Bioconjugate Chem. 2004, 15, 61−69
61
[18F]Galacto-RGD: Synthesis, Radiolabeling, Metabolic Stability, and Radiation Dose Estimates Roland Haubner,*,‡ Bertrand Kuhnast,‡ Christian Mang,§ Wolfgang A. Weber,‡ Horst Kessler,§ Hans-Ju¨rgen Wester,‡ and Markus Schwaiger‡ Department of Nuclear Medicine, Technische Universita¨t Mu¨nchen, 81675 Mu¨nchen, Germany, and Institute of Organic Chemistry and Biochemistry, Technische Universita¨t Mu¨nchen, 85747 Garching, Germany . Received September 18, 2003
It has been demonstrated in various murine tumor models that radiolabeled RGD-peptides can be used for noninvasive determination of Rvβ3 integrin expression. Introduction of sugar moieties improved the pharmacokinetic properties of these peptides and led to tracer with good tumor-tobackground ratios. Here we describe the synthesis, radiolabeling, and the metabolic stability of a glycosylated RGD-peptide ([18F]Galacto-RGD) and give first radiation dose estimates for this tracer. The peptide was assembled on a solid support using Fmoc-protocols and cyclized under high dilution conditions. It was conjugated with a sugar amino acid, which can be synthesized via a four-step synthesis starting from pentaacetyl-protected galactose. For radiolabeling of the glycopeptide, 4-nitrophenyl-2-[18F]fluoropropionate was used. This prosthetic group allowed synthesis of [18F]GalactoRGD with a maximum decay-corrected radiochemical yield of up to 85% and radiochemical purity >98%. The overall radiochemical yield was 29 ( 5% with a total reaction time including final HPLC preparation of 200 ( 18 min. The metabolic stability of [18F]Galacto-RGD was determined in mouse blood and liver, kidney, and tumor homogenates 2 h after tracer injection. The average fraction of intact tracer in these organs was approximately 87%, 76%, 69%, and 87%, respectively, indicating high in vivo stability of the radiolabeled glycopeptide. The expected radiation dose to humans after injection of [18F]Galacto-RGD has been estimated on the basis of dynamic PET studies with New Zealand white rabbits. According to the residence times in these animals the effective dose was calculated using the MIRDOSE 3.0 program as 2.2 × 10-2 mGy/MBq. In conclusion, [18F]GalactoRGD can be synthesized in high radiochemical yields and radiochemical purity. Despite the timeconsuming synthesis of the prosthetic group 185 MBq of [18F]Galacto-RGD, a sufficient dose for patient studies, can be produced starting with approximately 2.2 GBq of [18F]flouride. Moreover, the fast excretion, the suitable metabolic stability and the low estimated radiation dose allow to evaluate this tracer in human studies.
INTRODUCTION
Integrins are heterodimeric glycoproteins consisting of an R- and a β-subunit, which are involved in cell/cell and cell/matrix interactions (1). They possess large extracellular and short cytoplasmatic domains. The C-terminal end of the β-subunit is connected via cytoplasmatic proteins such as talin, vinculin, and R-actinin with the actin filament of the cytoskeleton. This connection allows inside out as well as outside in signaling. Thus, integrins are not only adhesion receptors linking cells with proteins of the extracellular matrix, but are also involved in signal transduction pathways (2). One of the most prominent members of this receptor class is the Rvβ3 integrin which is involved in several pathological processes such as osteoporosis (3), restenosis (4), inflammation (5), rheumatoid arthritis (6), metastasis (7), and tumor-induced angiogenesis (8). Thus, great efforts are being made to develop Rvβ3 antagonists for
several therapeutic approaches (9, 10). First, cyclic pentapeptides (11, 12) were introduced which contain the tripeptide sequence arginine-glycine-aspartic acid (RGD1 in single letter amino acid code). This is a recognition motif used by several extracellular matrix proteins, like fibronectin and vitronectin, to bind to a variety of integrins including Rvβ3 (13). It has been demonstrated in model systems that cyclo(-Arg-Gly-Asp-D-Phe-Val-), one member of this peptide class, can block tumorinduced angiogenesis. For example, it has been shown in a chick chorioallontoic membrane model that this RGD-peptide inhibits tumor induced formation of new blood vessels without affecting already existing adjacent vessels (14). Initial clinical trials evaluating the use of Rvβ3 antagonists as antiangiogenic therapy in patients with various malignant tumors have been initiated (15) (e.g. Cilengitide, a derivative of cyclo(-Arg-Gly-Asp-D-Phe-Val-) where valine is replaced by NMeVal, is in clinical phase
* To whom requests for reprints should be addressed, at Department of Nuclear Medicine, Klinikum rechts der Isar, Technische Universita¨t Mu¨nchen, Ismaninger Str. 22, D-81675 Mu¨nchen, Germany. Phone: +49-89-41406332; FAX: +49-8941404841. E-mail:
[email protected]. ‡ Department of Nuclear Medicine. § Institute of Organic Chemistry and Biochemistry.
