Propanoic Acid - SAGE Journals

RESEARCH ARTICLE

Click Synthesis and Biologic Evaluation of (R)- and (S)-2Amino-3-[1-(2-[18F]Fluoroethyl)-1H-[1,2,3]Triazol-4yl]Propanoic Acid for Brain Tumor Imaging with Positron Emission Tomography Jonathan McConathy, Dong Zhou, Stephany E. Shockley, Lynne A. Jones, Elizabeth A. Griffin, Hsiaoju Lee, Susan J. Adams, and Robert H. Mach

Abstract The (R)- and (S)-enantiomers of 2-amino-3-[1-(2-[18F]fluoroethyl)-1H-[1,2,3]triazol-4-yl]propanoic acid (4) were synthesized and evaluated in the rat 9L gliosarcoma brain tumor model using cell uptake assays, biodistribution studies, and micro–positron emission tomography (microPET). The (R)- and (S)-enantiomers of [18F]4 were radiolabeled separately using the click reaction in 57% and 51% decay-corrected yields, respectively. (S)-[18F]4 was a substrate for cationic amino acid transport and, to a lesser extent, system L transport in vitro. In vivo biodistribution studies demonstrated that (S)-[18F]4 provided higher tumor uptake and higher tumor to brain ratios (15:1 at the 30- and 60-minute time points) compared to the (R)-enantiomer (7:1 at the 30- and 60-minute time points). MicroPET studies with (S)-[18F]4 confirmed that this tracer provides good target to background ratios for both subcutaneous and intracranial 9L gliosarcoma tumors. Based on these results, the 1H-[1,2,3]triazole-substituted amino acid (S)-[18F]4 has promising PET properties for brain tumors and represents a novel class of radiolabeled amino acids for tumor imaging.

ADIOLABELED AMINO ACIDS are an important class of tumor imaging agents that target the increased rates of amino acid transport that occur in many types of tumor cells. A number of radiolabeled 11 amino acids, including L-[ C]methionine, (2[18F]fluoroethyl)-L-tyrosine (FET), 6-[18F]fluoro-3,4-dihydroxy-L-phenylalanine (FDOPA), 3-[123I]iodo-a-methyl18 L-tyrosine (IMT), and 3-[ F]fluoro-a-methyl-L-tyrosine (FMT) have proven utility for imaging gliomas in human patients.1–6 These tracers have improved sensitivity and specificity relative to 2-[18F]fluoro-2-deoxy-D-glucose (FDG) for detecting brain gliomas, particularly in the setting of recurrence after treatment. Radiolabeled amino

R

Department of Radiology (Radiologic Sciences), Washington University School of Medicine, St. Louis, MO; Division of Radiological Sciences, Mallinckrodt Institute of Radiology, St. Louis, MO; and Molecular Neuroimaging, New Haven, CT. Address correspondence to: Jonathan McConathy, MD, PhD, Mallinckrodt Institute of Radiology, Division of Nuclear Medicine, 510 South Kingshighway Boulevard, Campus Box 8223, St. Louis, MO 63110; e-mail: [email protected].

DOI 10.2310/7290.2010.00025 #

2010 Decker Publishing

acids also can increase the diagnostic yield of stereotactic biopsy and provide better delineation of tumor volumes for treatment planning than contrast-enhanced magnetic resonance imaging alone. Recent data suggest that radiolabeled amino acids may be useful for monitoring response to therapy and differentiating low-grade from high-grade gliomas. The use of amino acids for imaging tumors outside the brain is not as well established, but there are recent studies with promising results in selected types of tumors. For example, [18F]FDOPA appears to be useful for imaging carcinoid and other neuroendocrine tumors, and anti-3-[18F]fluoro-1-amino-1-cyclobutane carboxylic acid (FACBC) has shown promising preliminary results for imaging prostate cancer.7–10 Amino acids enter cells via membrane-associated carrier transport proteins. Over 20 mammalian amino acid transport systems have been described with varying substrate specificities, pH dependence, sodium dependence, and regulatory mechanisms. A number of amino acid transport subtypes, including the L-type amino acid transporter 1 (LAT1) and system ASC amino acid transporter 2 (ASCT2), have been shown to be upregulated in various human tumors, including breast cancers and gliomas, and their presence may have prognostic significance.11–14 System L

Molecular Imaging, Vol 9, No 6 (November–December 2010): pp 329–342

329

330

McConathy et al

and, to a lesser extent, system A amino acid transport substrates have been the major focus of radiolabeled amino acid development for positron emission tomography (PET) and single-photon computed tomography (SPECT). Few radiolabeled amino acids are suitable for PET and SPECT that target amino acid transport systems other than system L and system A, and the utility of radiolabeled amino acids that target other transport systems for tumor imaging remains largely unexplored. The synthesis and biologic evaluation of (R)- and (S)-2amino-3-[1-(2-[18F]fluoroethyl)-1H-[1,2,3]triazol-4-yl]propanoic acid (4) was motivated by both radiosynthetic and biologic considerations. From a synthetic perspective, fluorine 18–labeled 1H-[1,2,3]triazole derivatives can be prepared through the Huisgen 1,3-dipolar cycloaddition reaction, also known as the ‘‘click’’ reaction.15–17 This reaction occurs between terminal alkynes and primary azides in the presence of a copper (I) catalyst under mild reaction conditions to provide 1H-[1,2,3]triazole substituents, typically in high radiochemical yield. In terms of biologic activity, data are very limited on the biologic transport of amino acids containing this group and its derivatives. The aromatic 1H-[1,2,3]triazole five-membered ring could act as a phenyl analogue, and if so, the resulting amino acids would be expected to be substrates for system L like other aromatic amino acids, such as L-phenylalanine, L-tyrosine, and their derivatives. Alternatively, the 1H-[1,2,3]triazole group is a nitrogen-substituted imidazole group, and amino acids containing this group could be substrates for basic amino acid transport systems that transport the naturally occurring amino acids L-histidine, L-arginine, and L-lysine. Several 1H[1,2,3]triazole-substituted amino acids have been reported in the literature, including a fluorine 18–labeled Nmethylcarbonyl ethyl ester derivative of compound 4, but their properties as amino acid transport substrates and for tumor imaging have not been reported.16,17 To determine the potential of 1H-[1,2,3]triazolesubstituted amino acids for tumor imaging, (R)- and (S)-[18F]4 were prepared and evaluated in cell uptake assays, biodistribution studies, and microPET studies using the rat 9L gliosarcoma tumor model. A primary brain tumor model was selected because of the utility of other radiolabeled amino acids in human patients with brain gliomas. The 9L gliosarcoma tumor is an established model for human glioblastoma, the highest grade of gliomas, and has been used in the preclinical evaluation of other fluorine 18–labeled amino acids.18–22 Both enantiomers of [18F]4 were evaluated because stereochemistry is known to affect the transport of many radiolabeled amino acids.18,22–24

Experimental Methods All reagents, reactions, and materials were purchased from commercially available sources. Chemicals were purchased from Aldrich Chemicals Co. (Milwaukee, WI) and Sigma Chemical Co. (St. Louis, MO) unless otherwise specified, and solvents were purchased from Aldrich Chemicals and Fisher Scientific Products (Pittsburgh, PA). Flash chromatography was carried out using Merck Kieselgel silica gel 60 (SigmaAldrich, St. Louis, MO) (230–400 mesh). Thin-layer chromatography (TLC) analyses were performed with 250 mm UV254 silica gel backing on glass plates (Aldrich Chemicals Co.). The TLC plates were developed with ninhydrin and/or phosphomolybdic acid stains. Alumina and C-18 SepPak cartridges were purchased from Waters, Inc. (Milford, MA). Ion-retardation resin (AG 11A8 50-100 mesh) was purchased from BioRad (Hercules, CA). Melting points were measured with a MelTemp 3.0 apparatus (Barnstead International, Dubuque, IA) in capillary tubes and are uncorrected. Hydrogen nuclear magnetic resonance (1H NMR) spectra were recorded on a Varian 300 MHz spectrometer, and chemical shifts (d values) are reported as parts per million (ppm) downfield from tetramethylsilane and coupling values in hertz. Elemental analyses were performed by Atlantic Microlabs, Inc. (Norcross, GA) and were within 6 0.4% of the theoretical values. The phrase ‘‘usual workup’’ refers to removal of residual water with anhydrous magnesium sulfate followed by rotary evaporation. Chemistry For both the (R)- and the (S)-enantiomers, the same reaction conditions were employed separately for the preparations of each enantiomer as depicted in Figure 1. (R)- and (S)-2-[N-(Tert-Butoxycarbonyl)Amino]-4Pentynoic Acid (1) (R)-Propargyl glycine (500 mg, 4.42 mmol) was suspended in a solution of methanol (18 mL) and triethylamine (1.8 mL), and water (1.8 mL) was added to provide a homogeneous solution. Subsequently, di-tert-butyl dicarbonate (1.89 g, 8.66 mmol, 2 Eq) was added, and the reaction mixture was stirred at room temperature overnight. The reaction mixture was concentrated under reduced pressure, and the residue was partitioned between ethyl acetate (40 mL) and 0.2 M aqueous hydrochloric acid (30 mL). The organic phase was retained, and the aqueous phase was extracted with ethyl acetate (2 3 20 mL). The

