) and evaluated the fatty acid analogue in mouse heart.
EXPERIMENTAL SECTION Materials and Methods. Chemicals were obtained from Sigma-Aldrich (St. Louis, MO, USA). 1H NMR spectra were obtained on a Varian Unity Inova 500NB (500 MHz) spectrometer (Palo Alto, CA, USA) at the Cooperative Center for Research Facilities, Sungkyunkwan University (Suwon, Korea). Chemical shifts (δ) are reported as ppm downfield of tetramethylsilane. Electron impact (EI) and fast atom bombardment (FAB) mass spectra were obtained on a JMS-700 Mstation (JEOL Ltd., Tokyo, Japan) at the Korea Basic Science Institute (Seoul, Korea). HPLC was carried out using a Thermo Separa-
10.1021/bc800472a CCC: $40.75 2009 American Chemical Society Published on Web 05/26/2009
1140 Bioconjugate Chem., Vol. 20, No. 6, 2009
tion Products System (Fremont, CA, USA) equipped with a semipreparative column (YMC C18, 5 µ, 10 × 250 mm) or an analytical column (YMC C18, 5 µ, 4.6 × 250 mm). Eluates were simultaneously monitored using a UV (220 nm) detector and a NaI(T1) radioactivity detector. TLC was performed on Merck F254 silica plates and analyzed using a Bioscan radioTLC scanner (Washington, DC, USA). microPET imaging of mice was acquired using an Inveon microPET (Siemens Medical Solutions, Malvern, PA, USA). [18F]Fluoride was produced by the 18O(p,n)18F reaction using a cyclotron (GE Healthcare PETtrace, Uppsala, Sweden). Radioactivities were measured using a dose calibrator (Biodex Medical Systems, Shirley, NY, USA) and tissue radioactivities using a Wallac automated gamma counter (Boston, MA, USA). All animal experiments were performed in compliance with the requirements of the Samsung Medical Center Laboratory Animal Care, in accord with NIH guidelines. Synthesis of Methyl 17-Hydroxy-6-thia-heptadecanoate (2). 11-Mercapto-1-undecanol (540 mg, 2.64 mmol) and N,Ndiisopropylethylamine (DIEA; 920 µL, 5.28 mmol) were added to methyl 5-bromovalerate (413 µL, 2.91 mmol) in CH3CN (5 mL), and the reaction mixture was stirred under N2 at 85 °C for 18 h. After solvent had been removed in vacuo, the residue was extracted with CH2Cl2, washed with water, and dried over anhydrous Na2SO4. Flash column chromatography (3:1 hexane/ ethyl acetate) gave 2 (597 mg, 71%) as a white solid. 1H NMR (CDCl3, 500 MHz) δ 3.67 (s, 3H), 3.64 (t, 2H, J ) 6.5 Hz), 2.53-2.48 (m, 4H), 2.33 (t, 2H, J ) 7.0 Hz), 1.76-1.70 (m, 2H), 1.65-1.54 (m, 6H), 1.38-1.24 (m, 14H); MS (EI) m/z 318 (100), (M+); HRMS calcd for C17H34O3S 318.2229, found 318.2230. Synthesis of Methyl 17-Methanesulfonyloxy-6-thia-heptadecanoate (3). Compound 2 (500 mg, 1.57 mmol) was dissolved in CH2Cl2 (7 mL), and to this solution was added triethylamine (438 µL, 3.14 mmol). After stirring for 10 min at room temperature, methanesulfonyl chloride (243 µL, 3.14 mmol) was added dropwise at 0 °C. The reaction mixture was then stirred under N2 at room temperature for 2 h. After quenching with 1 N HCl, the mixture was extracted with CH2Cl2, washed with water, and dried over anhydrous Na2SO4. The oily residue so obtained was purified by flash column chromatography (3:1 hexane/ethyl acetate) to give 3 (504 mg, 81%) as a pale yellow solid. 