Synthesis and Evaluation of 18F-and 11C-Labeled Phenyl

Bioconjugate Chem. 2008, 19, 441–449

441

Synthesis and Evaluation of 18F- and 11C-Labeled Phenyl-Galactopyranosides as Potential Probes for in ViWo Visualization of LacZ Gene Expression using Positron Emission Tomography Sofie Celen,† Christophe Deroose,‡ Tjibbe de Groot,‡ Satish K. Chitneni,† Rik Gijsbers,§ Zeger Debyser,§ Luc Mortelmans,‡ Alfons Verbruggen,† and Guy Bormans*,† Laboratory for Radiopharmacy, Faculty of Pharmaceutical Sciences, and Department of Nuclear Medicine, U.Z. Gasthuisberg, and Division of Molecular Medicine, Katholieke Universiteit Leuven, Leuven, Belgium. Received June 18, 2007; Revised Manuscript Received October 22, 2007

3-Hydroxy-2-nitrophenyl 2,3,4,6-tetra-O-acetyl-β-D-galactopyranoside, a derivative of the chromogenic β-galactosidase (β-gal) substrate o-nitrophenyl β-D-galactopyranoside (ONPG) was synthesized using a Koenigs-Knorr glycosylation reaction. It was alkylated with 2-[18F]fluoroethyl triflate or [11C]methyl triflate, followed by deacetylation of the sugar hydroxyl groups to obtain radiolabeled 3-(2′-[18F]fluoroethoxy)-2-nitrophenyl β-Dgalactopyranoside ([18F]-2c) and 3-[11C]methoxy-2-nitrophenyl β-D-galactopyranoside ([11C]-3c), which were evaluated as potential reporter probes for in ViVo visualization of LacZ gene expression with positron emission tomography (PET). In Vitro, [18F]-2c and [11C]-3c were good substrates of β-gal and showed, respectively, a 7.5and 2.5-fold higher uptake into β-gal expressing cells (LacZ cells) compared to control cells. However, reversedphase HPLC analysis of the LacZ cell lysate and supernatant showed that labeled 3-(2′-[18F]fluoroethoxy)-2nitrophenol, the hydrolysis product formed by β-gal-mediated cleavage of [18F]-2c, substantially leaked out of the cells, which would lead to loss of PET signal. In a µPET study of [18F]-2c in a mouse with a β-gal expressing tumor, high retention was observed in liver and kidneys, but only negligible accumulation was seen in the tumor. As a general conclusion, it can be stated that the synthesized PET tracers [18F]-2c and [11C]-3c are not suitable for use as LacZ reporter probes. Further structural modifications to improve the diffusion over the tumor cell membrane and to increase retention in β-gal expressing cells may lead to more favorable in ViVo imaging probes.

INTRODUCTION Given progress in molecular biology and genetics as well as advances in imaging technology, it is possible to visualize cellular processes in space and time at a molecular or genetic level (1). One of the most studied molecular imaging targets to date is the expression of genes. In ViVo imaging of the location, the magnitude, and the time variation of gene and transgene expression is important in the development of transgenic animal models of disease, in trafficking and targeting of cells, and in gene therapy (2, 3). Many promising methods are being developed to image gene expression by using an appropriate combination of a reporter transgene and a reporter probe (4, 5). The most commonly used reporter systems include β-galactosidase (6–8), β-glucuronidase (6, 9), chloramphenicolacetyltransferase(6,10),greenfluorescentprotein(6,11,12), and luciferases (6, 13–15). Among these, the LacZ gene, which encodes E. coli β-galactosidase (β-gal), is one of the most attractive reporter genes due to its widespread use in different vectors, convenient in Vitro and histological detection methods, split protein systems, and its presence in many transgenic mouse models (16). Up to now, this reporter system has been frequently used for in Vitro evaluation of gene expression using substrates allowing colorimetric (17, 18), fluorescence (19–22), chemiluminescence (23), or bioluminescence (24) detection. All these optical imaging modalities are very useful in superficial tissues * Corresponding author. Address: Herestraat 49 bus 821 BE-3000 Leuven Belgium. Tel: +32 16 330447. Fax: +32 16 330449. E-mail: [email protected]. † Laboratory for Radiopharmacy, Faculty of Pharmaceutical Sciences. ‡ Department of Nuclear Medicine, U.Z. Gasthuisberg. § Division of Molecular Medicine.

and have extensive applications in small animals, but application to larger animals is limited by the depth of light penetration. Monitoring reporter gene expression in larger animals and ultimately in humans is possible with the use of specific radiotracers combined with nuclear imaging techniques like positron emission tomography (PET 2, 25) or single-photon emission computed tomography (SPECT (26)) and magnetic resonance imaging (MRI) in combination with paramagnetic reporter probes (27–29). The group of Mason et al. (30, 31) have been developing MRI probes by introducing a fluorine atom or a trifluoromethyl group in the colorimetric indicator ortho-nitrophenyl β-D-galactopyranoside (ONPG) and demonstrated the usefulness of these fluorogalactopyranosides to detect β-gal activity in Vitro with 19F NMR spectroscopy. Well-known and established PET and SPECT reporter systems are herpes simplex virus type 1 thymidine kinase (HSV1-tk) in combination with 18FHBG, human dopamine type 2 receptor (hD2R) in combination with [18F]fluoroethylspiperone, and human sodium iodide symporter (hNIS) in combination with 124I-, 123I-, or 99mTcO4- (4, 5). However, little progress has been made in the development of radioprobes that target LacZ gene expression. Lee et al. (32) described the use of a radiolabeled competitive inhibitor 2-(4-[125I/123I]iodophenyl) ethyl 1-thio-β-D-galactopyranoside to detect β-gal activity in mice bearing LacZ expressing tumors using SPECT. The disadvantage of this system is the low sensitivity using an inhibitor probe and the poor cell membrane permeability. Bormans et al. (33) synthesized 2′-[18F]fluorodeoxylactose (18FDL) using an enzymatic method starting from 2-[18F]fluoro2-deoxy-D-glucose (18FDG). The yield of the enzymatic synthesis was rather low, and a biodistribution study in Rosa-26

10.1021/bc700216d CCC: $40.75  2008 American Chemical Society Published on Web 01/08/2008

442 Bioconjugate Chem., Vol. 19, No. 2, 2008

mice compared to control mice showed a lack of tissue retention, probably due to the inability of 18FDL to cross the cell membrane. In our study, a derivative of ONPG was synthesized, labeled with 18F or 11C, and both radiolabeled agents were evaluated as potential reporter probes for in ViVo visualization of LacZ gene expression with PET.