Evaluation of (R)- and (S)-[18F]AFETP for Brain Tumor Imaging

331

(R)- and (S)-2-[N-(Tert-Butoxycarbonyl)Amino]-3-[1(2-Fluoroethyl)-1H-[1,2,3]Triazol-4-yl]Propanoic Acid Tert-Butyl Ester (3)

Figure 1. Synthesis of labeling precursors (R)- and (S)-2 and (R)- and (S)-4 HCl salt. Reaction steps: (a) 1.5 eq (Boc)2O, 9:1 CH3OH, Et3N; (b) Cl3CC(5NH)OtBu, CH2Cl2; (c) FCH2CH2N3, CuI, DMF/CH3OH; (d) concentrated aqueous HCl, EtOH, 60uC.

combined organic extracts were washed with water (2 3 30 mL) followed by the usual workup to provide a crude, colorless, thick oil (1.06 g) suitable for use in the next step without further purification. 1H NMR (CDCl3) d 1.47 (s, 9H), 2.09 (t, J 5 2.9 Hz, 1H), 2.75 to 2.86 (br m, 2H), 4.49 to 4.57 (m, 1H), and 5.34 (d, J 5 8.1 Hz, 1H). For the (S)-enantiomer, a 200 mg portion of (S)propargyl glycine (200 mg, 1.77 mmol) provided crude (S)-1 (430 mg) suitable for use in the next step without further purification. The 1H NMR spectrum of (S)-1 was the same as for (R)-1. (R)- and (S)-2-[N-(Tert-Butoxycarbonyl)Amino]-4Pentynoic Acid Tert-Butyl Ester (2) The crude N-Boc acid (R)-1 (640 mg, 2.99 mmol) was dissolved in anhydrous dichloromethane (4 mL). An excess of tert-butyl-2,2,2-trichloroacetimidate (1.96 g, 8.97 mmol, 3 eq) was added, and the reaction mixture was stirred at room temperature overnight. The supernatant containing the product was isolated, and the remaining solid was washed with 3:97 ethyl acetate to hexane (4 3 5 mL). The supernatant and the washes containing the crude product were combined and concentrated under reduced pressure. Purification by silica gel flash chromatography (1:9 ethyl acetate to hexane) provided (R)-2 as a white solid (685 mg, 88%). m.p. 62.0 to 63.0uC. 1H NMR (CDCl3) d 1.45 (s, 9H), 1.49 (s, 9H), 2.02 (t, J 5 2.7 Hz, 1H), 2.69 to 2.71 (m, 2H), 4.31 to 4.37 (m, 1H), 5.35 (d, J 5 7.5, 1H). Anal. (C14H23NO4) C,H,N. For the (S)-enantiomer, the same reaction performed with (S)-1 (350 mg, 1.64 mmol) provided (S)-2 (339 mg, 77%) as a white solid. The 1H NMR spectrum of (S)-1 was the same as for (R)-1. m.p. 62.5 to 63.5uC. Anal. (C14H23NO4) C,H,N.

This reaction was performed based on conditions reported by Glaser and Arstad.17 A portion of 1-(4-methylbenzenesulfonate)-2-fluoroethanol (200 mg, 0.917 mmol, 1.07 Eq) in N,N-dimethylformamide (4 mL) was stirred with a suspension of sodium azide (179 mg, 2.75 mmol, 3.2 eq.) at room temperature. After 48 hours, the solution was filtered, and the crude 1-azido-2-fluoroethane was used immediately in the next step without further purification. In a separate flask, copper (I) iodide (873 mg, 4.58 mmol, 5.4 Eq) was suspended in methanol (2 mL) under an argon atmosphere with vigorous stirring. In rapid succession, (R)-2 (229 mg, 0.85 mmol, 1.0 Eq) dissolved in methanol (1 mL), the crude 1-azido-2-fluoroethane dissolved in N,N-dimethylformamide (4 mL), and triethylamine (640 mL, 4.59 mmol, 5.4 Eq) were added. The reaction mixture was stirred overnight at room temperature. The reaction mixture was then partitioned between saturated aqueous sodium bicarbonate (20 mL) and diethyl ether (20 mL). The organic layer was retained, and the aqueous layer was extracted with additional diethyl ether (2 3 20 mL). The combined organic layers were washed with water (3 3 20 mL) followed by the usual workup. Purification by silica gel flash chromatography (1:1 ethyl acetate) provided (R)-3 as a colorless oil (239 mg, 78%), which solidified on prolonged standing. m.p. 78.0 to 78.5uC. 1H NMR (CDCl3) d 1.42 (s, 9H), 1.43 (s, 9H), 3.22 (d, J 5 5.4 Hz, 2H), 4.45 to 4.51 (m, 1H), 4.59 to 4.62 (m, 1H), 4.67 to 4.72 (m, 2H), 4.84 to 4.87 (m, 1H), 5.43 (br d, J 5 8.1 Hz, 1H), and 7.50 (s, 1H). Anal. (C16H27FN4O4). For the (S)-enantiomer, the reaction was performed as for the (R)-enantiomer with slightly higher excesses of reagents (1.5 Eq 1-(4-methylbenzenesulfonate)-2-fluoroethanol, 4.5 Eq of sodium azide, 6.4 Eq of copper (I) iodide, 6.4 Eq of triethylamine). Using this procedure, (S)2 (82 mg, 0.30 mmol) provided (S)-3 (100 mg, 92%) as a colorless oil, which solidified on prolonged standing. m.p. 77.5 to 78.0uC. The 1H NMR spectrum of (S)-3 was the same as for (R)-3. Anal. (C16H27FN4O4). (R)- and (S)-2-Amino-3-[1-(2-Fluoroethyl)-1H[1,2,3]Triazol-4-yl]Propanoic Acid Hydrochloride Salt (4) A portion of (R)-3 (45 mg, 0.13 mmol) was dissolved in a mixture of methanol (2 mL) and 1 M aqueous hydrochloric acid (2 mL) and then sealed in a screw-top vial.

332

McConathy et al

The reaction mixture was heated at 60uC for 3 hours and then concentrated under reduced pressure. The residue was washed with diethyl ether (3 3 3 mL). The resulting oil was dissolved in a small amount of ethanol (< 0.5 mL), and the solid product was precipitated by adding diethyl ether (10 mL). The supernatant was removed to provide (R)-4 as a faintly yellow solid (23 mg, 74%). m.p. 201.5uC (decomposed). 1H NMR (D6–dimethyl sulfoxide) d 0.96 (t, J 5 6.9 Hz, 1H), 3.14 to 3.16 (m, 2H), 4.10 to 4.18 (br m, 1H), 4.54 to 4.58 (m, 1H), 4.62 to 4.67 (m, 2H), 4.78 to 4.81 (m, 1H), and 8.53 (br s, 3H). Anal. (C7H12ClFN4O2). For the (S)-enantiomer, (S)-3 (30 mg, 0.083 mmol) was dissolved in 1 M aqueous hydrochloric acid (3 mL), sealed in a screw-top vial, and then heated at 60uC for 3 hours. The solvent was removed under reduced pressure, and the residue was triturated with ethanol (0.5 mL) to provide a white solid. Residual ethanol was removed under reduced pressure, and the resulting white solid was washed with diethyl ether (3 3 3 mL) to provide (S)-4 (17 mg, 86%) as a white solid. The 1H NMR spectrum of (S)-4 was the same as for (R)-4. m.p. 202uC (decomposed). Anal. (C7H12ClFN4O2). The enantiomeric purity of (R)- and (S)-4 was evaluated by chiral high-performance liquid chromatography (HPLC) analysis with a 4.6 3 150 mm Chirex 3126 (D)-Penicillamine column (Phenomenex, Torrance, CA). The mobile phase consisted of 85:15 aqueous 3 mM copper (II) sulfate pentahydrate to acetonitrile with a flow rate of 1.5 mL/min and ultraviolet (UV) detection (l 5 254 nm). The same HPLC system was used for analysis of the fluorine 18–labeled products. 2-Azido-1-(4-Methylbenzenesulfonate)-Ethanol (5) This reaction was performed based on the procedure of Demko and Sharpless with minor modifications.15 2Bromo-1-ethanol (4.0 g, 32.0 mmol) was added to water (10 mL) followed by sodium azide (2.5 g, 38.5 mmol), and the reaction mixture was heated at reflux overnight. After cooling, the reaction mixture was extracted with dichlormethane (3 3 10 mL), and the combined organic phases