1H NMR (CDCl3, 500 MHz) δ 4.22 (t, 2H, J ) 7.0 Hz), 3.67 (s, 3H), 3.00 (s, 3H), 2.53-2.48 (m, 4H), 2.34 (t, 2H, J ) 7.0 Hz), 1.78-1.70 (m, 4H), 1.65-1.54 (m, 4H), 1.42-1.28 (m, 14H); MS (EI) m/z 396 (9.6) (M+); HRMS calcd for C18H36O5S 396.2004, found 396.2011. Synthesis of Methyl 17-Azido-6-thia-heptadecanoate (4). Sodium azide (224 mg, 3.44 mmol) was added to 3 (455 mg, 1.15 mmol) in EtOH (6 mL), and the reaction mixture was stirred at 85 °C for 10 h. After removing solvent in vacuo, the residue was extracted with CH2Cl2, washed with water, and dried over anhydrous Na2SO4. The oily residue obtained was purified by flash column chromatography (3:1 hexane/ethyl acetate) to give 4 (327 mg, 83%) as a white solid. 1H NMR (CDCl3, 500 MHz) δ 3.67 (s, 3H), 3.25 (t, 2H, J ) 7.0 Hz), 2.53-2.48 (m, 4H), 2.33 (t, 2H, J ) 7.0 Hz), 1.76-1.70 (m, 2H), 1.65-1.54 (m, 6H), 1.37-1.26 (m, 14H); MS (EI) m/z 343 (3.1) (M+); HRMS calcd for C17H33O2N3S 343.2293, found 343.2288. Synthesis of Methyl 17-[4-(2-Hydroxyethyl)-1H-1,2,3-triazol-1-yl]-6-thia-heptadecanoate (5). Compound 4 (100 mg, 0.291 mmol) was added to 3-butyn-1-ol (22.1 µL, 0.291 mmol) in a mixture of tert-butanol (2 mL) and water (1 mL). To this solution was then added a water solution (1 mL) containing CuSO4 · 5H2O (26.8 mg, 0.08 mmol) and sodium ascorbate (40 mg, 0.20 mmol). The reaction mixture was then stirred at room temperature for 1 h, and at the end of reaction, it was extracted
Kim et al.
with CH2Cl2, washed with water, and dried over Na2SO4. The crude product obtained was purified by flash column chromatography (30:1 CH2Cl2/methanol) to give 5 (96 mg, 80%) as a white solid. 1H NMR (CDCl3, 500 MHz) δ 7.39 (br s, 1H), 4.33 (t, 2H, J ) 7.0 Hz), 3.67 (s, 3H), 2.97-2.95 (m, 2H), 2.53-2.48 (m, 4H), 2.33 (t, 2H, J ) 7.5 Hz), 1.91-1.88 (m, 4H), 1.76-1.70 (m, 2H), 1.65-1.53 (m, 4H), 1.38-1.26 (m, 14H); MS (EI) m/z 413 (32.1) (M+); HRMS calcd for C21H39O3N3S 413.2712, found 413.2709. Synthesis of Methyl 17-[4-(2-Fluoroethyl)-1H-1,2,3-triazol1-yl]-6-thia-heptadecanoate (6). Compound 5 (80 mg, 0.19 mmol) was dissolved in CH2Cl2 (5 mL), and diethylaminosulfur trifluoride (DAST; 51.1 µL, 0.39 mmol) was added at -40 °C. After stirring the reaction mixture under N2 at -40 °C for 2 h, it was warmed up to room temperature, concentrated in vacuo, and purified by flash column chromatography (1:2 hexane/ethyl acetate) to give 6 (41 mg, 52%) as a pale yellow solid. 1H NMR (CDCl3, 500 MHz) δ 7.48-7.46 (m, 1H), 4.72 (dt, 2H, J ) 47.0, 6.0 Hz), 4.37-4.34 (m, 2H), 3.67 (s, 3H), 3.18 (dt, 2H, J ) 26.0, 6.0 Hz), 2.75-2.64 (m, 2H), 2.55-2.50 (m, 2H), 2.42-2.34 (m, 2H), 1.93-1.91 (m, 2H), 1.87-1.55 (m, 6H), 1.38-1.28 (m, 14H); MS (EI) m/z 415 (22.2) (M+); HRMS calcd for C21H38O2N3FS 415.2669, found 415.2674. Synthesis of 17-[4-(2-Fluoroethyl)-1H-1,2,3-triazol-1-yl]6-thia-heptadecanoic Acid (1). Compound 6 (35 mg, 0.084 mmol) was dissolved in ethanol (4 mL) and treated with 0.5 N NaOH (1 mL). After stirring at 80 °C for 2 h, the reaction mixture was cooled, neutralized with 0.1 N HCl in an ice bath, and then extracted with CH2Cl2. The organic fraction was dried over anhydrous Na2SO4 and purified by flash column chromatography (10:1 CH2Cl2/MeOH) to give 1 (15 mg, 44%) as a pale yellow solid. 