MATERIALS AND METHODS General. All reagents and solvents were obtained commercially from Acros (Geel, Belgium), Aldrich, Fluka, Sigma (Sigma-Aldrich, Bornem, Belgium), Merck (Darmstadt, Germany), or Fischer Bioblock Scientific (Tournai, Belgium) unless otherwise specified. For ascending thin layer chromatography (TLC), precoated aluminum-backed plates (silica gel 60 with fluorescent indicator, 0.2 mm thickness; Macherey-Nagel, Düren, Germany) were used and developed using mixtures of ethyl acetate and hexane, CH2Cl2 and hexane, or acetonitrile as mobile phase. After evaporation of the solvent, compounds were detected under UV light (254 nm). The sugar compounds were additionally visualized by staining in oxidative medium (5% H2SO4 in ether), followed by heating. 1H and 13C NMR spectra were recorded on a Gemini 200 MHz spectrometer (Varian, Palo Alto, CA, USA) using CDCl3 or DMSO-d6 as solvent. Chemical shifts are reported in parts per million relative to tetramethylsilane (δ ) 0). Coupling constants are reported in hertz. Splitting patterns are defined by s (singlet), br s (broad singlet), d (doublet), dd (double–doublet), t (triplet), dt (double triplet), or m (multiplet). High-performance liquid chromatography (HPLC) analysis was performed on an XTerra RP C18 column (5 µm, 4.6 mm × 250 mm; Waters, Milford, USA) eluted with mixtures of water or 0.1 M ammonium acetate pH 4 and acetonitrile. The column effluent was monitored for absorbance with a UV-spectrometer (Waters 2487 Dual γ absorbance detector) set at 254 nm and for radioactivity using a 2 in. NaI(Tl) scintillation detector coupled to a single-channel analyzer (Medilab-Select, Mechelen, Belgium). Output of both detectors was analyzed using RaChel analysis software (Lablogic, Sheffield, UK). The radioactivity counting for biodistribution studies, log P determinations, and cell-uptake studies was done using an automated gamma counter equipped with a 3 in. NaI(Tl) well crystal coupled to a multichannel analyzer (Wallac 1480 Wizard 3″, Wallac, Turku, Finland). Exact mass measurements were performed on a time-of-flight spectrometer (LCT, Micromass, Manchester, UK) equipped with a standard electrospray ionization (ESI) interface. Samples were infused in acetonitrile/ water mixtures with a Harvard 22 syringe pump (Harvard Apparatus, Holliston, Massachusetts, USA). Accurate mass determination was done by coinfusion with a 10 µg/mL solution of ONPG in acetonitrile/water as an internal calibration mass in positive mode (ES+). Acquisition and processing of the data were done with Masslynx software (version 3.5, Waters). Melting points (mp) were determined using an IA9000 digital melting point apparatus (Electrothermal, Southend-on-Sea, England). The experiments in mice were carried out in compliance with the national laws relating to the conduct of animal experimentation and with approval of the local ethical committee for animal experiments. Synthesis.3-Hydroxy-2-nitrophenyl 2,3,4,6-tetra-O-acetyl-β-Dgalactopyranoside (1). To a solution of 0.42 g (2.7 mmol) 2-nitroresorcinol in a mixture of 10 mL quinoline and 5 mL CH2Cl2 was added 1 g (3.65 mmol) Ag2CO3 and 1 g (2.43 mmol) 2,3,4,6-tetra-O-acetyl-R-D-galactopyranosyl bromide. The flask was shielded from light, and the suspension was stirred for 3 h. The reaction mixture was poured into 50 mL methanol/ dichloromethane (1:9 v/v) followed by filtration over celite. The

Celen et al.

filtrate was successively washed with 50 mL HCl (1 M), 50 mL water, 50 mL saturated NaHCO3 solution, and 50 mL brine. After drying over MgSO4, the solution was filtered and the filtrate was concentrated in vacuo and applied on a silica gel column eluted with ethyl acetate/hexane (1:1 v/v). Yield: 0.37 g (0.76 mmol, 31%) of a yellow solid. 1 H NMR (CDCl3) δ 2.02, 2.07, 2.12, 2.19 (12H, 4× s, 4× CH3CO), 4.09 (1H, t, 3JH5-H6 ) 6 Hz, H-5), 4.17–4.28 (2H, m, 2× H-6), 5.10 (2H, d, 3JH1-H2 ) 8 Hz, H-1; dd, 3JH3-H2 ) 10.6 Hz, 3JH3-H4 ) 3 Hz, H-3), 5.48 (1H, d, 3JH4-H3 ) 3 Hz, H-4), 5.58 (1H, dd, 3JH2-H3 ) 10.6 Hz, 3JH2-H1 ) 8 Hz, H-2), 6.82 (1H, d, 3J ) 8.4 Hz, Har-4), 6.86 (1H, d, 3J ) 8.4 Hz, Har-6), 7.39 (1H, t, 3J ) 8.4 Hz, Har-5). 13 C NMR (CDCl3) δ 20.52, 61.28, 66.68, 67.80, 70.56, 71.42, 100.31 (anomeric C-1), 109.05, 113.73, 134.97, 154.64, 170.24. Exact Mass (ESI-MS) for C20H22NO13Na [M + Na]+: found 508.1048, calculated 508.1062. mp: 167–169 °C. 3-(2′-Fluoroethoxy)-2-nitrophenol (2a). A solution of 0.25 g (1.6 mmol) 2-nitroresorcinol, 0.094 mL (1.6 mmol) 2-fluoroethanol, and 0.42 g (1.6 mmol) triphenylphosphine in 15 mL tetrahydrofuran was stirred for 10 min at room temperature under an atmosphere of nitrogen, and 0.32 mL (1.6 mmol) diisopropylazodicarboxylate was added in one portion. After stirring for 4 h at room temperature under nitrogen, a small amount of silica gel was added, and the solvent was evaporated. The silica gel impregnated with the crude product was applied on a silica gel column that was eluted with a mixture of ethyl acetate/hexane (3:7 v/v), yielding 0.18 g (0.9 mmol, 56%) of a yellow solid. 1 H NMR (CDCl3) δ 4.26 (2H, dt, 3JH-F ) 26.8 Hz, 3J ) 4.2 Hz, -CH2CH2F), 4.74 (2H, dt, 2JH-F ) 46.8 Hz, 3J ) 4.2 Hz, -CH2CH2F), 6.47 (1H, d, 3J ) 8.4 Hz, Har-6), 6.68 (1H, d, 3J ) 8.4 Hz, Har-4), 7.32 (1H, t, 3J ) 8.4 Hz, Har-5), 10.1 (br s, OH). 13 C NMR (CDCl3) δ 70.05, 83.01, 105.02, 111.54, 135.39, 156.46. 3-(2′-Fluoroethoxy)-2-nitrophenyl 2,3,4,6-tetra-O-acetyl-βD-galactopyranoside (2b). The title compound was prepared from 3-(2′-fluoroethoxy)-2-nitrophenol (2a) using the same procedure as described for compound 1. Yield: 0.19 g (0.35 mmol, 31%) of a white solid. 1 H NMR (CDCl3) δ 2.01, 2.08, 2.13, 2.18 (12H, 4× s, 4× CH3CO), 4.1 (1H, t, 3JH5-H6 ) 6 Hz, H-5), 4.18–4.40 (4H, m, 2× H-6; -CH2CH2F), 4.72 (2H, dt, 2JH-F ) 46.8 Hz, 3J ) 4.2 Hz, -CH2CH2F), 5.02 (1H, d, 3JH1-H2 ) 8 Hz, H-1), 5.07 (1H, dd, 3JH3-H2 ) 10.6 Hz, 3JH3-H4 ) 3.2 Hz, H-3), 5.42–5.52 (2H, m, 3JH2-H3 ) 10.6 Hz, 3JH2-H1 ) 8 Hz, H-2; H-4), 6.8 (1H, d, 3J ) 8.4 Hz, Har-6), 6.93 (1H, d, 3J ) 8.4 Hz, Har-4), 7.34 (1H, t, 3J ) 8.4 Hz, Har-5). 13 C NMR (CDCl3) δ 20.39, 61.25, 66.65, 67.71, 68.74, 70.47, 71.36, 82.98, 100.88 (anomeric C-1), 108.84, 110.72, 131.09, 149.03, 150.69, 170.15. 3-(2′-Fluoroethoxy)-2-nitrophenyl β-D-galactopyranoside (2c). To a solution of 0.136 g (0.26 mmol) 3-(2′-fluoroethoxy)-2nitrophenyl 2,3,4,6-tetra-O-acetyl-β-D-galactopyranoside (2b) in 14 mL methanol was added 0.385 mL of a solution of 100 mg sodium in 20 mL methanol, and stirring was continued for 30 min. A small amount of silica gel was added, and the solvent was evaporated. The silica gel impregnated with the crude product was applied on a silica gel column that was eluted with acetonitrile, yielding 0.065 g (0.178 mmol, 70%) of a white solid. 1 H NMR (DMSO) δ 3.45–3.69 (6H, m, H-2; H-3 ; H-4; H-5; 2× H-6), 4.38 (2H, dt, 3JH-F ) 30.2 Hz, 3J ) 3.6 Hz, -CH2CH2F), 4.69 (2H, dt, 2JH-F ) 47.6 Hz, 3J ) 3.6 Hz, -CH2CH2F), 4.99 (1H, d, 3JH1-H2 ) 7.4 Hz, H-1), 6.95 (1H, d,