were dried over anhydrous magnesium sulfate and filtered. A 1.5 Eq portion of p-toluenesulfonyl chloride (9.15 g, 48.0 mmol) and 2.0 Eq portion of triethylamine (8.9 mL, 64 mmol) were added to the filtrate, and the reaction mixture was stirred at room temperature overnight. The reaction mixture was then washed with water (2 3 20 mL) followed by the usual workup. Purification by silica gel flash chromatography (2:8 ethyl acetate to hexane) provided 5 as a faintly yellow oil (5.1 g, 66%) at room temperature, which formed a solid when stored at 230uC. 1H NMR (300 MHz, CDCl3), 2.47 (3H, s), 3.47 to 3.51 (2H, m), 4.15 to 4.18 (2H, m), 7.36 to 7.39 (2H, m), and 7.80 to 7.84 (2H, m). Radiosynthesis The identical procedure was used for the radiosynthesis of (R)- and (S)-[18F]4 starting from the enantiomerically pure alkynes (R)- and (S)-2, respectively, as shown in Figure 2. The [18F]fluoride used for radiosyntheses was produced from [18O]H2O using the 18O(p,n)18F reaction in a CTI RDS 111 cyclotron at Washington University in the St. Louis Cyclotron Facility. Typical radiosyntheses were performed with 1.8 to 5.6 GBq of potassium [18F]fluoride eluted from a trap and release cartridge in a 0.1 M aqueous solution of potassium carbonate (200– 1,000 mL). A portion of K222 Krypotofix (5 mg, 16 mmol) in anhydrous acetonitrile (1 mL) and potassium carbonate (1 mg, 7.2 mmol) in water (0.1 mL) were added to the aqueous K[18F]F in a borosilicate vial with a screw-top cap. The solution was then azeotropically dried with acetonitrile. The 2-[18F]fluoroethyl azide 6 was prepared by adding the tosylate precursor 5 (1.5 mg, 6.2 mmol) in anhydrous acetonitrile (0.2 mL) to the dried [18F]fluoride. The reaction vial was sealed and heated to 85uC for 15 minutes. The reaction vial was then allowed to cool at ambient temperature for 2 to 3 minutes prior to use in the next step. The incorporation of [18F]fluoride could be estimated at this point by spotting the crude reaction mixture on a silica TLC plate (6:4 ethyl acetate to hexane, retention factor (Rf) 5 0.6).

Figure 2. Radiolabeling of (R)- and (S)-[18F]4. Reaction steps: (a) 1.5 eq TsCl, Et3N, CH2Cl2; (b) K[18F]F, K222, K2CO3; (c) [18F]6, CuSO4, sodium Lascorbate; (d) 1 M aqueous HCl, microwave heating.


The cycloaddition reaction was performed by adding a mixture of 1.5 M aqueous sodium L-ascorbate (50 mL) and 0.45 M of aqueous copper (II) sulfate pentahydrate (50 mL) in a syringe prepared immediately before use to the crude [18F]6 followed by the alkyne precursor 2 (5 mg, 19 mmol) in N,N-dimethylformamide (0.1 mL). The reaction vial was sealed, and the reaction was allowed to proceed at room temperature for 15 minutes with intermittent gentle shaking. The reaction mixture was diluted with acetonitrile (2 mL) and then passed in series through a cotton plug and an Alumina N light SepPak Plus cartridge (preconditioned with 10 mL of acetonitrile). Formation of the crude intermediate [18F]3 was evaluated by silica TLC (6:4 ethyl acetate to hexane, Rf 5 0.2). The eluate was concentrated to approximately 1 mL by blowing nitrogen over the surface while heating at 105uC. Purification of the intermediate [18F]3 was performed by diluting the solution with water (2 mL) and injecting the resulting solution through a 0.45 mm nylon filter onto a Zorbax SB C-18 HPLC column (10 3 250 mm, 5 mm particle size; Agilent Technologies, Santa Clara, CA). The elution was performed with a mobile phase consisting of 5:6 acetonitrile to 0.1 M ammonium acetate, a flow rate of 3 mL/min, and radiometric and UV detection (l 5 220 nm). The HPLC fractions containing radioactivity eluting at the appropriate retention time (17 minutes) were collected separately and combined. Solid-phase extraction (SPE) of the purified intermediate was performed by diluting the combined fractions with water (15:1 water to eluate volume) and passing the solution through a classic C-18 SepPak cartridge preconditioned with acetonitrile (10 mL) and then water (10 mL). The intermediate [18F]3 was eluted from the C-18 SepPak cartridge with portions of acetonitrile (0.5 mL each) with almost all of the radiolabeled product typically eluting in the second and third fractions. The acetonitrile solution containing the radiolabeled intermediate was concentrated by blowing nitrogen over the surface while heating at 105uC in a borosilicate glass vial. A portion of 1 M aqueous hydrochloric acid (0.5 mL) was added to the residue, the vial was sealed with a screw-top cap, and the reaction mixture was heated with a 60 W microwave for 30 seconds. Two different methods were used for formulating the final product [18F]4 for biologic studies. For rodent biodistribution and microPET studies, the aqueous hydrochloric acid solution was neutralized by adding an equal volume of 100 mg/mL of sodium bicarbonate in sterile 0.9 % saline. After addition of the sodium bicarbonate, the desired concentration of [18F]4 was achieved through dilution with 0.9 % saline. The final

333

doses were passed through a 0.22 mm nylon filter prior to use in rodent studies. For cell uptake studies, the aqueous hydrochloric acid solution containing the final product was added to and rapidly eluted with water through a 8 3 60 mm column of AG 11A8 ion-retardation resin preconditioned with 50 mL of water. Fractions containing radioactivity were collected and diluted to the desired concentration with water for use in cell uptake studies. The radiochemical purity, enantiomeric purity, and specific activity of the final product [18F]4 were evaluated after dose formulation. TLC analysis was performed with silica plates developed with 1:1 methanol to water. HPLC analysis was performed with 4.6 3 150 mm Chirex 3126 (D)-penicillamine chiral column eluted with a mobile phase consisting of 85:15 aqueous 3 mM copper (II) sulfate pentahydrate to acetonitrile, a flow rate of 1.5 mL/min, and radiometric and UV detection (l 5 254 nm). The identity of the product was confirmed by HPLC coinjection of nonradioactive (R)- and (S)-4. The specific activity of the product was determined by comparing the UV absorbance associated with the radiolabeled product to a standard curve obtained by injecting varying amounts of nonradioactive (R)- or (S)-4. Cell Uptake Assays 9L gliosarcoma cells were cultured in Earle’s Minimal Essential Medium (MEM) with Earle’s basic salt solution and 2 mM L-glutamine supplement with 10% newborn calf serum, 2 mM L-glutamine, 1% MEM vitamin solution, and 0.1 mM nonessential amino acids. The cells were passaged every 2 to 3 days prior to use, with no more than a total of 20 passages. Two to 3 days before the uptake assays, aliquots of 2.5 or 5.0 3 104 cells suspended in culture medium were added to each well of Costar 24-well plates to achieve log growth phase with approximately 70% confluency at the time of the uptake assay (approximately 105 cells per well). Cell uptake assays were performed using the cluster tray method reported in the literature.25 Two buffers were used for the assays: a phosphate-buffered saline solution and a sodium-free phosphate-buffered choline chloride solution. The sodium buffer consisted of 105 mM sodium chloride, 3.8 mM potassium chloride, 1.2 mM potassium bicarbonate, 25 mM sodium phosphate dibasic, 0.5 mM calcium chloride dihydrate, 1.2 mM magnesium sulfate, and 5.6 mM D-glucose. The choline buffer was the same as the sodium buffer except choline chloride was substituted for sodium chloride and choline phosphate dibasic was substituted for sodium phosphate dibasic. A 250 mM

334

McConathy et al

solution of choline phosphate dibasic was prepared by boiling a mixture of 75% aqueous choline bicarbonate (95.6 mL, 0.500 mol) and 85% aqueous phosphoric acid (16.9 mL, 0.247 mol) in 750 mL of water for 60 minutes. The solution was brought to a total volume of 1 L with water and brought to pH 7.4 with concentrated aqueous hydrochloric acid. The pH of the sodium and choline buffer solutions was adjusted to 7.4 with concentrated aqueous hydrochloric acid prior to use. The following inhibitors were added to the appropriate sodium or choline buffer: N-methyl-a-aminoisobutyric acid (MeAIB, 10 mM), a mixture of L-alanine/L-serine/Lcysteine (ASC, 3.3 mM of each amino acid), (R,S)(endo,exo)-2-aminobicyclo(2,2,1)-heptane-2-carboxylic acid (BCH, 10 mM), L-arginine (Arg, 10 mM), L-lysine (Lys, 10 mM), L-histidine (His, 10 mM), and a mixture of Arg/Lys/His (RKH, 3.3 mM of each amino acid). For control conditions, 10 mM of sucrose was added to maintain consistent osmolality relative to the inhibitor conditions. The pH of each solution of inhibitor in buffer was adjusted as necessary to 7.40 prior to use in cell uptake assays. The assay buffers with inhibitors were then formulated with radiotracer by adding 20 mL/mL of the radiolabeled amino acid [18F]4 at a concentration of 37 to 74 MBq/mL to each control and inhibitor condition. From each assay condition with radiotracer, a 100 mL sample was taken as a standard to determine the total amount of radioactivity added to each well. Each 24-well plate was used for six assay conditions with each condition performed in quadruplicate. The cell uptake assays were initiated by rinsing the cells with 2 3 2 mL of the appropriate sodium or choline buffer without inhibitor at 37uC and then adding 0.4 mL of the appropriate assay buffer at 37uC with inhibitor and (R)- or (S)-[18F]4. Uptake was allowed to proceed for 30 seconds and then terminated by rinsing the cell wells with 3 3 2 mL of the appropriate ice-cold buffer. Residual fluid was removed by pipette, and 200 mL aliquots of aqueous 0.2 M NaOH/0.2% sodium dodecyl sulfate lysis buffer was added to each cell well. The plate was then agitated at room temperature for 30 minutes, and 100 mL of the lysate was taken from each well for counting. Additional 20 mL aliquots were taken in triplicate from each well for protein concentration determination using the Pierce bicinchoninic acid protein assay kit method (Rockford, IL). The amounts of radioactivity in each sample from each well and the standard counts for each condition were measured as counts per minute (cpm) using a gamma counter and decay corrected for elapsed time. The cpm values of each well were normalized to the amount of