1H NMR (CDCl3, 500 MHz) δ 7.52 (s, 1H), 4.69 (dt, 2H, J ) 47.0, 6.0 Hz), 4.34 (t, 2H, J ) 7.5 Hz), 3.12(dt, 2H, J ) 26.0, 6.0 Hz), 2.75-2.67 (m, 2H), 2.55-2.50 (m, 2H), 2.39-2.31 (m, 2H), 1.93-1.88 (m, 2H), 1.86-1.54 (m, 6H), 1.39-1.22 (m, 14H); MS (FAB) m/z 402 (16.3), (M++H); HRMS calcd for C20H37O2N3FS 402.5900, found 402.5891. Synthesis of 17-Hydroxy-6-thia-heptadecanoic Acid (7). Compound 7 was synthesized as described in the literature (24). 11-Mercapto-1-undecanol (800 mg, 3.91 mmol) and KOH (439 mg, 7.83 mmol) in EtOH (8 mL) were added to 5-bromovaleric acid (711.8 mg, 3.91 mmol), and stirred under N2 at 85 °C for 18 h. At the end of the reaction, the mixture was acidified with 2.5 M H2SO4. After removing solvent in vacuo, the residue was extracted with chloroform, washed with water, and dried over anhydrous Na2SO4. Flash column chromatography (30:1 CH2Cl2/ methanol) gave 7 (821 mg, 69%) as a white solid. 1H NMR (CDCl3, 500 MHz) δ 3.65 (t, 2H, J ) 6.5 Hz), 2.54-2.49 (m, 4H), 2.50 (t, 2H, J ) 7.5 Hz), 2.38 (t, 2H, J ) 7.5 Hz), 1.78-1.72 (m, 2H), 1.68-1.51(m, 6H), 1.39-1.28 (m, 14H); MS (FAB) m/z 305 (100), (M++H); HRMS calcd for C16H33O3S 305.2150, found 305.2155. Synthesis of 17-Methanesulfonyloxy-6-thia-heptadecanoic Acid (8). Compound 8 was prepared using the procedure described for 3. Flash column chromatography (2:1 hexane/ ethyl acetate) gave 8 (350 mg, 63%) as a pale yellow solid. 1H NMR (CDCl3, 500 MHz) δ 4.22 (t, 2H, J ) 6.5 Hz), 3.00 (s, 3H), 2.54-2.49 (m, 4H), 2.39 (t, 2H, J ) 7.0 Hz), 1.78-1.72 (m, 4H), 1.66-1.57 (m, 4H), 1.40-1.24 (m, 14H); MS (FAB) m/z 383 (22.0), (M++H); HRMS calcd for C17H35O5S2 383.1926, found 383.1931. Synthesis of 17-Azido-6-thia-heptadecanoic Acid (9). Compound 9 was prepared using the procedure described for 4. Flash column chromatography (50:1 CH2Cl2/methanol) gave 9 (165 mg, 64%) as a pale yellow solid. 1H NMR (CDCl3, 500 MHz) δ 3.25 (t, 2H, J ) 7.0 Hz), 2.54-2.49 (m, 4H), 2.39 (t, 2H, J ) 7.0 Hz), 1.77-1.73 (m, 4H), 1.68-1.56 (m, 4H), 1.54-1.27
PET Radiotracer for Evaluation of Fatty Acid Metabolism
(m, 14H); MS (FAB) m/z 330 (55.0), (M++H); HRMS calcd for C16H32O2N3S 330.2215, found 330.2204. Synthesis of 4-Tosyloxy-1-butyne (10). 3-Butyn-1-ol (175 µL, 2.31 mmol) was dissolved in distilled CH2Cl2 (5 mL), and to this solution was added triethylamine (323 µL, 2.31 mmol). After stirring at room temperature for 1 h, the solution was reacted with tosyl chloride (90 mg, 0.45 mmol) for 3 h at 0 °C. The mixture was then extracted with CH2Cl2 (30 mL × 3), washed with water, and dried over Na2SO4. Flash column chromatography (4:1 hexane/ethyl acetate) gave 10 (374.2 mg, 72%) as a pale yellow oil. 1H NMR (CDCl3, 500 MHz) δ 1.97 (t, 1H, J ) 3.0 Hz), 2.46 (s, 3H), 2.54-2.58 (m, 2H), 4.11 (t, 2H, J ) 7.5 Hz), 7.35 (d, 2H, J ) 8.0 Hz), 7.81 (d, 2H, J ) 8.5 Hz); MS (FAB) m/z 225 (100), (M + H)+; HRMS calcd for C11H13O3S 225.0586, found 225.0585. Radiochemical Synthesis of [18F]1. After loading a QMA cartridge with [18F]F-, radioactivity was eluted using a 1:1 solution of water and CH3CN (600 µL) containing K2,2,2 (13 mg, 0.03 mmol) and K2CO3 (3 mg, 0.02 mmol). Solvents were removed under N2 at 90 °C, and the remaining water was removed by two azeotropic distillations using 100-200 µL aliquots of CH3CN at 90 °C under a gentle