18

F/11C-Labeled Phenyl-Galactopyranosides as Potential Probes

J ) 8.4 Hz, Har-6), 6.98 (1H, d, 3J ) 8.4 Hz, Har-4), 7.45 (1H, t, 3J ) 8.4 Hz, Har-5). 13 C NMR (DMSO) δ 60.33, 68.1, 69.07, 70.04, 73.54, 75.9, 83.551, 101.09 (anomeric C-1), 107.07, 108.53, 131.63, 149.29, 150.12. Exact Mass (ESI-MS) for C14H17FNO9Na [M + Na]+: found 386.0844, calculated 386.0858. mp: 123–125 °C. 3-Methoxy-2-nitrophenol (3a). The same procedure was used as described for compound 2a, using methanol instead of 2-fluoroethanol. Yield 0.17 g (1 mmol, 62%) of a yellow solid. 1 H NMR (CDCl3) δ 3.95 (3H, s, -OCH3), 6.55 (1H, d, 3J ) 8.4 Hz, Har-6), 6.71 (1H, d, 3J ) 8.4 Hz, Har-4), 7.41 (1H, t, 3 J ) 8.4 Hz, Har-5), 10.22 (s, OH). 3-Methoxy-2-nitrophenyl 2,3,4,6-tetra-O-acetyl-β-D-galactopyranoside (3b). The title compound was prepared from 3-methoxy2-nitrophenol (3a) using the same procedure as described for compound 1. Yield: 0.13 g (0.26 mmol, 39%) of a yellow solid. 1 H NMR (CDCl3) δ 2.1 (12H, 4× s, 4× CH3CO), 3.82 (3H, s, -OCH3), 4.01 (1H, t, 3JH5-H6 ) 6.2 Hz, H-5), 4.1–4.24 (2H, m, 2× H-6), 4.92 (1H, d, 3JH1-H2 ) 8 Hz, H-1), 5.0 (1H, dd, 3 JH3-H2 ) 10.6 Hz, 3JH3-H4 ) 3.2 Hz, H-3), 5.36–5.45 (2H, m, 3 JH2-H3 ) 10.6 Hz, 3JH2-H1 ) 8 Hz, 3JH4-H3 ) 3.2 Hz, H-2; H-4), 6.71 (1H, d, 3J ) 8.4 Hz, Har-6), 6.82 (1H, d, 3J ) 8.4 Hz, Har-4), 7.27 (1H, t, 3J ) 8.4 Hz, Har-5). 13 C NMR (CDCl3) δ 20.55, 56.51, 61.28, 66.65, 67.65, 70.47, 71.29, 100.83 (anomeric C-1), 107.35, 109.78, 131.12, 151.73, 170.42. 3-Methoxy-2-nitrophenyl-β-D-galactopyranoside (3c). The same procedure was used as described for compound 2c, using 3-methoxy-2-nitrophenyl 2,3,4,6-tetra-O-acetyl-β-D-galactopyranoside (3b) as starting material. Yield 0.03 g (0.09 mmol, 45%) of a yellow solid. 1 H NMR (DMSO) δ 3.38–3.7 (6H, m, H-2; H-3; H-4; H-5; 2× H-6), 3.86 (3H, s, -OCH3), 4.58 (d, OH), 4.66 (t, OH), 4.87 (d, OH), 4.98 (1H, d, 3JH1-H2 ) 7.4 Hz, H-1), 5.11 (d, OH), 6.92 (1H, d, 3J ) 8.4 Hz, Har-6), 6.96 (1H, d, 3J ) 8.4 Hz, Har-4), 7.45 (1H, t, 3J ) 8.4 Hz, Har-5). 13 C NMR (DMSO) δ 56.81, 60.33, 68.13, 70.04, 73.54, 75.9, 101.06 (anomeric C-1), 106.07, 108.14, 131.69, 149.2, 150.21. Exact Mass (ESI-MS) for C13H16NO9Na [M + Na]+: found 354.0823, calculated 354.0796. mp: 130.5–132 °C. Radiolabeling. Production of [18F]fluoride and Radiosynthesis of [18F]-2c using 2-[18F]fluoroethyl Triflate ([18F]FEtOTf). [18F]F- was obtained using an 18O(p,n)18F reaction by irradiation of 0.5 mL of 97% enriched [18O]H2O (Rotem HYOX18, Rotem Industries, Beer Sheva, Israel) in a niobium target with 18 MeV protons from a Cyclone 18/9 cyclotron (Ion Beam Applications, Louvain-la-Neuve, Belgium). After the irradiation, [18F]F- was separated from [18O]H2O using a SepPak Light Accell plus QMA anion exchange cartridge (Waters). The [18F]F- was then eluted from the cartridge with a solution containing 2.47 mg potassium carbonate and 27.92 mg Kryptofix 222 dissolved in 0.75 mL of H2O/CH3CN (5:95 v/v) into a reaction vial. After evaporation with a stream of helium of the solvent from the reaction vial, [18F]F- was further dried by azeotropic distillation using 1 mL of dry acetonitrile. A volume of 5 µL 2-bromoethyl triflate in 0.7 mL o-dichlorobenzene was added to the vial containing [18F]F-. The resulting 2-[18F]fluoroethyl bromide was then distilled at 120 °C with a helium flow (3–4 mL/min) and subsequently passed over a silver triflate column (6 mm × 50 mm) at 250 °C to obtain [18F]FEtOTf. The resulting [18F]FEtOTf was bubbled into another reaction vial containing 1 mg of the precursor (1) and 5 µL NaOH 1 M in 0.3 mL acetone. After distillation of a sufficient amount of radioactivity in the solution, the reaction vial was closed and heated at 90 °C for 10 min. After cooling, deprotection of the sugar hydroxyl 3

Bioconjugate Chem., Vol. 19, No. 2, 2008 443

groups was done by addition of 0.3 mL NaOH 33 mM and incubation for 5 min at room temperature. The reaction mixture was subsequently neutralized with 0.5 mL 0.5 M phosphate buffer pH 5, and the mixture was heated at 90 °C for 4 min with a flow of helium to remove the unreacted [18F]FEtOTf or [18F]FEtOH which is formed as a side product. The resulting mixture was purified by HPLC on a µBondapak C18 column (10 µm, 7.8 mm × 300 mm; Waters), eluted with 0.05 M sodium acetate buffer pH 5.5/ethanol (85:15 v/v) at a flow rate of 4 mL/min. The radiolabeled compound [18F]-2c eluted after 6.35 min and was obtained with a decay-corrected radiochemical yield of approximately 59% relative to starting activity of [18F]FEtOTf. Radiochemical purity and specific activity were assayed using HPLC on an XTerra RP C18 column (5 µm, 4.6 mm × 250 mm; Waters) eluted with gradient mixtures of water/ acetonitrile (0 min, 90:10 v/v; 20 min, 30:70 v/v; linear gradient) at a flow rate of 1 mL/min. [18F]-2c elutes after 11 min and was obtained with an average specific activity of 64.6 GBq/ µmol at the end of synthesis (EOS) and a radiochemical purity of >99%. Production of [11C]methyl Triflate ([11C]MeOTf) and Radiosynthesis of [11C]-3c. Carbon-11 was produced by a 14 N(p,R)11C nuclear reaction. The target gas (a mixture of 95% N2 and 5% H2) was irradiated using 18 MeV protons at a beam current of 25 µA for about 30 min, to yield [11C]CH4. The [11C]CH4 was then reacted with vaporous I2 at 650 °C in a home-built recirculation synthesis module to convert it to [11C]MeI. Subsequently, [11C]MeOTf was obtained by passage of the [11C]MeI over a silver triflate column (6 mm × 50 mm) at 180 °C. The resulting [11C]MeOTf was bubbled through a solution of 1 mg precursor (1) in a mixture of 5 µL NaOH 1 M and 0.3 mL acetone. When the amount of radioactivity in the vial had stabilized, the sugar hydroxyl groups were deprotected by addition of 0.3 mL NaOH 33 mM and incubation for 5 min at room temperature. The reaction mixture was subsequently neutralized with 0.5 mL 0.5 M phosphate buffer pH 5 and purified by HPLC on a µBondapak C18 column (Waters), eluted with 0.1 M ammonium acetate pH 4/ethanol (90:10 v/v) at a flow rate of 4 mL/min. [11C]-3c eluted after 10.2 min and was obtained with a decay-corrected radiochemical yield of approximately 94% relative to [11C]MeOTf starting activity. The product was analyzed by HPLC on an XTerra RP C18 column (Waters) eluted with gradient mixtures of water/acetonitrile (0 min, 90:10 v/v; 20 min, 30:70 v/v; linear gradient) at a flow rate of 1 mL/min. [11C]-3c elutes after 10.5 min and was obtained with an average specific activity of 26.1 GBq/µmol at the end of synthesis (EOS) and a radiochemical purity of >99%. Log P (n-Octanol/Phosphate Buffer pH 7.4). The partition coefficient was determined using a modification of the method described by Vanbilloen et al. (34). An aliquot of 30 µL of RP-HPLC purified tracer solution containing 185 kBq of [18F]2c or 555 kBq [11C]-3c was added to a polypropylene tube (Sarstedt, Nümbrecht, Germany) containing 2 mL 0.025 M sodium phosphate buffer pH 7.4 and 2 mL n-octanol (density ) 0.827 g/mL). The tube was vortexed at room temperature for 2 min, followed by centrifugation at 3000 rpm (1837 g) for 5 min (Eppendorf centrifuge 5810, Eppendorf, Westbury, USA). Aliquots of 50 µL and 500 µL were drawn from the n-octanol and aqueous phases, respectively, taking care to avoid crosscontamination between the two phases. The samples were weighed, and radioactivity was counted using an automated gamma counter. The experiments were performed at least in triplicate. In Vitro Hydrolysis of [18F]-2c and [11C]-3c by β-gal. An aliquot of 0.4 mL HPLC purified [18F]-2c (5.9 MBq) or 0.4 mL HPLC purified [11C]-3c (9.6 MBq) was added to 0.6 mL of a solution containing HEPES buffer pH 7.0 (120 mM),