radioactivity added to each well and the protein concentration in the well and expressed as percent uptake relative to the sodium control condition. The data from each plate were analyzed with a one-way analysis of variance (ANOVA) with Tukey posttests using GraphPad Prism software (GraphPad Software, La Jolla, CA) with p values # .05 considered statistically significant. Biodistribution Studies with (R)- and (S)-[18F]4 in Rats with Subcutaneous 9L Tumors All animal experiments were conducted under Institutional Animal Care and Use Committee–approved protocols in compliance with the National Institutes of Health (NIH) guidelines for the care and use of research animals established by the Washington University Medical School Animal Studies Committee. All animals were housed in accordance with the NIH and Association for Assessment and Accreditation of Laboratory Animals Care guidelines in facilities maintained by the Division of Comparative Medicine. Biodistribution studies were performed with both enantiomers of [18F]4 using Fischer 344 rats (Charles River Laboratories, Wilmington, MA) implanted with subcutaneous 9L gliosarcoma tumors. The tumors were implanted by subcutaneous injection of a suspension of approximately 1 3 106 9L gliosarcoma cells into the flanks of the rats. The tumors were allowed to grow for 3 weeks prior to the study. Anesthesia was achieved using a mixture of 1% isoflurane in oxygen, and 1.1 to 1.9 MBq of the radiolabeled amino acid (R)- or (S)-[18F]4 was injected intravenously through a tail vein. The animals were euthanized in groups of four for each enantiomer at 5, 30, or 60 minutes after injection. The tumors, organs, and tissues of interest were dissected and weighed, and the amounts of activity were measured with an automated well counter. Standard dilutions of the doses were also measured, and the data were decay corrected normalized as the percentage of total dose per gram of tissue (%ID/g) for each sample. The uptake in thyroid tissue was expressed as the percent dose per organ owing to the technical difficulty of dissecting the thyroid free of the trachea. The data were analyzed using GraphPad Prism software with one-way ANOVA for each enantiomer at each time point with Tukey posttests. The kidney and pancreas were excluded from this ANOVA analysis owing to their much higher uptake values and variance compared to the other organs. The tumor and brain uptake values were directly compared for both enantiomers at each time


point with t-tests corrected for multiple comparisons; p values # .05 were considered statistically significant. MicroPET with (S)-[18F]4 in Rats with Subcutaneous and Intracranial 9L Gliosarcoma Tumors MicroPET studies were performed in two rats implanted with subcutaneous 9L tumors and one rat implanted with an intracranial 9L tumor. The subcutaneous implantations were performed as in the biodistribution studies except that the tumors were allowed to grow for 11 days prior to imaging. The intracranial tumor was implanted in the left midcerebrum using a template method as described previously26 using a total of 5 3 104 cells suspended in 5 mL total volume. Anesthesia was achieved with a mixture of 1% isoflurane/oxygen during image acquisition. Computed tomographic (CT) images were acquired with a MicroCAT II System (ImTek Inc., Knoxville, TN), and dynamic PET data were acquired with MicroPET-FOCUS 120 and 220 scanners (Concorde MicroSystems, Knoxville, TN) for 120 minutes after the intravenous tail vein injection of 8.9 to 11.5 MBq of (S)-[18F]4. Images were reconstructed with attenuation and scatter correction. The PET and CT images were fused and analyzed using an Amira software package (Visage Imaging, Inc., San Diego, CA), and time-activity curves were generated from manually drawn volumes of interest using ASIPro PET data analysis software (Concorde MicroSystems).

Results and Discussion Chemistry and Radiosynthesis of (R)- and (S)-[18F]4 The (R)- and (S)-enantiomers of nonradioactive 4 were prepared separately using the same reaction steps starting from commercially available enantiomerically pure (R)and (S)-propargyl glycine as depicted in Figure 1. The synthesis of (R)- and (S)-N-Boc t-butyl–protected propargyl glycine 2 was readily achieved in good yield in two steps. The protected fluoroethyl triazoles (R)- and (S)-3 were prepared using the method described by Glaser and Arstad17 with in situ formation of 2-fluoroethyl azide followed by the cycloaddition reaction. The protecting groups were readily removed by heating (R)- and (S)-3 in a mixture of methanol and 1 M aqueous hydrochloric acid followed by rotary evaporation to provide (R)- and (S)-4 as the hydrochloride salt in analytically pure form. No racemization was detected by chiral HPLC analysis, and the enantiomeric excess was greater than 98% for each enantiomer.

335

The initial step of the radiosynthesis of (R)- and (S)[18F]4 was performed using a method similar to that described by Glaser and Arstad.17 The preparation of the [18F]fluoroethyl azide 6 was accomplished from a tosylate precursor using standard nucleophilic substitution conditions with potassium [18F]fluoride. It was not necessary to isolate [18F]6 by distillation to achieve adequate yields in the subsequent cycloaddition reaction. The formation of [18F]6 could be assessed by TLC, although in some cases, the volatility of the product led to underestimation of the yield of this intermediate. The click reaction was performed at room temperature in the same vessel with (R)- or (S)-2 using a copper (II) sulfate/sodium Lascorbate system. The radiolabeling reactions are depicted in Figure 2. The crude cycloaddition product (R)- or (S)[18F]3 was passed through a cotton plug and alumina SepPak cartridge to remove solids and unreacted [18F]fluoride prior to purification. The crude intermediate (R)- or (S)-[18F]3 was purified using C-18 reverse-phase HPLC, and the intermediate was recovered from the HPLC eluate using SPE with a C-18 cartridge. The intermediate was deprotected quantitatively using 1 M aqueous hydrochloric acid with microwave heating to provide (R)- or (S)-[18F]4. Dose formulation for animal studies was achieved by adding a nearequimolar amount of sodium bicarbonate in 0.9% saline. For cell studies, the dose formulation was performed using an ion-retardation resin instead of sodium bicarbonate or similar agent to minimize the introduction of sodium into the sodium-free choline buffer conditions or solutes that could alter the osmolality of the uptake buffers. The total synthesis time was approximately 2.5 hours with (R)- and (S)-[18F]4 obtained in decay-corrected yields of 57 6 6% (n 5 3) and 516 8% (n 5 5), respectively. The radiochemical purity as assessed by TLC and analytical HPLC was greater than 99% after acidic deprotection and dose formulation with sodium bicarbonate for both enantiomers used for biodistribution and microPET studies. Analytical samples after dose forumulation of both enantiomers of [18F]4 were stable for at least 30 minutes as assessed by TLC and HPLC analysis. For the cell uptake studies, an ion-retardation resin was used to minimize the presence of sodium ions associated with the final product that could affect the assay conditions. Some minor decomposition of the final product was observed with the use of the ion-retardation resin, particularly if the final product was left on the column for a prolonged period of time. The radiochemical purity of the final product after the ion-retardation column was greater than 94%.