stream of N2. The resulting K[18F]F was dissolved in CH3CN (1 mL) and transferred to a vial containing 4-tosyloxy-1-butyne (23). This reaction mixture was then stirred at 100 °C for 20 min, while 4-[18F]fluoro-1-butyne was distilled with acetonitrile into the second vial containing the azido precursor 9 (1.1 mg, 5.36 µmol), CuI (3.1 mg, 0.02 mmol), and sodium ascorbate (21 mg, 0.11 mmol) at -50 °C (dry ice in an acetone bath). After completing the distillation, 4-[18F]fluoro-1-butyne was obtained in 43-47% decay-corrected radiochemical yields based on [18F]fluoride ion. The reaction mixture was then warmed to room temperature, and DIEA (14 µL, 0.12 mmol) and water (250 µL) were added. After stirring for 15 min at 90 °C, the resulting mixture was concentrated to remove acetonitrile under N2 at 50 °C (water bath), diluted with water (500 µL), and then injected onto a HPLC column, which was then eluted with a linear gradient (from 30:70 to 0:100) of water containing acetic acid (2.6%) and methanol over 30 min at a flow rate of 3 mL/ min. The desired fraction, which eluted at 20-21 min, was collected, and solvents were removed using a rotary evaporator. Overall, decay-corrected radiochemical yields of [18F]1 from [18F]fluoride ion were 20-26%. Radiotracer [18F]1 was then redissolved in ethanol and diluted in saline containing 1% bovine serum albumin (BSA) at 40 °C to give a final solution of 10% ethanol in saline containing 1% BSA. Specific activity was determined by comparing the UV peak area of the desired radiolabeled HPLC peak with those of the standard curve obtained from HPLC analyses of different concentrations of nonradiolabeled compound 1. This HPLC analysis was performed, as described above using an analytical column at a flow rate of 1 mL/min. An aliquot of [18F]1 was coinjected with 1 into a HPLC system to confirm its identity. Tissue Distribution. Institute of Cancer Research (ICR) mice (male, 25-28 g, 4 mice per time point) were fasted for 14 h before injecting [18F]1 (3.0 MBq/mouse) in 0.2 mL of 10% ethanol-saline containing 1% BSA via a tail vein. Mice were sacrificed by cervical dislocation at 1, 5, 10, 30, and 60 min postinjection. Samples of blood, heart, lung, liver, kidney, and bone were removed, weighed, and counted. Data were expressed as percentages of injected dose per gram of tissue (%ID/g). In another set of experiments, 14 h fasted ICR mice (male, 25-28 g, 4 mice per time point) were injected intraperitoneally with etomoxir (40 mg/kg) at 2 h prior to a tail-vein injection of [18F]1 (3.0 MBq/mouse) (16). The mice were sacrificed by
Bioconjugate Chem., Vol. 20, No. 6, 2009 1141
cervical dislocation at 10 and 30 min postinjection. The remainder of the procedure was the same as that described above. Metabolite Analysis in Mouse Hearts. ICR mice were injected with [18F]1 (14.3 MBq) and heart tissue and urine samples were collected at 10 and 30 min postinjection. Separately, a group of mice were injected intraperitoneally with etomoxir (40 mg/kg) at 2 h before a tail-vein injection of radiotracer, and heart tissue and urine samples were collected at 30 min postinjection. Heart tissue samples were homogenized with a 2:1 mixture of CHCl3-CH3OH (1.25 mL) at 0 °C, and to this homogenate