MgSO4 (4 mM), NaCl (180 mM), and 6 µg (1 U) β-gal (E. coli, E.C.3.2.1.23, 165 U/mg, Fluka). The solution was incubated at 37 °C in an oil bath, and 50 µL samples were collected at different time points ranging from 30 s to 45 min after the start of the incubation. Enzymatic activity was stopped by cooling the samples in ice. Each sample was analyzed for release of the radiolabeled aglycon using HPLC coupled to radiometric detection using an XTerra RP C18 column eluted with mixtures of 40% acetonitrile ([18F]-2c) or 30% acetonitrile ([11C]-3c) in 0.1 M ammonium acetate pH 4 at a flow rate of 1 mL/min. Intact [11C]-3c elutes at 4 min, the carbon-11 labeled aglycon 3-[11C]methoxy-2-nitrophenol ([11C]-3a) resulting from enzymatic hydrolysis elutes at 11.5 min. Intact [18F]-2c elutes at 3.5 min and the fluorine-18 labeled hydrolysis product 3-(2′[18F]fluoroethyl)-2-nitrophenol ([18F]-2a) at 8 min. The percentage of conversion of [18F]-2c and [11C]-3c as a function of the incubation period was calculated from the integrated HPLC chromatograms. Biodistribution Study in Normal Mice. Solutions of [18F]2c and [11C]-3c obtained after RP-HPLC purification were diluted using 0.9% NaCl for injection to concentrations of 3.7 MBq/mL and 37 MBq/mL, respectively. The biodistribution of both tracers was determined in male NMRI mice (body mass 30–40 g). The mice were anesthetized by intraperitoneal injection of 0.1 mL of a solution containing 3 mg ketamine and 0.225 mg xylazine. A volume of 0.1 mL of the diluted tracer solution was then injected via a tail vein, and the mice were sacrificed by decapitation at 2 or 60 min postinjection (p.i., n ) 3 per time point). All organs and body parts were dissected. Blood was collected in tared tubes, and the masses of all organs and body parts were determined. The radioactivity in each organ was counted using a 3 in. NaI(Tl) well counter, corrected for background radioactivity and expressed as a percentage of the injected dose (% ID) or as a percentage of the injected dose per gram tissue (% ID/g). For the calculation of total radioactivity in blood, blood mass was assumed to be 7% of the body mass. Metabolism of [18F]-2c and [11C]-3c in Normal Mice. The metabolic stability of [18F]-2c and [11C]-3c was studied in normal mice by determination of the relative amounts of the parent tracer and metabolites in plasma. An aliquot of 3.7 MBq of [18F]-2c or 18 MBq of [11C]-3c was injected intravenously via a tail vein of anesthetized mice. The animals (n ) 2 per tracer) were sacrificed by decapitation at 30 min p.i., and blood was collected into a BD vacutainer (containing lithium heparin; BD, Franklin Lakes, USA). After centrifugation at 3000 rpm (1837 g) for 5 min (Eppendorf centrifuge 5810), plasma was collected, diluted with an equal volume of acetonitrile, and again centrifuged for 5 min at 3000 rpm. The plasma was filtered through a 0.22 µm filter (Millipore, Bedford, USA), and the filter was rinsed with an equal volume of water. Finally, the filtered plasma was mixed with authentic 2a and 2c or authentic 3a and 3c and analyzed using HPLC on an XTerra RP C18 column eluted with mixtures of 40% acetonitrile ([18F]-2c) or 30% acetonitrile ([11C]-3c) in 0.1 M ammonium acetate pH 4 at a flow rate of 1 mL/min. After passing through an inline UV detector, the HPLC-eluate was collected in 1 mL fractions, and their radioactivity was measured using a gamma counter. Cell Uptake Studies. A lentiviral vector (LV) encoding the cDNA of β-gal linked to the puromycin-N-acetyl transferase (pac) gene from Streptomyces alboniger through an encephalomyocarditis virus internal ribosome entry site (IRES) was produced as described (35) and denominated LV-LacZ-I-P. A similar vector encoding the firefly luciferase (fluc) gene or the varicella zoster virus thymidine kinase (VZV-tk) was used as the control (LV-fluc-I-P and LV-VZV-tk-I-P).

Celen et al.

Human embryonic kidney cells (293T) transduced with LVLacZ-I-P and with LV-fluc-I-P were maintained in culture in medium containing 1 µg/mL puromycin. They were plated in triplicate at a density of 200 000 cells per well in 24-well plates. After 24 h, the medium was discarded, and 0.25 mL of fresh medium with HPLC-purified tracer (185 kBq of [18F]-2c or 1.11 MBq of [11C]-3c per well) was added. The cells were then incubated at 37 °C in a 5% CO2 atmosphere for time intervals of 30, 60, 90, or 120 min (n ) 3 per time point, per vector). Earlier time points were not investigated. Following incubation and removal of the medium, the cells were washed three times with 0.4 mL of ice-cold phosphate-buffered saline (PBS). The cells were then lysed with 0.25 mL of Cell Culture Lysis Reagent 1× solution (Promega Corporation, Madison, USA) for 10 min, after which the lysate was collected, followed by a 0.125 mL rinse using the same solution. The cell fractions (lysate and rinse) as well as the wash fractions (medium and PBS) were collected separately for each well, and the radioactivity was measured using a gamma counter. This procedure was repeated for each incubation period. The protein concentration of each cell fraction was determined using the Bio-Rad Protein Assay (Biorad, München, Germany) and a spectrophotometer (Bio Synchron-Anthos 2010, Anthos LabTec Instruments, Austria) at 595 nm. Tracer uptake was normalized for total protein content in the cell fraction for each individual well and expressed as percentage of total radioactivity per milligram protein. HPLC Analysis of Cell Lysates and Supernatants of LacZ Gene- and VZV-tk Gene-Transduced Cells after 30 min Incubation with [18F]-2c. Cells expressing β-gal or VZVTK as control were grown to near confluency in 75 cm2 flasks, in the conditions described above. The medium was discarded, and the monolayer was incubated with 4 mL of fresh medium containing 22 MBq of [18F]-2c for each flask. After 30 min incubation at 37 °C and removal of the medium, the cells were washed with cold PBS (3 × 5 mL). Intracellular components from the cells in both flasks were harvested by treatment with 1.5 mL of Cell Culture Lysis Reagent solution, followed by a rinse with 1 mL of the same solution. An aliquot of 1 mL of the cell suspension was diluted with an equal volume of acetonitrile, and the mixture was centrifuged at 3000 rpm (1837 g) for 5 min (Eppendorf centrifuge 5810). 1.5 mL of the centrifuged cell suspension was filtered through a 0.22 µM filter (Millipore), and the filter was rinsed with an equal volume of water. Finally, the filtrate was mixed with authentic 2a and 2c and analyzed using HPLC on an XTerra RP C18 eluted with a mixture of 30% acetonitrile in 0.1 M ammonium acetate pH 4 at a flow rate of 1 mL/min. After passing through an inline UV detector, the HPLC eluate was collected in 1 mL fractions, and the radioactivity of all fractions was measured using a gamma counter. The supernatant from both flasks was analyzed in the same way, with the only difference that these fractions were first centrifuged (prior to addition of acetonitrile) to remove the small number of cells that might have been detached from the bottom of the flask during the rinse steps. µPET Imaging in a β-gal Expressing Tumor Mouse Model. Five million SHSY5Y (human neuroblastoma) cells stably transduced with LV-LacZ-I-P were implanted in the flank of female nude mice. Tumor growth and body mass were monitored on a regular basis. The mice were housed on a 12 h/12 h light/darkness cycle and had access to food and water ad libitum. When the tumor reached a volume over 500 mm3, imaging was performed. Dynamic µPET imaging was performed with a Focus 220 micro-PET scanner (Concorde Microsystems, Knoxville, TN, USA). The mouse was anesthetized with isoflurane (2%) in oxygen at a flow rate of 1 L/min and subsequently injected with 13 MBq [18F]-2c via a lateral tail vein. A µPET image was acquired for 215 min, and reconstruc-