336

McConathy et al

The estimated enantiomeric purity of both the (R)- and the (S)-[18F]4 final products was greater than 98% enantiomeric excess, with the undesired enantiomer undetectable by chiral HPLC. The estimated specific activities of the final products used in biologic studies were at least 37 GBq/mmole for the (S)-enantiomer and at least 11 GBq/mmole for the (R)-enantiomer. These values represent the minimum specific activities as the sensitivity of UV detection was limited for the low concentrations of nonradioactive (R)- and (S)-4 in the final product. For both enantiomers, no mass peak corresponding to the radioactive peak of the desired product was detected during analytical HPLC analysis. The actual specific activities of both (R)- and (S)-[18F]4 are expected to be substantially higher than these minimum values. The decision to purify the intermediates (R)- and (S)[18F]3 by HPLC rather than purify the final product was based on several considerations. The protecting groups of the intermediate allowed purification using a standard reverse-phase C-18 column rather than a specialized column for unprotected amino acids. Importantly, the acidic deprotection step provided (R)- or (S)-[18F]4 in a radiochemically pure form not requiring further purification. Additionally, the final product in aqueous hydrochloric acid was readily formulated for use in cell uptake and biodistribution studies. Cell Uptake Assays with (R)- and (S)-[18F]4 Cell uptake assays were performed with 9L gliosarcoma cells using both enantiomers of [18F]4, and these results are depicted in Table 1. These assays were performed with a sodium buffer and a sodium-free choline buffer to assess the sodium dependence of the transport of (R)- and (S)[18F]4. The assays were performed using a cluster tray method, which allowed the simultaneous performance of six assay conditions each in quadruplicate with short incubation times. The short assay uptake time of 30 seconds was chosen to evaluate the initial influx of radiotracer. Some amino acid transport systems, including system L and system ASC, are bidirectional and can mediate substrate efflux, which can be minimized by using very short incubation times. A variety of inhibitor conditions were used to determine which amino acid transport system or systems were responsible for cellular uptake of (R)- and (S)-[18F]4. The N-methylated amino acid MeAIB is a selective system A substrate and has been used extensively as a competitive inhibitor of system A transport.27,28 The bulky neutral amino acid BCH has been extensively used as a competitive

Table 1. In Vitro Uptake of (R)- and (S)-[18F]4 by 9L Gliosarcoma Cells in the Presence and Absence of Amino Acid Transport Inhibitors Inhibitor Neutral amino acid transport Na sucrose Na ASC Na MeAIB Na BCH Cho sucrose Cho BCH Basic amino acid transport Na sucrose Na ASC Na Arg Na Lys Na His Na RKH

(S)-[18F]4

(R)-[18F]4

100 6 23 11 6 7 82 6 7 74 6 3 58 6 7 30 6 6

100 6 4.8 64 6 6.3 97 6 7.3 110 6 11 107 6 14 95 6 3.9

100 6 15 264 50 6 9 46 6 4 31 6 5 37 6 6

100 6 10 67 6 15 83 6 7 104 6 8 91 6 10 82 6 4

ASC 5 L-Ala; BCH 5 2-aminobicyclo(2,2,1)-heptane-2-carboxylic acid (system L inhibitor); Cho 5 choline buffer; MeAIB 5 N-methyl aaminoisobutyric acid (system A inhibitor); Na 5 sodium buffer, L-Ser, LCys mixture; RKH 5 L-Arg, L-Lys, L-His mixture. The uptake data were normalized based on the amount of activity added to each well and to the amount of protein present in each well. The data are expressed as percent uptake relative to control conditions with standard deviation. See the Methods section for experimental details. The Na sucrose and Cho sucrose conditions represent controls for the sodium-containing and sodium-free conditions, respectively. (S)-[18F]4, neutral amino acid inhibitors one-way ANOVA results: for Na sucrose vs Na ASC, Na sucrose vs Cho BCH, Na MeAIB vs Na ASC, Na MeAIB vs Cho BCH, Na BCH vs Na ASC, and Na ASC vs Cho sucrose, p , .001; for Na sucrose vs Cho sucrose and Na BCH vs Cho BCH, p , .01; for Cho sucrose vs Cho BCH, p , .05. (R)-[18F]4, neutral amino acid inhibitors one-way ANOVA results: for Na sucrose vs Na ASC, Na MeAIB vs ASC, Na BCH vs Na ASC, and Na ASC vs Cho sucrose, p , .001; for Na ASC vs Cho BCH, p , .01. (S)-[18F]4, basic amino acid inhibitors one-way ANOVA results: for Na sucrose vs Na ASC, Na sucrose vs Na Arg, Na sucrose vs Na Lys, Na sucrose vs Na His, Na sucrose vs RKH, Na ASC vs Na Arg, Na ASC vs Na Lys, and Na ASC vs Na RKH, p , .001; for Na ASC vs Na His, p , .01. (R)-[18F]4, basic amino acid inhibitors one-way ANOVA results: for Na sucrose vs Na ASC and Na ASC vs Na Lys, p , .01; for Na ASC vs Na His, p , .05.

inhibitor of system L transport, and under sodium-free conditions, BCH is a selective system L inhibitor.29 In the presence of sodium, BCH is also an inhibitor of system B0,+, which is a sodium-dependent amino acid transporter that typically is a minor transport system in most tissues. A mixture of L-Ala, L-Ser, and L-Cys (ASC) was used as a nonspecific inhibitor condition, and this combination of amino acids is expected to inhibit a broad range of amino acid transport systems, including system ASC. Initial cell uptake assays were performed with MeAIB, BCH, and ASC inhibitor conditions primarily to evaluate


the role of system A and system L transport in the uptake of both enantiomers of [18F]4 by 9L gliosarcoma cells. In the case of (S)-[18F]4, there was no significant inhibition of uptake by MeAIB, indicating that (S)-[18F]4 is not a substrate for system A transport under these assay conditions. In contrast, the ASC condition inhibited 89% of uptake relative to the sodium control, indicating that almost all of the cell uptake of (S)-[18F]4 was mediated by amino acid transport. Additionally, substituting choline for sodium led to a 42% inhibition of uptake, consistent with a sodium-dependent component of transport of (S)[18F]4. In the sodium-free conditions, the addition of BCH led to an approximately 50% reduction of the total uptake relative to the choline control. The inhibition of uptake by BCH under sodium-free conditions indicates that (S)[18F]4 is a substrate for system L transport, accounting for approximately 28% of total uptake. The initial assays with (S)-[18F]4 demonstrated that this compound enters cells via sodium-dependent and sodium-independent amino acid transport with approximately 30% of uptake mediated by system L. Because of the results of initial cell uptake assays and the structural relationship of the 1H-[1,2,3]triazole group to the imidazole side chain of L-His, additional uptake assays using L-His as well as the basic side chain amino acids LArg and L-Lys individually and together (RKH) as competitive inhibitors were performed. These conditions inhibited the uptake of (S)-[18F]4 relative to control, with the magnitude of inhibition ranging from 50% with L-Arg to 69% with L-His. These data are consistent with (S)-[18F]4 entering 9L gliosarcoma cells via cationic amino acid transport. In contrast, the same uptake assays performed with (R)-[18F]4 did not demonstrate significant uptake mediated by system L or cationic amino acid transport systems and only partial inhibition with the nonspecific ASC condition. These results suggest that the majority of cell uptake of (R)-[18F]4 under these assay conditions was mediated by amino acid transport systems not affected by the inhibitors used in these studies or alternatively was not mediated by amino acid transport systems. These results also demonstrate that stereochemistry at the a-carbon has a substantial effect on the biologic transport of [18F]4. In summary, the in vitro uptake of (S)-[18F]4 by 9L gliosarcoma cells is mediated by a combination of cationic amino acid transport and, to a lesser extent, system L transport. The entry of cationic amino acids into cells can be mediated by multiple sodium-independent amino acid transporter systems, including the cationic amino acid transporter (CAT) family and the system y+L family.

337

Transport of (S)-[18F]4 by one or more of these transport systems is compatible with the sodium-independent transport observed in the uptake assays. Cationic amino acids can also be transported by amino acid transport systems b0,+ AT and ATB0,+ in a sodium-dependent fashion, which may account for some or all of the sodium-dependent transport of (S)-[18F]4.30,31 More extensive amino acid transport assays will be needed to determine which of these transport systems mediate the uptake of (S)-[18F]4 by 9L gliosarcoma cells. Biodistribution of (R)- and (S)-[18F]4 in Rats with Subcutaneous 9L Gliosarcoma Tumors The biodistribution results obtained with (R)- and (S)[18F]4 are presented in Table 2 and Table 3. (S)-[18F]4 demonstrated superior properties for brain tumor imaging compared to (R)-[18F]4. At all time points, the magnitude of tumor uptake was higher with (S)-[18F]4 than with the (R)-enantiomer (p , .04 at 5 minutes, p , .03 at 30 minutes, p , .0003 at 60 minutes). Both compounds demonstrated peak tumor uptake at the 30-minute time point with 0.83 %ID/g observed with (S)-[18F]4 and 0.29 %ID/g with (R)-[18F]4. The brain uptake ranged from 0.02 to 0.09 %ID/g, with significantly higher uptake of (S)[18F]4 at 30 and 60 minutes than with the (R)-enantiomer (p , .02 and p , .0003, respectively). These properties led to maximal tumor to brain ratios of 15.2:1 with (S)-[18F]4 at 60 minutes postinjection and 7.4 with (R)-[18F]4 at 30 minutes postinjection. For both enantiomers, the uptake of activity in tumor was significantly greater than in normal brain at all time points except with (R)-[18F]4 at 5 minutes. These results indicate that (S)-[18F]4 is the more biologically active transport substrate with higher uptake in both tumor and normal brain as well as better tumor to brain ratios in the 9L model. The distribution of (R)- and (S)-[18F]4 in the normal organs and tissues demonstrated the highest uptake in the kidneys followed by the pancreas at all time points. This pattern of biodistribution has been observed with a wide range of radiolabeled amino acids.2,18,19,21,32 The very high uptake in the kidneys with relatively low small bowel activity at 60 minutes postinjection suggests a renal route of excretion. As in the brain, (S)-[18F]4 demonstrated superior properties for tumor imaging outside the brain compared to the (R)-enantiomer with higher tumor to normal tissue ratios. At 60 minutes postinjection of (S)[18F]4, the magnitude of tumor uptake of radioactivity was significantly higher than in the normal organs and tissues (p , .01) except in the kidneys and pancreas. The tumor