were added 40% urea (0.35 mL) and 5% H2SO4 (0.35 mL). This mixture was then vortexed and centrifuged at 3000 g for 5 min (15). The resulting organic and aqueous fractions, and tissue pellets were then counted. Aqueous fractions and urine samples were analyzed by HPLC using a linear gradient (from 30:70 to 0:100) of water containing acetic acid (2.6%) and methanol over 30 min at a flow rate of 1 mL/ min. Organic fractions were then analyzed by TLC using CH2Cl2-CH3OH (15:1) as the developing solvent. An aliquot of the organic fractions were treated with 10 N KOH at 60 °C for 1 h and analyzed by TLC as described above. On the other hand, residual tissue pellet was dissolved in 1 N NaOH at 90 °C for 1 h and treated with 50% TCA, and supernatants and precipitates were counted (15). microPET Imaging of [18F]1 in Normal Mice. ICR mice were fasted for 14 h before injecting [18F]1 (22.9 MBq), and microPET imaging of two anesthesized mice (28 g) was performed using an Inveon microPET. Dynamic scan of the mice was obtained for 30 min (12 frames × 10 s, 6 frames × 30 s, 5 frames × 60 s, 5 frames × 120 s, and 2 frames × 300 s), and the reconstruction was performed using ordered subset expectation maximization 3D (OSEM3D).
RESULTS Chemistry. Nonradiolabeled compound 1 was synthesized by the Cu(I)-catalyzed 1,3-cycloaddition of azido precursor 4 and 3-butyn-1-ol, followed by OH group fluorination and methyl ester removal (Scheme 1). The methyl ester at the other end of 5 was found to be necessary for OH group fluorination using DAST to avoid side product formation. The resulting fluoro compound 6 was obtained in higher yield at low temperature (52% at -40 °C vs 11% at 0 °C). Radiotracer [18F]1 was synthesized as above using click chemistry from the 17-azido6-thia fatty acid analogue 9 and 4-[18F]fluoro-1-butyne (Scheme 2) and purified by reverse-phase HPLC; although [18F]1 was purifiable by normal-phase HPLC, the eluted peak was very broad. The radiotracer [18F]1 obtained using this technique had high radiochemical purity (>99%) and specific activity (55.4 ( 4.2 GBq/µmol), probably because unreacted 4-[18F]fluoro-1butyne was easily removed due to its low boiling point (45 °C) (25). The decay-corrected radiochemical yields of [18F]1 were 20-26%, and the total synthesis time required, including HPLC purification, was 80-85 min. Radiotracer [18F]1 was identified by coeluting it with nonradiolabeled compound 1 on HPLC. Use of 4-[18F]fluoro-1-butyne for click chemistry with azide has advantages, such as a little perturbation of biomolecule properties and purification by distillation of 4-[18F]fluoro-1-butyne, although one of the reasons for relatively low yield of [18F]1 could be derived from incomplete transfer of 4-[18F]fluoro-1butyne to the second reaction vial, due to its low boiling point. Tissue Distribution. Since fatty acids are predominantly metabolized in the fasting state, mice were fasted for 14 h prior to experiments. After injecting [18F]1 into mice, high levels of radioactivity accumulated in blood, heart, lung, liver, and kidneys, and this heart uptake washed out slowly (3.70%ID/g at 1 min, 3.44%ID/g at 5 min, 3.28%ID/g at 10 min, 3.26%ID/g
1142 Bioconjugate Chem., Vol. 20, No. 6, 2009
Kim et al.