18



Figure 1. Structure of ONPG, 3-hydroxy-2-nitrophenyl 2,3,4,6-tetra-O-acetyl-β-D-galactopyranoside (1) and the radiolabeled probes 3-(2′[18F]fluoroethoxy)-2-nitrophenyl β-D-galactopyranoside ([18F]-2c) and 3-[11C]methoxy-2-nitrophenyl β-D-galactopyranoside ([11C]-3c). Scheme 1. Radiolabeling of Precursor 1 and Subsequent Deprotection Yielding [18F]-2c and [11C]-3c

tion was done using filtered back-projection with a RAMP 0.5 filter. The images were displayed in AMIDE (36).

RESULTS AND DISCUSSION β-Galactosidase (β-gal), the enzyme encoded by the LacZ gene, catalyzes hydrolysis of galactopyranosides by cleavage of the C-O bond between the sugar and the aglycon with overall retention of anomeric configuration (37, 38). As indicated by the diversity of substrates that can be used to detect β-gal activity, this enzyme has a broad substrate specificity. The only prerequisite is the conservation of the galactose structure, since it has been observed that hydrogen bonding between the active site of the enzyme and the hydroxyl groups of the glycosidic substrate is important in the formation of the enzyme–substrate complex (37–40). On the basis of these findings, we derivatized in the present study o-nitrophenyl β-D-galactopyranoside (ONPG) in the phenyl moiety in order to minimize interference with its binding affinity, labeled it using [18F]FEtOTf or [11C]MeOTf and evaluated the two tracers as potential reporter probes for in ViVo visualization of LacZ gene expression with PET. The structures of ONPG, the precursor 1, and the PET-tracers [18F]2c and [11C]-3c are shown in Figure 1. The ONPG-derived precursor 1 was synthesized following the procedure described by Oyama et al. (41), which involved a Koenigs-Knorr glycosylation reaction of 2-nitroresorcinol with acetobromo-R-D-galactose in the presence of silver carbonate in quinoline at room temperature. Inversion of the configuration at the anomeric center resulted in the sole formation of the β-anomer in a 30% yield. Compound 1 was identified using mass spectrometry, and its structure was confirmed with 1H and 13 C NMR. The presence of the anomeric H-1 at 5.1 ppm as a doublet with an axial-axial coupling (3JH1-H2) of 8 Hz, the coupling of H-2 with H-3 with a coupling constant (3JH2-H3) of 10.6 Hz in the 1H NMR spectrum, and the resonance of the anomeric carbon atom at 100.31 ppm in the 13C NMR spectrum proved the β-configuration of the precursor (42). To prevent alkylation of the sugar hydroxyl groups, hydrolysis of the protective acetate groups was performed after radiolabeling. Since the synthesis of the cold reference compounds by alkylation of 1 with 2-fluoroethyl tosylate or methyl triflate and subsequent deprotection failed or resulted in a negligible yield, we first synthesized the alkylated phenols (2a and 3a) via a Mitsunobu coupling between 2-nitroresorcinol and the corresponding alcohol (2-fluoroethanol or methanol) using triph-

enylphosphine and diisopropylazodicarboxylate in tetrahydrofuran. The structure of these compounds was confirmed with 1 H and 13C NMR. In a second step, the alkylated phenols were coupled to acetobromo-R-D-galactose in the same way as described for the synthesis of 1, yielding 2b and 3b in a moderate yield. Subsequent deprotection of the sugar hydroxyl groups using sodium methoxide resulted in the desired reference compounds 2c and 3c. The anomeric β-configuration, essential for substrate affinity for β-galactosidase, was again confirmed by the characteristic chemical shifts (2c, δH-1 ) 4.99 ppm, δC-1 ) 101.09 ppm; 3c, δH-1 ) 4.98 ppm, δC-1 ) 101.06 ppm) and coupling constants (3JH1-H2 ) 7.4 Hz) in the 1H and 13C NMR spectra. High-resolution mass spectrometry (HRMS) yielded the exact mass as calculated for the precursor (1) and the reference compounds (2c, 3c). Previous attempts to label a derivative of ONPG bearing a tosyl-ethoxy substituent in ortho position of the nitro group directly with [18F]fluoride failed because of decomposition of the precursor under the applied labeling conditions (Kryptofix 222 in combination with K2CO3 or KHCO3 at 110 or 60 °C). These failures prompted us to use the secondary labeling agents 2-[18F]fluoroethyl triflate ([18F]FEtOTf) and [11C]methyl triflate ([11C]MeOTf), which do not require the use of a strongly alkaline solution of kryptofix and a potassium salt at elevated temperatures. [18F]FEtOTf and [11C]MeOTf were synthesized following reported procedures (43, 44). Labeling with carbon11 was done by alkylation of precursor 1 with [11C]MeOTf in acetone in the presence of NaOH at room temperature. Labeling with fluorine-18 was done in a similar way with [18F]FEtOTf as alkylating agent. However, heating to 90 °C was required to effect the reaction. At 70 °C, slightly lower yields were obtained, and at room temperature, less than 20% of the protected tracer was formed (Scheme 1). For both tracers, labeling was followed by deprotection of the sugar hydroxyl groups under basic conditions. For the fluorine-18 labeled tracer, the unreacted [18F]EtOTf or [18F]EtOH was evaporated, and both tracers were purified using preparative RP-HPLC. Chemical and radiochemical purity of both HPLC purified tracers was examined on an analytical RP C18 column and were found to be >99%. The identity of the tracers was confirmed by coelution with authentic nonradioactive compounds after coinjection on analytical RPHPLC. The lipophilicity of the tracers was determined by measuring their n-octanol/water partition coefficients, which were found


Celen et al.