338

McConathy et al

Table 2. Biodistribution of (S)-[18F]4 in Fischer Rats with Subcutaneous 9L Gliosarcoma Tumors

Table 3. Biodistribution of (R)-[18F]4 in Fischer Rats with Subcutaneous 9L Gliosarcoma Tumors

Organ/Tissue

Organ/Tissue

Blood Bone Brain Fat Heart Kidney Large intestine Liver Lung Muscle Pancreas Salivary gland Small intestine Spleen Testes Thyroid Tumor

5 min

30 min

0.98 6 0.15 0.47 6 0.065 0.08 6 0.011 0.09 6 0.020 0.43 6 0.048 11.0 6 1.2 0.14 6 0.012 0.34 6 0.035 0.87 6 0.12 0.18 6 0.030 2.75 6 0.63 0.74 6 0.043 0.69 6 0.076 0.72 6 0.054 0.21 6 0.013 0.71 6 0.12 0.47 6 0.069

0.47 6 0.10 0.36 6 0.068 0.06 6 0.008 0.05 6 0.010 0.38 6 0.066 11.4 6 1.9 0.15 6 0.028 0.43 6 0.091 0.83 6 0.18 0.18 6 0.037 2.20 6 0.75 0.66 6 0.14 0.54 6 0.073 0.75 6 0.12 0.17 6 0.033 0.44 6 0.10 0.83 6 0.14

60 min 0.23 0.23 0.05 0.04 0.30 4.79 0.10 0.32 0.56 0.16 1.12 0.36 0.31 0.47 0.11 0.30 0.72

6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6

0.015 0.011 0.001 0.007 0.011 0.38 0.012 0.021 0.085 0.006 0.15 0.037 0.024 0.035 0.006 0.024 0.098

Blood Bone Brain Fat Heart Kidney Large intestine Liver Lung Muscle Pancreas Salivary gland Small intestine Spleen Testes Thyroid Tumor

5 min 0.90 0.32 0.09 0.10 0.42 14.0 0.15 0.54 0.74 0.18 0.53 0.46 0.32 0.33 0.18 0.64 0.27

6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6

0.17 0.083 0.015 0.044 0.076 3.8 0.043 0.15 0.14 0.049 0.13 0.083 0.079 0.065 0.042 0.092 0.042

30 min 0.28 0.13 0.04 0.03 0.13 6.76 0.08 0.58 0.54 0.08 0.50 0.20 0.30 0.18 0.08 0.21 0.29

6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6

0.037 0.027 0.005 0.012 0.014 0.83 0.060 0.074 0.211 0.020 0.10 0.016 0.15 0.004 0.011 0.023 0.13

60 min 0.14 0.08 0.02 0.02 0.08 2.39 0.05 0.48 0.37 0.05 0.43 0.15 0.35 0.14 0.05 0.12 0.17

6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6

0.015 0.007 0.001 0.002 0.006 0.14 0.002 0.023 0.10 0.006 0.063 0.009 0.015 0.002 0.002 0.010 0.009

The data are expressed as %ID/g 6 standard deviation except for the thyroid, which is expressed as a percentage of the total dose per organ. n 5 3 or 4 for each value.

The data are expressed as %ID/g 6 standard deviation except for the thyroid, which is expressed as a percentage of the total dose per organ. n 5 3 or 4 for each value.

to normal tissue ratios at 60 minutes were at least 2:1 with (S)-[18F]4 except in the lung and the spleen, which demonstrated ratios of 1.3:1 and 1.6:1, respectively. The uptake of radioactivity observed in bone was relatively low with both compounds, indicating that significant in vivo defluorination did not occur during the time course of the studies. Comparisons of tumor to normal tissue ratios at each time point obtained with (R)- and (S)-[18F]4 are depicted in Table 4. For most tissues, including the 9L tumors, (S)-[18F]4 had similar to higher tissue to blood ratios than the (R)-enantiomer, which was most pronounced at 60 minutes as depicted in Table 5. These biodistribution results in conjunction with the cell uptake assays indicate that 1H-[1,2,3]triazole-substituted amino acids are biologically active and represent a promising class of nonnatural amino acids for tumor imaging. To date, substrates for cationic amino acid transport systems have not been systematically evaluated as potential tumor imaging agents. Although (S)-[18F]4 was not entirely selective for cationic amino acid transport, the 50 to 69% inhibition of uptake by L-His, L-Arg, and L-Lys is consistent with a substantial component of cationic amino acid transport. The ability to measure the rates of cationic amino acid transport by tumor cells may provide clinically relevant information beyond tumor detection not available with radiolabeled amino acids that target other

transport systems. The cationic amino acid transport substrate L-Arg is semiessential and cannot be synthesized in adequate amounts to meet the needs of rapidly proliferating cells.33 Additionally, L-Arg plays roles in angiogenesis through nitric oxide synthesis and in polyamine synthesis in proliferating cells.33,34 Thus, radiolableled amino acids targeting cationic amino acids could provide a noninvasive means to characterize important metabolic pathways in tumor cells. The component of system L transport observed with (S)[18F]4 may also play a role in the tumor uptake of this compound given the efficacy of system L substrates for brain tumor imaging. However, tumor to brain ratios observed with (S)-[18F]4 are higher than with other, more selective system L substrates, which are typically on the order of 2:1 to 3:1. For example, in rats with subcutaneously implanted 9L gliosarcoma tumors, [18F]FET demonstrated a tumor to brain ratio of 2.3:1 at 60 minutes postinjection,35 and [18F]fluoroalkyl tyrosine derivatives demonstrated tumor to brain ratios ranging from 1.9 to 2.9:1 at the same time point.36 Similarly, the nonnatural amino acid FACBC demonstrated a tumor to brain ratio of 6.6:1 at 60 minutes after injection in the 9L model.21 The higher tumor to brain ratio observed with (S)-[18F]4 may be due to the component of cationic amino acid transport. The reason for the relatively high uptake in lung and spleen observed


339

Table 4. Tumor to Organ Ratios Observed with (R)-[18F]4 and (S)-[18F]4 5 min

Blood Bone Brain Fat Heart Kidney Large intestine Liver Lung Muscle Pancreas Small intestine Spleen Salivary gland Testes

30 min

(S)-[18F]4

(R)-[18F]4

0.52 6 0.12 1.1 6 0.23 6.3 6 0.80 5.7 6 1.1 1.2 6 0.25 0.04 6 0.002 3.7 6 0.5 1.5 6 0.31 0.58 6 0.14 3.0 6 0.62 0.17 6 0.03 0.72 6 0.15 0.67 6 0.15 0.63 6 0.13 2.3 6 0.45

0.33 6 0.08 0.99 6 0.30 3.3 6 0.18 3.6 6 1.6 0.70 6 0.11 0.02 6 0.01 2.1 6 0.98 0.59 6 0.16 0.38 6 0.07 1.8 6 0.53 0.58 6 0.17 1.0 6 0.15 0.90 6 0.20 0.63 6 0.13 1.7 6 0.42

(S)-[18F]4 1.8 2.2 15.1 16.7 2.2 0.07 5.5 1.8 1.0 4.6 0.41 1.5 1.1 1.3 5.0

6 6 6 6 6 6 6 6 6 6 6 6 6 6 6

0.41 0.55 4.4 3.6 0.54 0.01 1.2 0.35 0.20 0.98 0.15 0.26 0.21 0.34 1.1

60 min (R)-[18F]4

(S)-[18F]4

(R)-[18F]4

1.0 6 0.48 2.4 6 1.0 7.4 6 3.0 8.8 6 5.7 2.2 6 1.2 0.04 6 0.02 3.5 6 1.1 0.51 6 0.19 0.55 6 0.06 3.9 6 1.7 0.6 6 0.21 2.2 6 2.7 1.6 6 0.73 1.5 6 0.78 3.8 6 1.6

3.2 6 0.61 3.1 6 0.50 15.2 6 2.0 20.9 6 5.7 2.4 6 0.38 0.15 6 0.03 7.0 6 1.3 2.2 6 0.40 1.3 6 0.31 4.5 6 0.68 0.65 6 0.14 2.4 6 0.42 1.6 6 0.29 2.0 6 0.46 6.4 6 1.1

1.2 6 0.07 2.1 6 0.11 6.9 6 0.27 9.8 6 0.75 2.0 6 0.13 0.07 6 0.001 3.8 6 0.28 0.4 6 0.03 0.50 6 0.15 3.3 6 0.44 0.41 6 0.09 0.49 6 0.03 1.2 6 0.05 1.2 6 0.12 3.3 6 0.22

The data are expressed as the average ratios of tumor uptake (%ID/g) divided by the corresponding normal organ/tissue uptake (%ID/g) for each animal at each time point 6 standard deviation. n 5 3 or 4 for each value.

with (S)-[18F]4 is unclear but may be due to cationic amino acid transport by the endothelium in these organs. Further studies are needed to determine which of the amino acid transporters that mediate cationic amino transport are involved in the uptake of (S)-[18F]4 in vivo. These effects of the a-carbon stereochemistry on the mechanism of transport and biodistribution of (R)- and

(S)-[18F]4 are in keeping with results obtained with a number of other nonnatural radiolabeled amino acids showing that stereochemistry can have important effects on the biologic behavior.18,22–24 The (R)- and (S)enantiomer of [18F]4 demonstrated different in vitro transport properties, and the (S)-enantiomer had superior properties for tumor imaging in the 9L gliosarcoma model.