Scheme 1a
a Reaction conditions: (a) DIEA, CH3CN, 85 °C, 18 h; (b) MsCl, Et3N, CH2Cl2, room temperature, 2 h; (c) NaN3, EtOH, 85 °C, 10 h; (d) CuSO4, sod. ascorbate, tert-butyl alcohol, H2O, room temperature, 1 h; (e) DAST, CH2Cl2, -40 °C, 2 h (f) 0.5 N NaOH, EtOH, 80 °C, 2 h.
Scheme 2a
a Reaction conditions: (a) KOH, EtOH, 85 °C, 18 h; (b) MsCl, Et3N, CH2Cl2, room temperature, 2 h; (c) NaN3, EtOH, 85 °C, 10 h; (d) K[18F]F/ K2.2.2., CH3CN, 100 °C, 20 min; (e) 9, CuI, DIEA, sod. ascorbate, CH3CN, H2O, 90 °C, 15 min.
Table 1. Tissue Distributions of [18F]1 in Micea %ID/g
a
tissue
1 min
5 min
10 min
30 min
60 min
blood heart lung liver kidney bone
2.17 ( 0.21 3.70 ( 0.50 3.08 ( 0.10 11.81 ( 0.11 8.66 ( 0.41 0.68 ( 0.02
0.62 ( 0.04 3.44 ( 0.31 2.60 ( 0.16 12.88 ( 0.69 8.08 ( 0.54 0.53 ( 0.03
0.66 ( 0.04 3.28 ( 0.05 3.41 ( 0.28 11.18 ( 0.47 4.68 ( 0.12 0.64 ( 0.07
0.66 ( 0.02 3.16 ( 0.36 3.04 ( 0.10 5.32 ( 0.16 3.13 ( 0.20 0.77 ( 0.06
0.73 ( 0.05 3.01 ( 0.15 3.09 ( 0.13 4.73 ( 0.25 2.66 ( 0.19 1.19 ( 0.09
Values are given as means ( SD of 4 mice.
at 30 min, and 3.01%ID/g at 60 min postinjection) with an elimination half-life of >60 min (Table 1). On the other hand, blood clearance of radioactivity was almost complete within 5 min, and the heart-to-blood uptake ratios of [18F]1 were 5.55 at 5 min and 4.97 at 10 min postinjection. Constant and low bone uptake suggested that [18F]1 underwent little metabolic defluorination in vivo (0.53-0.77%ID/g up to 30 min and 1.19%ID/g at 60 min postinjection). Moreover, pretreatment of mice with etomoxir (a CPT I inhibitor) decreased the myocardial uptake of [18F]1 by 30% and 53% at 10 and 30 min, respectively (Table 2). Metabolite Analysis in Mouse Hearts. When heart tissues were homogenized in and extracted with a 2:1 mixture of CHCl3-CH3OH, 40% urea, and 5% H2SO4 (extraction efficiency of 93%), most of the radioactivity in heart tissues was found in tissue pellet (63.4:32.8:3.8 and 62.1:33.4:4.5 (tissue pellet/organic fraction/aqueous fraction) at 10 and 30 min postinjection, respectively) (Table 3). Analysis of aqueous fractions showed a radioactive peak with an HPLC retention time of 19.4-20.1 min, which was identified as [18F]1 (Figure 1A,C), although most of the radioactivity was not extracted into the aqueous fraction. HPLC analysis of the urine samples also
Table 2. Tissue Distribution of [18F]1 in Etomoxir-Pretreated Micea %ID/g
a
tissue
10 min
30 min
blood heart lung liver kidney bone
0.95 ( 0.24 2.31 ( 0.13 3.77 ( 0.20 12.43 ( 0.64 4.23 ( 0.13 0.77 ( 0.14
0.78 ( 0.06 1.47 ( 0.08 3.40 ( 0.23 8.09 ( 0.42 2.30 ( 0.14 0.90 ( 0.05
Values are given as means ( SD of 4 mice.