Scheme 2. Schematic Representation of Enzymatic Cleavage of the Galactopyranosides into Their Corresponding Phenol-Aglycons and D-Galactose

to be -0.8 and -0.7, respectively, for [18F]-2c and [11C]-3c. This reflects the polarity of the sugar backbone. The optimal theoretical log P value of a compound for passive diffusion over the blood-brain barrier is situated within the range 1–2.5 (45, 46). Extrapolating this to diffusion over a cell membrane suggests that the labeled galactose derivatives may not be lipophilic enough to efficiently enter the cells through passive diffusion. In literature, no evidence of active transport for this type of galactose derivative was found. However, the cell uptake studies (Vide infra) suggest that [18F]-2c and [11C]-3c do enter the cells. In an in Vitro test (Scheme 2), both tracers were found to be good substrates of β-gal with results in a comparable range as reported by Cui et al. (30) for a similar compound (Figure 2). After 30 s of incubation at 37 °C, already 89% of [18F]-2c was hydrolyzed. Complete hydrolysis was reached after about 15 min. For the C-11 labeled tracer, the enzymatic conversion was somewhat slower. After 30 s of incubation, 60% of [11C]-3c was hydrolyzed, and almost complete hydrolysis (90%) was reached after 15 min. In a similar test with the reference substrate ONPG (1 mg), 100% conversion was also reached within 15 min (results not shown). In a control experiment under identical conditions, but in the absence of β-gal, no hydrolysis of the labeled galactose derivatives was observed. These results suggest that the labeled galactose derivatives were specifically hydrolyzed by β-gal, with a rate comparable to that for ONPG. The biodistribution of [18F]-2c and [11C]-3c was evaluated in normal male NMRI mice at 2 and 60 min p.i. for both tracers. Data are shown in Tables 1 and 2. Both tracers have a similar behavior in ViVo, but the clearance from blood, defined as the 2-min-to-60-min ratio of the blood activity, was slower for [18F]2c than for [11C]-3c (1.7 and 4.7, respectively). Blood clearance occurred mainly through the renal pathway with 33.2% of the injected dose (ID) ([18F]-2c) and 45.8% ID ([11C]-3c) in the urine at 60 min p.i. Negligible uptake was seen in the brain, which was expected in view of the negative log P value. The slow blood clearance may interfere with successful in ViVo imaging.

Figure 2. In Vitro hydrolysis of [18F]-2c (() and [11C]-3c (O) by β-gal (1U) in HEPES buffered solution at 37 °C.

Table 1. Biodistribution of [18F]-2c in Normal NMRI Mice 2 and 60 min p.i.a % IDb 18

% ID/gc

[ F]-2c

2 min

60 min

2 min

60 min

urine kidneys liver intestines spleen + pancreas lungs heart stomach brain blood

0.3 ( 0.1 10.1 ( 1.1 20.0 ( 2.3 8.9 ( 0.4 0.9 ( 0.1 1.5 ( 0.0 0.5 ( 0.1 1.1 ( 0.2 0.1 ( 0.0 21.4 ( 2.2

33.2 ( 0.9 3.3 ( 1.2 5.1 ( 0.7 14.3 ( 0.9 0.5 ( 0.0 0.5 ( 0.0 0.4 ( 0.0 0.6 ( 0.1 0.4 ( 0.0 11.6 ( 3.0

21.3 ( 3.6 10.7 ( 0.9 3.0 ( 0.3 3.6 ( 0.1 7.2 ( 0.6 5.2 ( 0.6 2.1 ( 0.7 0.5 ( 0.1 8.0 ( 0.8

7.0 ( 2.8 2.9 ( 0.1 4.9 ( 0.7 2.2 ( 0.3 3.0 ( 0.3 4.6 ( 0.5 1.1 ( 0.3 1.9 ( 0.2 4.6 ( 1.4

a Data are expressed as mean ( SD; n ) 3 per time point; p.i. ) post injection. b Percentage of injected dose calculated as cpm in organ/ total cpm recovered. c Percentage of injected dose per gram tissue.

Table 2. Biodistribution of [11C]-3c in Normal NMRI Mice 2 and 60 min p.i.a % IDb 11

% ID/gc

[ C]-3c

2 min

60 min

2 min

60 min

urine kidneys liver intestines spleen + pancreas lungs heart stomach brain blood

0.3 ( 0.2 7.8 ( 1.6 22.0 ( 1.5 10.1 ( 1.0 1.0 ( 0.2 1.4 ( 0.1 0.6 ( 0.1 1.0 ( 0.1 0.1 ( 0.0 19.5 ( 2.5

45.8 ( 7.9 3.5 ( 2.2 14.9 ( 2.6 10.2 ( 2.2 0.6 ( 0.1 0.3 ( 0.0 0.2 ( 0.0 0.6 ( 0.1 0.1 ( 0.0 4.3 ( 0.8

14.4 ( 4.5 9.9 ( 1.3 3.0 ( 0.4 3.2 ( 0.7 5.9 ( 0.2 3.6 ( 0.4 1.3 ( 0.1 0.3 ( 0.0 7.4 ( 0.9

5.8 ( 3.7 6.8 ( 0.9 3.0 ( 0.9 1.9 ( 0.2 1.2 ( 0.2 1.2 ( 0.2 0.7 ( 0.1 0.4 ( 0.1 1.6 ( 0.4

a Data are expressed as mean ( SD; n ) 3 per time point; p.i. ) post injection. b Percentage of injected dose calculated as cpm in organ/ total cpm recovered. c Percentage of injected dose per gram tissue.

The metabolic stability was determined in normal mice, having low levels of endogenous β-galactosidase (47). Plasma analysis of [18F]-2c and [11C]-3c at 30 min p.i. revealed that, respectively, 98% and 97% of the activity was present as intact tracer, indicating high in ViVo stability. Negligible traces of hydrolysis product were found, indicating little degradation by endogenous mammalian β-galactosidase. The results of the cell uptake studies are shown in Figure 3a ([18F]-2c),b ([11C]-3c). A maximal uptake was observed after 30 min incubation for [18F]-2c with a 7.5-fold higher uptake in the β-gal-expressing cells (3.2%/mg protein) compared to the control cells (0.4%/mg protein). This uptake ratio is comparable with the maximal uptake ratio of the iodine-123 labeled β-gal inhibitor of the group of Lee et al. (32). For [11C]-3c, the maximal uptake was reached after 90 min of incubation with a 2.5-fold higher uptake in the β-gal expressing cells (5.2%/mg protein) compared to the control cells (2.1%/mg protein). [18F]-2c was further evaluated in Vitro and in ViVo. The uptake of [18F]-2c in the β-gal-expressing cells decreases significantly as a function of time, from 3.2%/mg protein after

18


Figure 3. a. Time-dependent uptake levels of [18F]-2c in LacZ gene (()- and fluc gene (O)-transduced 293T cells. Data are expressed as % tracer/mg protein and are mean values of triplicate samples. Values with asterisk are significant different; p < 0.006, unpaired bidirectional Student t test. b. Time-dependent uptake levels of [11C]-3c in LacZ gene (()- and fluc gene (O)-transduced 293T cells. Data are expressed as % tracer/mg protein and are mean values of triplicate samples. Values with asterisk are significant different; p < 0.005, unpaired bidirectional Student t test.

30 min of incubation to 1.5%/mg protein after 120 min of incubation (Figure 3a). This may be due to leakage of the radiolabeled aglycon ([18F]-2a) out of the transduced cells. To test this hypothesis and to check the nature of the metabolites present in the transduced cells, the cell lysates and supernatant fractions of both the β-gal expressing cells as well as the control cells were analyzed using RP-HPLC, after incubation with [18F]2c for 30 min. The HPLC chromatograms of the supernatants and cell lysates of the LV-LacZ-I-P and LV-VZV-tk-I-P transduced cells are shown in Figure 4a (supernatants) and b (cell lysates). The cold intact (2c) and hydrolyzed (2a) reference compounds were added to the cell lysates and supernatant fractions prior to HPLC analysis. In this way, the signals in the reconstructed radiometric HPLC chromatogram could be assigned by correlation to the signals in the UV absorption chromatogram. The component eluting in fractions 4 to 6 is the intact tracer [18F]-2c. The metabolite eluting in fractions 14 to 16 is the aglycon 3-(2′-[18F]fluoroethoxy)-2-nitrophenol ([18F]-2a). In the LacZ cell lysate, 63.5% of the radioactivity was recovered as radiolabeled aglycon (Table 3). In the VZVtk control cells, only negligible amounts of hydrolysis product were found in the cells and in the supernatants. This suggests that the labeled aglycon (50.7%) found in the supernatant fraction of the β-gal expressing cells must have leaked out of the LacZ expressing cells, which would result in loss of PETsignal. This back-diffusion can be explained in terms of the pKa of the hydrolysis product ([18F]-2a). The pKa of nonsubstituted o-nitrophenol is 7.23 (48); the dissociation constant of [18F]-2a is therefore estimated to be in a similar range. Ideally, the dissociation constant should be an order of magnitude lower than 7.4 (physiological pH) to direct the equilibrium to the dissociated form, resulting in a higher retention of the ionized aglycon in LacZ transduced cells. To test the potential of [18F]-2c as a PET reporter probe for detecting LacZ gene expression in ViVo, a mouse growing


Figure 4. a. Analysis of the supernatants of LacZ gene- and VZV-tk gene-transduced 293T cells after 30 min incubation with [18F]-2c. b. Analysis of the cell lysates of LacZ gene- and VZV-tk gene-transduced 293T cells after 30 min incubation with [18F]-2c. Table 3. Results of the RP-HPLC Analysis of the Cell Lysate and Supernatant Fractions of LacZ Gene- and VZV-tk Gene-Transduced Cells Recovered after 30 min Incubation with [18F]-2c cell lysatea [18F]-2c (intact) [18F]-2a (aglycon) a

supernatanta

LacZ

VZV-tk

LacZ

VZV-tk

36.5 63.5

97.8 2.2

49.3 50.7

94.4 5.6

Data are expressed as % of recovered activity.