Table 5. Tissue to Blood Ratios Observed with (R)-[18F]4 and (S)-[18F]4 5 min (S)-[18F]4 Bone Brain Fat Heart Kidney Large intestine Liver Lung Muscle Pancreas Small intestine Spleen Salivary gland Testes Tumor

0.48 6 0.01 0.08 6 0.01 0.09 6 0.01 0.44 6 0.03 11.4 6 2.5 0.14 6 0.01 0.35 6 0.02 0.89 6 0.04 0.18 6 0.01 2.89 6 1.0 0.71 6 0.04 0.74 6 0.08 0.77 6 0.11 0.22 6 0.02 0.39 6 0.28

30 min (R)-[18F]4 0.35 0.10 0.11 0.47 15.4 0.17 0.59 0.83 0.20 0.59 0.36 0.36 0.51 0.20 0.33

6 0.03 6 0.02 6 0.05 6 0.03 6 1.6 6 0.06 6 0.07 6 0.08 6 0.04 6 0.04 6 0.04 6 0.01 6 0.03 6 0.01 6 0.08

(S)-[18F]4 0.78 6 0.07 0.12 6 0.02 0.11 6 0.0 0.82 6 0.1 24.4 6 1.1 0.32 6 0.01 0.92 6 0.1 1.76 6 0.11 0.38 6 0.01 4.59 6 0.68 1.16 6 0.09 1.60 6 0.08 1.41 6 0.10 0.35 6 0.01 1.33 6 0.9

60 min (R)-[18F]4 0.49 0.14 0.11 0.47 24.4 0.30 2.14 2.00 0.31 1.85 1.13 0.67 0.74 0.30 0.75

6 6 6 6 6 6 6 6 6 6 6 6 6 6 6

0.16 0.01 0.03 0.03 2.2 0.21 0.5 0.82 0.11 0.6 0.65 0.1 0.15 0.08 0.6

(S)-[18F]4 1.02 0.21 0.16 1.31 21.0 0.46 1.42 2.46 0.70 4.91 1.34 2.06 1.56 0.50 3.19

6 6 6 6 6 6 6 6 6 6 6 6 6 6 6

0.10 0.02 0.03 0.11 1.4 0.04 0.04 0.47 0.05 0.50 0.07 0.12 0.09 0.03 0.61

(R)-[18F]4 0.58 6 0.03 0.17 6 0.01 0.12 6 0.004 0.59 6 0.05 16.8 6 0.90 0.32 6 0.039 3.36 6 0.36 2.60 6 0.83 0.38 6 0.07 3.07 6 0.68 2.45 6 0.27 1.01 6 0.09 1.03 6 0.16 0.36 6 0.04 1.21 6 0.07

The data are expressed as the average ratios of tissue uptake (%ID/g) divided by the concentration of activity in the blood (%ID/g) for each animal at each time point with standard deviation. n 5 3 or 4 for each value.

340

McConathy et al

The (S)-enantiomer of [18F]4 was evaluated in microPET studies in rats with subcutaneously and intracranially implanted 9L gliosarcoma tumors. These results corroborated the biodistribution studies and confirmed the favorable in vivo tumor imaging properties of (S)-[18F]4 in this tumor model. The time-activity curves are depicted in Figure 3, and representative images are shown in Figure 4. Near-maximal tumor uptake occurred with both the intracranial and subcutaneous tumors at approximately 30 minutes postinjection, with maximal uptake at 27.5 and 32.5 minutes, respectively. During the remainder of the 2hour study, the amount of radioactivity in both the intracranial and subcutaneous tumors decreased slowly. In contrast, the uptake in the adjacent normal tissues (brain, paraspinal muscle, and spinal column) peaked within 5 minutes postinjection and washed out rapidly. The tumor to normal tissue ratios measured in the microPET studies were in good agreement with those observed in the biodistribution studies. The subcutaneous tumor to paraspinal muscle microPET uptake ratio was approximately 5.0:1 at 30 minutes compared to the biodistribution tumor to muscle ratio of 4.6:1 at 30

minutes. Similarly, the tumor to bone (spinal column) ratio was 2.5:1 at 30 mintes in the microPET study compared to a tumor to bone (femur) ratio of 2.2:1 in the biodistribution study at the same time point. Finally, the maximal tumor to brain ratios in the microPET study were approximately 9:1 compared to 15:1 in the biodistribution study. This difference in tumor to brain ratios between the intracranial and subcutaneous tumors is due at least in part to partial volume effect in the microPET study. The relatively small size of the intracranial tumor and the low uptake in adjacent normal brain are expected to accentuate partial volume effects with the intracranial tumor. The blood-brain barrier (BBB) may play an important role in the relatively high tumor to brain ratios observed with (S)-[18F]4 in the biodistribution and microPET studies. The relatively low normal brain uptake of both enantiomers of [18F]4 may be due to exclusion of these tracers by the BBB and contribute to the high tumor to normal brain ratios. In contrast, the BBB is not an issue for in vivo tracer uptake by 9L gliosarcoma cells as intracranially implanted 9L tumors do not have a functional BBB as demonstrated by gadoliniumenhanced magnetic resonance imaging studies.37,38 However, the differences in the biologic behaviors of the (R)- and (S)enantiomers in the cell uptake assays and the in vivo biodistribution studies provide strong evidence that the high tumor to brain ratios observed with (S)-[18F]4 are not due to passive BBB effects alone. Failure to cross the normal BBB may limit the role of (S)-[18F]4 for imaging gliomas with relatively intact BBBs, in particular low-grade gliomas. The efficacy of (S)-[18F]4 for imaging non–contrast-enhancing

A

B

This difference is most likely due to the fact that the position of the 1H-[1,2,3]triazole group of (S)-[18F]4 corresponds to the configuration of the side chains of natural L-amino acids, including L-His, L-Arg, and L-Phe. MicroPET Studies with (S)-[18F]4 in Rats with Subcutaneous and Intracranial 9L Gliosarcoma Tumors

Figure 3. A, Time-activity curve from a microPET study performed with (S)-[18F]4 in a Fischer rat with an intracranially implanted 9L gliosarcoma tumor. The data are presented as a percentage of the injected dose per cubic cm of tissue (%ID/cc) calculated as (Bq/cc of tissue)/ (injected dose in Bq) 3 100%. B, Time-activity curve from a microPET study performed with (S)-[18F]4 in a Fischer rat with a subcutaneously implanted 9L gliosarcoma tumor. The data are presented as a percentage of the injected dose per cubic centimeter of tissue (%ID/cc) calculated as (Bq/cc of tissue)/(injected dose in Bq) 3 100%.


341

Figure 4. A, PET (left) and fused PET-CT (right) images from a microPET study performed with (S)[18F]4 in a Fischer rat with an intracranial 9L gliosarcoma tumor. The PET portion of this image represents summed data from 60 to 120 minutes postinjection. The intracranial focus of high uptake represents the 9L gliosarcoma tumor. B, Fused PET-CT image from a microPET study performed with (S)-[18F]4 in a Fischer rat with a subcutaneous 9L gliosarcoma tumor. The PET portion of this image represents summed data from 60 to 120 minutes postinjection. The focus of high uptake in the right flank (left side of image) represents the 9L gliosarcoma tumor. High uptake is also seen in the bladder (inferior aspect of image at midline) and the right renal collecting system (upper left aspect of image).

A

B

gliomas will need to be addressed experimentally. The potential effects of the BBB are not an issue for imaging tumors outside the central nervous system.