showed the unmetabolized [18F]1 (Figure 1B,C). TLC analyses of the organic fractions showed that [18F]1 was being converted into an unidentified nonpolar compound (41:59 and 35:65 at 10 and 30 min postinjection) (Figure 2). The residual tissue pellets were dissolved in 1 N NaOH at 90 °C, which was mostly precipitated with addition of 50% TCA (88%). In addition, the amount to precipitate was reduced to 71% when mice were pretreated with etomoxir at 2 h before [18F]1 administration. microPET Imaging of [18F]1 in Normal Mice. Early PET images (first 2 min) of mice injected with [18F]1 showed high accumulation of the radioactivity in blood pool, which decreased
PET Radiotracer for Evaluation of Fatty Acid Metabolism
Bioconjugate Chem., Vol. 20, No. 6, 2009 1143
Table 3. Distribution of Radioactivity in the Heart Homogenates Obtained after Injecting Mice with [18F]1
10 min 30 min 30 min (etomoxir-pretreated)
% organic fraction
% aqueous fraction
% tissue pellet
32.8 33.4 54.0
3.8 4.5 7.2
63.4 62.1 38.8
rapidly with time. Heart uptake of [18F]1 was clearly shown during the whole time of the study (30 min after radiotracer injection) (Figure 3), although a high level of radioactivity was also accumulated in liver and kidneys. This result showed high myocardial uptake and prolonged myocardial retention of [18F]1 in mice.
DISCUSSION Click chemistry concerns triazole formation via the Cu(I)catalyzed 1,3-cycloaddition of azides and terminal alkynes and has been found to be useful for synthesizing various triazolecontaining compounds (17). Furthermore, we synthesized 4[18F]fluoro-1-butyne, a synthon for click chemistry, and found that the 4-(2-[18F]fluoroethyl)-1H-1,2,3-triazolyl group is metabolically stable (23). The radiotracer [18F]1 was prepared from precursor 9 and 4-[18F]fluoro-1-butyne using click chemistry, which did not require carboxylic acid protection (Scheme 2). Because unsubsitituted fatty acid analogues undergo rapid myocardial clearance, a sulfur atom was inserted into the fatty acid chain to stop β-oxidation beyond the sulfur-containing site. The tissue distribution of [18F]1 showed high uptake by heart tissue (3.44%ID/g at 5 min) and high heart-to-blood uptake ratios in mice. In addition, [18F]1 underwent little in vivo defluorination, whereas odd-numbered ω-[18F]fluorofatty acid and ω-[18F]fluoro-4-thia-hexadecanoic acid have been shown
Figure 1. HPLC profiles of a water-soluble heart tissue sample (A) and a urine sample (B) obtained at 10 min after injecting mice with [18F]1 and HPLC profile of [18F]1 (C). HPLC analysis was conducted using an analytical column (YMC C18, 5 µ, 4.6 × 250 mm) and a NaI(T1) radioactivity detector.
to suffer from in vivo defluorination in rodents (16, 26). In the present study, myocardial retention of [18F]1 was found to be greater than that of other unsubstituted fatty acid analogues (