Figure 5. µPET image, selected from a 3D data set (summed image from 0 to 215 min p.i.), of the distribution of [18F]-2c in a β-galexpressing tumor mouse model. The position of the tumor is marked with a cross.

β-galactosidase-expressing SHSY5Y-LacZ-I-P xenografts was injected with [18F]-2c and a µPET image was acquired for 215 min. Figure 5 shows the summed image from 0 to 215 min. In analogy with the mice biodistribution study, high uptake is seen in the liver and the kidneys. Only negligible accumulation is observed in the LacZ-transduced tumor, despite the pronounced difference in uptake between β-gal expressing cells and control cells in Vitro. The absence of tracer accumulation in the LacZ tumor can be due to poor passive diffusion over the tumor cell membrane. This would be in accordance with the results of Liu


et al. (49) who needed to perform direct intratumoral injection of their fluoronitrophenyl galactopyranoside MRI probe to observe any contrast.

CONCLUSION A derivative of the chromogenic β-gal substrate ONPG was efficiently synthesized and labeled with [18F]FEtOTf or [11C]MeOTf to obtain [18F]-2c and [11C]-3c, which were evaluated as reporter probes for detection of LacZ gene expression with PET. Metabolic stability analysis of plasma collected 30 min after injection of [18F]-2c or [11C]-3c in mice showed high in ViVo stability for both tracers. In an in Vitro experiment, [18F]-2c and [11C]-3c were found to be good substrates with complete conversion after about 15 min of incubation with β-galactosidase. Cell uptake studies revealed that [18F]-2c had a 7.5-fold higher uptake into LacZ genetransduced cells compared to control cells. For [11C]-3c, this difference was only 2.5. RP-HPLC analysis of the cell lysates and supernatant fractions from β-gal expressing cells and control cells after 30 min of incubation with [18F]-2c showed substantial leakage of the β-gal cleavage product of [18F]-2c out of the transduced cells, which would lead to loss of PET signal. In a µPET study of [18F]-2c in a LacZ expressing tumor mouse model, only negligible accumulation was observed in the LacZ transduced tumor, despite the significant difference in uptake between β-gal expressing cells and control cells in Vitro. As a general conclusion, it can be stated that the synthesized PET tracers [18F]-2c and [11C]-3c are not suitable for use as LacZ reporter probes. Further modifications to improve diffusion over the tumor cell membrane and to increase the retention in β-gal expressing cells may lead to better in ViVo imaging results.

ACKNOWLEDGMENT We gratefully thank Marva Bex and Sigrid Ruymen for their assistance in the labeling experiments and Christelle Terwinghe, Kim Deliège, and Peter Vermaelen for their help in the in ViVo studies. This work was supported by SBO grant (IWT-30 238) of the Flemisch Institute supporting Scientific-Technological Research in industry (IWT), the IDO grant (IDO/02/012) of the Katholieke Universiteit Leuven, and by the EC-FP6-project DiMI, LSHB-CT-2005-512146.

LITERATURE CITED (1) Doubrovin, M., Serganova, I., Mayer-Kuckuk, P., Ponomarev, V., and Blasberg, R. G. (2004) Multimodality in ViVo moleculargenetic imaging. Bioconjugate Chem. 15, 1376–1388. (2) Herschman, H. R. (2004) PET reporter genes for noninvasive imaging of gene therapy, cell tracking and transgenic analysis. Crit. ReV. Oncol. Hematol. 51, 191–204. (3) Acton, P. D., and Zhou, R. (2005) Imaging reporter genes for cell tracking with PET and SPECT. Q. J. Nucl. Med. Mol. Imaging 49, 349–360. (4) Gambhir, S. S., Barrio, J. R., Herschman, H. R., and Phelps, M. E. (1999) Assays for noninvasive imaging of reporter gene expression. Nucl. Med. Biol. 26, 481–490. (5) Serganova, I., and Blasberg, R. (2005) Reporter gene imaging: potential impact on therapy. Nucl. Med. Biol. 32, 763–780. (6) Wood, K. V. (1995) Marker proteins for gene expression. Curr. Opin. Biotechnol. 6, 50–58. (7) Lin, W. C., Pretlow, T. P., Pretlow, T. G., and Culp, L. A. (1990) Bacterial lacZ gene as a highly sensitive marker to detect micrometastasis formation during tumor progression. Cancer Res. 50, 2808–2817. (8) Cui, C., Wani, M. A., Wight, D., Kopchick, J., and Stambrook, P. J. (1994) Reporter genes in transgenic mice. Transgenic Res. 3, 182–194.

Celen et al. (9) Jefferson, R. A., Bevan, M., and Kavanagh, T. (1987) The use of the Escherichia coli β-glucuronidase gene as a gene fusion marker for studies of gene expression in higher plants. Biochem. Soc. Trans. 15, 17–18. (10) Gorman, C. M., Moffat, L. F., and Howard, B. H. (1982) Recombinant genomes which express chloramphenicol acetyltransferase in mammalian cells. Mol. Cell. Biol. 2, 1044–1051. (11) Chalfie, M., Tu, Y., Euskirchen, G., Ward, W. W., and Prasher, D. C. (1994) Green fluorescent protein as a marker for gene expression. Science 263, 802–805. (12) Hoffman, R. M. (2002) Green fluorescent protein imaging of tumour growth, metastasis, and angiogenesis in mouse models. Lancet Oncol. 3, 546–556. (13) O’Connell-Rodwell, C. E., Burns, S. M., Bachmann, M. H., and Contag, C. H. (2002) Bioluminescent indicators for in ViVo measurements of gene expression. Trends Biotechnol. 20, S19– S24. (14) Rehemtulla, A., Stegman, L. D., Cardozo, S. J., Gupta, S., Hall, D. E., Contag, C. H., and Ross, B. D. (2000) Rapid and quantitative assessment of cancer treatment response using in ViVo bioluminescence imaging. Neoplasia 2, 491–495. (15) Williams, T. M., Burlein, J. E., Ogden, S., Kricka, L. J., and Kant, J. A. (1989) Advantages of firefly luciferase as a reporter gene: application to the interleukine-2 gene promoter. Anal. Biochem. 176, 28–32. (16) Rossi, F., Charlton, C. A., and Blau, H. M. (1997) Monitoring protein–protein interactions in intact eukaryotic cells by β-galactosidase complementation. Proc. Natl. Acad. Sci. U.S.A. 94, 8405–8410. (17) Sanchez-Ramos, J., Song, S., Dailey, M., Cardozo-Pelaez, F., Hazzi, C., Stedeford, T., Willing, A., Freeman, T. B., Saporta, S., Zigova, T., Sanberg, P. R., and Snyder, E. Y. (2000) The X-gal caution in neural transplantation studies. Cell Transplant. 9, 657–667. (18) Aamlid, K. H., Lee, G., Smith, B. V., Richardson, A. C., and Price, R. G. (1990) New colorimetric substrates for the assay of glycosidases. Carbohydr. Res. 205, c5–c9. (19) Tung, C.-H., Zeng, Q., Shah, K., Kim, D.-E., Schellingerhout, D., and Weissleder, R. (2004) In ViVo imaging of β-galactosidase activity using far red fluorescent switch. Cancer Res. 64, 1579– 1583. (20) Rotman, B., Zderic, J. A., and Edelstein, M. (1963) Fluorogenic substrates for β-D-galactosidases and phosphatases derived from fluorescein (3,6-dihydroxyfluoran) and its monomethyl ether. Proc. Natl. Acad. Sci. U.S.A. 50, 1–6. (21) Nolan, G. P., Fiering, S., Nicolas, J.-F., and Herzenberg, L. A. (1988) Fluorescence activated cell-analysis and sorting of viable mammalian cells based on β-D-galactosidase activity after transduction of Escherichia coli lacZ. Proc. Natl. Acad. Sci. U.S.A. 85, 2603–2607. (22) Zhang, Y.-Z., Naleway, J. J., Larison, K. D., Huang, Z., and Haugland, R. P. (1991) Detecting lacZ gene expression in living cells with new lipophilic, fluorogenic β-galactosidase substrates. FASEB J. 5, 3108–3113. (23) Jain, V. K., and Magrath, I. T. (1991) A chemiluminescent assay for quantitation of β-galactosidase in the femtogram range. Application to quantitation of β-galactosidase in lacZ-transfected cells. Anal. Biochem. 199, 119–124. (24) Wehrman, T. S., Von Degenfeld, G., Krutzik, P. O., Nolan, G. P., and Blau, H. M. (2006) Luminescent imaging of β-galactosidase activity in living subjects using sequential reporter-enzyme luminescence. Nat. Methods 3, 295–301. (25) Gambhir, S. S. (2002) Molecular imaging of cancer with positron emission tomography. Nat. ReV. Cancer 2, 683 693. (26) Bailey, D. L., and Parker, J. A. (2004) Single photon emission computed tomography, Nuclear Medicine in Clinical Diagnosis and Treatment, 3rd ed., Vol. 2 (Ell, P. J., and Gambhir, S. S., Eds.) pp 1815–1826,Chapter 128, Churchill Livingstone, Edinburgh, GB.