Conclusions The radiolabeled amino acid (S)-[18F]4 has good tumor imaging properties in the 9L gliosarcoma tumor model as demonstrated through cell uptake, biodistribution, and microPET studies. These results also indicate that radiolabeled amino acids substituted with the 1H[1,2,3]triazole group are biologically active and represent a novel promising class of tumor imaging agents. Cell uptake assays demonstrated that (S)-[18F]4 is a substrate for cationic amino acid transport and, to a lesser degree, system L transport. The biologic behavior of [18F]4 is influenced by the stereochemistry at the a-carbon in both in vitro cell uptake and in vivo biodistribution studies, with the (S)-enantiomer possessing superior imaging properties. The tumor to brain ratios obtained with (S)[18F]4 in biodistributions studies were 15:1 at the 30- and 60-minute time points compared to 7:1 at these time

points with the (R)-enantiomer. Additionally, the uptake in the tumor was significantly higher at 60 minutes than in most normal tissues with the exception of the kidney and pancreas, indicating that (S)-[18F]4 may also be useful for imaging tumors outside the brain. MicroPET images in rats with subcutaneous and intracranial 9L gliosarcoma tumors confirmed the biodistribution study results and illustrate the promising imaging properties of (S)-[18F]4.

Acknowledgment Financial disclosure of authors and reviewers: The radiotracers reported in this article are subject to an option for licensing from Washington University in St Louis (J McConathy and RH Mach, inventors) by Isotrace Technologies, Inc. (St. Charles, MO).

References 1. Singhal T, Narayanan TK, Jain V, et al. [11C]-L-Methionine positron emission tomography in the clinical management of cerebral gliomas. Mol Imaging Biol 2008;10:1–18.

342

McConathy et al

2. Langen KJ, Hamacher K, Weckesser M, et al. O-(2[18F]Fluoroethyl)-L-tyrosine: uptake mechanisms and clinical applications. Nucl Med Biol 2006;33:287–94. 3. Chen W, Silverman DH, Delaloye S, et al. 18F-FDOPA PET imaging of brain tumors: comparison study with 18F-FDG PET and evaluation of diagnostic accuracy. J Nucl Med 2006;47:904–11. 4. Langen KJ, Pauleit DCoenen HH. 3-[123I]Iodo-alpha-methyl-Ltyrosine: uptake mechanisms and clinical applications. Nucl Med Biol 2002;29:625–31. 5. McConathy J, Goodman MM. Non-natural amino acids for tumor imaging using positron emission tomography and single photon emission computed tomography. Cancer Metastasis Rev 2008;27: 555–73. 6. Inoue T, Shibasaki T, Oriuchi N, et al. 18F Alpha-methyl tyrosine PET studies in patients with brain tumors. J Nucl Med 1999;40: 399–405. 7. Becherer A, Szabo M, Karanikas G, et al. Imaging of advanced neuroendocrine tumors with 18F-FDOPA PET. J Nucl Med 2004; 45:1161–7. 8. Montravers F, Grahek D, Kerrou K, et al. Can fluorodihydroxyphenylalanine PET replace somatostatin receptor scintigraphy in patients with digestive endocrine tumors? J Nucl Med 2006;47: 1455–62. 9. Schuster DM, Votaw JR, Nieh PT, et al. Initial experience with the radiotracer anti-1-amino-3-18F-fluorocyclobutane-1-carboxylic acid with PET/CT in prostate carcinoma. J Nucl Med 2007;48:56–63. 10. Oka S, Hattori R, Kurosaki F, et al. A preliminary study of anti-1amino-3-18F-fluorocyclobutyl-1-carboxylic acid for the detection of prostate cancer. J Nucl Med 2007;48:46–55. 11. Fuchs BC, Finger RE, Onan MC, et al. ASCT2 silencing regulates mammalian target-of-rapamycin growth and survival signaling in human hepatoma cells. Am J Physiol Cell Physiol 2007;293:C55–63. 12. Fuchs BC, Bode BP. Amino acid transporters ASCT2 and LAT1 in cancer: partners in crime? Semin Cancer Biol 2005;15:254–66. 13. Esseghir S, Reis-Filho JS, Kennedy A, et al. Identification of transmembrane proteins as potential prognostic markers and therapeutic targets in breast cancer by a screen for signal sequence encoding transcripts. J Pathol 2006;210:420–30. 14. Nawashiro H, Otani N, Shinomiya N, et al. L-type amino acid transporter 1 as a potential molecular target in human astrocytic tumors. Int J Cancer 2006;119:484–92. 15. Demko ZP, Sharpless KB. An intramolecular [2 + 3] cycloaddition route to fused 5-heterosubstituted tetrazoles. Org Lett 2001;3:4091–4. 16. Mindt TL, Struthers H, Brans L, et al. ‘‘Click to chelate’’: synthesis and installation of metal chelates into biomolecules in a single step. J Am Chem Soc 2006;128:15096–7. 17. Glaser M, Arstad E. ‘‘Click labeling’’ with 2-[18F]fluoroethylazide for positron emission tomography. Bioconjug Chem 2007;18:989–93. 18. Martarello L, McConathy J, Camp VM, et al. Synthesis of syn- and anti-1-amino-3-[18F]fluoromethyl-cyclobutane-1-carboxylic acid (FMACBC), potential PET ligands for tumor detection. J Med Chem 2002;45:2250–9. 19. McConathy J, Martarello L, Malveaux EJ, et al. Radiolabeled amino acids for tumor imaging with PET: radiosynthesis and biological evaluation of 2-amino-3-[18F]fluoro-2-methylpropanoic acid and 3-[18F]fluoro-2-methyl-2-(methylamino)propanoic acid. J Med Chem 2002;45:2240–9. 20. McConathy J, Martarello L, Malveaux EJ, et al. Synthesis and evaluation of 2-amino-4-[18F]fluoro-2-methylbutanoic acid

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31. 32.

33.

34.

35.

36.

37.

38.

(FAMB): relationship of amino acid transport to tumor imaging properties of branched fluorinated amino acids. Nucl Med Biol 2003;30:477–90. Shoup TM, Olson J, Hoffman JM, et al. Synthesis and evaluation of [18F]1-amino-3-fluorocyclobutane-1-carboxylic acid to image brain tumors. J Nucl Med 1999;40:331–8. Yu W, McConathy J, Olson J, et al. Facile stereospecific synthesis and biological evaluation of (S)- and (R)-2-amino-2-methyl-4[123I]iodo-3-(E)-butenoic acid for brain tumor imaging with single photon emission computerized tomography. J Med Chem 2007;50: 6718–21. Makrides V, Bauer R, Weber W, et al. Preferred transport of O-(2[18F]fluoroethyl)-D-tyrosine (D-FET) into the porcine brain. Brain Res 2007;1147:25–33. Langen KJ, Hamacher K, Bauer D, et al. Preferred stereoselective transport of the D-isomer of cis-4-[18F]fluoro-proline at the bloodbrain barrier. J Cereb Blood Flow Metab 2005;25:607–16. Gazzola GC, Dall’Asta V, Franchi-Gazzola R, et al. The cluster-tray method for rapid measurement of solute fluxes in adherent cultured cells. Anal Biochem 1981;115:368–74. La Regina MC, Culbreth VO, Higashikubo R, et al. An alternative method to stereotactic inoculation of transplantable brain tumours in large numbers of rats. Lab Anim 2000;34:265–71. Shotwell MA, Kilberg MSOxender DL. The regulation of neutral amino acid transport in mammalian cells. Biochim Biophys Acta 1983;737:267–84. Christensen HN, Oxender DL, Liang M, et al. The use of Nmethylation to direct route of mediated transport of amino acids. J Biol Chem 1965;240:3609–16. Palacin M, Estevez R, Bertran J, et al. Molecular biology of mammalian plasma membrane amino acid transporters. Physiol Rev 1998;78:969–1054. Closs EI, Boissel JP, Habermeier A, et al. Structure and function of cationic amino acid transporters (CATs). J Membr Biol 2006;213: 67–77. Closs EI. Expression, regulation and function of carrier proteins for cationic amino acids. Curr Opin Nephrol Hypertens 2002;11:99–107. Jager PL, Vaalburg W, Pruim J, et al. Radiolabeled amino acids: basic aspects and clinical applications in oncology. J Nucl Med 2001;42:432–45. Wheatley DN. Arginine deprivation and metabolomics: important aspects of intermediary metabolism in relation to the differential sensitivity of normal and tumour cells. Semin Cancer Biol 2005;15: 247–53. Wu G, Bazer FW, Davis TA, et al. Arginine metabolism and nutrition in growth, health and disease. Amino Acids 2009;37: 153–68. Lee TS, Ahn SH, Moon BS, et al. Comparison of 18F-FDG, 18F-FET and 18F-FLT for differentiation between tumor and inflammation in rats. Nucl Med Biol 2009;36:681–6. Moon BS, Lee TS, Lee KC, et al. Syntheses of F-18 labeled fluoroalkyltyrosine derivatives and their biological evaluation in rat bearing 9L tumor. Bioorg Med Chem Lett 2007;17:200–4. Carvlin MJ, Rajan SS, Rosa L, et al. Efficacy of nonionic lowosmolar gadodiamide injection in animals with intracranial mass lesions. J Magn Reson Imaging 1992;2:19–24. Wilkins DE, Raaphorst GP, Saunders JK, et al. Correlation between Gd-enhanced MR imaging and histopathology in treated and untreated 9L rat brain tumors. Magn Reson Imaging 1995;13:89–96.