18


(27) Bogdanov, A., and Weissleder, R. (1998) The development of in ViVo imaging systems to study gene expression. Trends Biotechnol. 16 (1), 5–10. (28) Weissleder, R., Moore, A., Mahmood, U., Bhorade, R., Benveniste, H., Chiocca, E. A., and Basilion, J. P. (2000) In ViVo magnetic resonance imaging of transgene expression. Nat. Med. 6, 351–354. (29) Louie, A. Y., Hüber, M. M., Ahrens, E. T., Rothbächer, U., Moats, R., Jacobs, R. E., Fraser, S. E., and Meade, T. J. (2000) In vivo visualization of gene expression using magnetic resonance imaging. Nat. Biotechnol. 18, 18321–325. (30) Cui, W., Otten, P., Li, Y., Koeneman, K. S., Yu, J., and Mason, R. P. (2004) Novel NMR approach to assessing gene transfection: 4-fluoro-2-nitrophenyl-β-D-galactopyranoside as a prototype reporter molecule for β-galactosidase. Magn. Res. Med. 51, 616–620. (31) Yu, J., Liu, L., Kodibagkar, V. D., Cui, W., and Mason, R. P. (2006) Synthesis and evaluation of novel enhanced gene reporter molecules: Detection of β-galactosidase activity using 19F NMR of trifluoromethylated aryl β-D-galactopyranosides. Bioorg. Med. Chem. 14, 326–333. (32) Lee, K.-H., Byun, S. S., Choi, J. H., Paik, J.-Y., Choe, Y. S., and Kim, B.-T. (2004) Targeting of LacZ reporter gene expression with radioiodone-labelled phenylethyl-β-D-thiogalactopyranoside. Eur. J. Nucl. Med. Mol. Imaging 31, 433–438. (33) Bormans, G., and Verbruggen, A. (2001) Enzymatic synthesis and biodistribution in mice of β-O-D-galactopyranosyl-(1,4′)2′-[18F]fluoro-2′-deoxy-D-glucopyranose (2′-[18F]fluorodeoxylactose). J. Labelled Cmpd. Radiopharm. 44, 417–423. (34) Vanbilloen, H. P., Kieffer, D., Cleynhens, B. J., Bormans, G., Mortelmans, L., and Verbruggen, A. M. (2005) Development and biological evaluation of 99mTc-BAT-tropane esters. Nucl. Med. Biol. 32, 607–612. (35) Geraerts, M., Michiels, M., Baekelandt, V., Debyser, Z., and Gijsbers, R. (2005) Upscaling of lentiviral vector production by tangential flow filtration. J. Gene Med. 7, 1299– 1310. (36) Andreas Loening, Crump Institute for Molecular Imaging, UCLA School of Medicine, CA, USA, http://sourceforge.net/ projects/amide/. (37) Withers, S. G., and Aebersold, R. (1995) Approaches to labeling and identification of active site residues in glycosidases. Protein Sci. 4, 361–372. (38) McCarter, J. D., Adam, M. J., and Withers, S. G. (1992) Binding energy and catalysis. Fluorinated and deoxygenated glycosides as mechanistic probes of Escherichia coli (LacZ) β-galactosidase. Biochem. J. 286, 721–727.

Bioconjugate Chem., Vol. 19, No. 2, 2008 449 (39) Namchuk, M. N., and Withers, S. G. (1995) Mechanism of Agrobacterium β-glucosidase: Kinetic analysis of the role of noncovalent enzyme/substrate interactions. Biochemistry. 34, 16194–16202. (40) Yoon, S., Kim, H. G., Chun, K. H., and Shin, J. E. N. (1996) 4-Deoxy-analogs of p-nitrophenyl β-D-galactopyranosides for specificity study with β-galactosidase from Escherichia coli. Bull. Korean Chem Soc. 17, 599–604. (41) Oyama, K. I., and Kondo, T. (2004) Total synthesis of apigenin 7,4′-di-O-β-glucopyranoside, a component of blue flower pigment of Salvia patens, and seven chiral analogues. Tetrahedron. 60, 2025–2034. (42) Yu, J., Otten, P., Ma, Z., Cui, W., Liu, L., and Mason, R. P. (2004) Novel NMR approach to assessing gene transfection: Synthesis and evaluation of fluorinated phenyl-β-D-galactosides with potential application for assessing LacZ gene expression. Bioconjugate Chem. 15, 1334–1341. (43) Zhang, M. R., Furutsuka, K., Yoshida, Y., and Suzuki, K. (2003) How to increase the reactivity of [F-18]fluoroethyl bromide: [F-18]fluoroethylation of amine, phenol and amide functional groups with [F-18]FEtBr, [F-18]FEtBr/NaI and [F-18]FEtOTf. J. Labelled Cmpd. Radiopharm. 46, 587–598. (44) Jewett, D. M. (1992) A simple synthesis of [C-11]methyl triflate. Appl. Radiat. Isot. 43, 1383–1385. (45) Dishino, D. D., Welch, M. J., Kilbourn, M. R., and Raichle, M. E. (1983) Relationship between lipophilicity and brain extraction of C-11-labeled radiopharmaceuticals. J. Nucl. Med. 24, 1030–1038. (46) Young, R. C., Mitchell, R. C., Brown, T. H., Ganellin, C. R., Griffiths, R., Jones, M., Rana, K. K., Saunders, D., Smith, I. R., Sore, N. E., and Wilks, T. J. (1998) Development of a new physicochemical model for brain penetration and its application to the design of centrally acting H-2-receptor histamineantagonists. J. Med. Chem. 31, 656–671. (47) Weiss, D. J., Liggitt, D., and Clark, J. G. (1999) Histochemical discrimination of endogenous mammalian β-galactosidase activity from that resulting from lac-Z gene expression. Histochem. J. 31, 231–236. (48) pKa data compiled by Williams, R. http://research.chem. psu.edu/brpgroup/pKa_compilation.pdf. (49) Liu, L., Kodibagkar, V. D., Yu, J.-X., and Mason, R. P. (2007) 19 F-NMR detection of lacZ gene expression via the enzymatic hydrolysis of 2-fluoro-4-nitrophenyl β-D-galactopyranoside in ViVo in PC3 prostate tumor xenografts in the mouse. FASEB J. 21, 2014–2019. BC700216D

Synthesis and Evaluation of 18F-and 11C-Labeled Phenyl

Recommend Documents