1 Abbreviations: EOB, end of bombardment; ESI-MS, electrospray ionization mass spectroscopy; [18F]FDG, 2-[18F]fluorodeoxyglucose; FP, 2-fluoropropionyl; NMeVal, NR-methylated valine; PET, positron emission tomography; RGD, single letter code for the amino acids arginine (R), glycine (G), and aspartate (D); SAA, sugar amino acid; SSTR, somatostatin receptor; TEMPO, 2,2,6,6-tetramethylpiperidine-1-yloxy.
10.1021/bc034170n CCC: $27.50 © 2004 American Chemical Society Published on Web 12/13/2003
62 Bioconjugate Chem., Vol. 15, No. 1, 2004
II studies (16)). Thus, radiolabeled RGD-peptides may be of advantage to document Rvβ3 expression of the tumors prior to administration of Rvβ3 antagonists allowing appropriate selection of patients entering clinical trials and determination of the optimal dose of Rvβ3 antagonists. Moreover, Rvβ3 expression has been reported to be an important factor determining the invasiveness and metastatic potential of malignant tumors in experimental tumor models as well as in patient studies (17, 18). Therefore, noninvasive imaging of Rvβ3 expression using radiolabeled RGD-peptides and positron emission tomography may provide a unique means of characterizing the biological aggressiveness of a malignant tumor in an individual patient. On the basis of cyclo(-Arg-Gly-Asp-D-Phe-Val-), several radiohalogenated and radiometalated RGD-peptides have been synthesized during the past few years (for review see ref 19). First tracers resulted from radioiodination of cyclo(-Arg-Gly-Asp-D-Tyr-Val-) and cyclo(-Arg-Gly-AspD-Phe-Tyr-) (20). These compounds showed high Rvβ3affinity and selectivity in vitro and receptor specific tumor uptake in vivo. However, due to their lipophilic character, high activity concentration was found in the liver. To reduce the lipophilicity of the tracer, the peptide was modified allowing conjugation with a sugar amino acid. The resulting radioiodinated glycopeptide GlucoRGD showed higher initial retention in the blood, higher uptake in the tumor and clearly reduced activity concentration in the liver (21). On the basis of these results, a 18 F-labeled glycopeptide was synthesized using 4-nitrophenyl-2-[18F]fluoropropionate as prosthetic group. It has already been demonstrated that this compound allows noninvasive determination of Rvβ3-expression in a murine tumor model using a dedicated small animal PET scanner (22). Here we describe the detailed synthesis of the precursor including synthesis of the corresponding sugar amino acid, the optimization of the labeling strategy, the metabolic stability, and first radiation dose estimates on [18F]Galacto-RGD. MATERIALS AND METHODS
All chemicals were used as supplied without further purification. 9-Fluorenylmethoxycarbonyl (Fmoc) amino acids were purchased from Bachem (Heidelberg, Germany) or Novabiochem (Bad Soden, Germany). The trityl chloride polystyrene (TCP) resin was purchased form PepChem (Tu¨bingen, Germany). 1-Hydroxybenzotriazole (HOBt), O-(1H-benzotriazol-1-yl)-N,N,N′,N′-tetramethyluronium tetrafluoroborate (TBTU), and diphenyl phosphorazidate (DPPA) were purchased from Aldrich (Steinheim, Germany) or Alexis (Gru¨nberg, Germany). 1-Hydroxy-7-azabenzotriazole (HOAt) and O-(7-azabenzotriazol-1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate (HATU) were purchased from PerSeptive Biosystems (Hamburg, Germany). All other organic reagents were purchased from Merck (Darmstadt, Germany), Aldrich, or Fluka (Neu-Ulm, Germany). Human M21 melanoma cells were a kind gift of D. A. Cheresh, Departments of Immunology and Vascular Biology, The Scripps Research Institute, La Jolla, CA. Mass spectra were recorded on the LC-MS system LCQ from Finnigan (Bremen, Germany) using the HewlettPackard series 1100 HPLC system. NMR-spectra were recorded on a Bruker AC 250 or Bruker AMX 500 (Karlsruhe, Germany) at 300 K. For all experiments, the solvent signal was used for calibration. Analytical reversed-phase high-performance liquid chromatography (RP-HPLC), using several acetonitrile/water
Haubner et al.
Figure 1. Synthesis of the Fmoc-protected sugar amino acid (Fmoc-SAA-OH). Synthesis starts with the transformation of the anomeric acteyl group of the pentaacetyl-protected galactose into a cyanide group using trimethylsilyl cyanide (TMSCN) under Lewis acidic conditions. The nitrile group is reduced to an aminomethylene unit using lithiumaluminum hydride. Subsequently, Fmoc-protection via standard protocols was carried out. Finally, the primary hydroxyl group is oxidized using a TEMPO-catalyzed oxidation.
gradients with 0.1% trifluoroacetic acid, was performed on Sykam equipment (Gilching, Germany) using columns with YMC-Pack ODS-A (5 µm, 250 × 4 mm) (YMC Co., Ltd, Kyoto, Japan). For radioactivity monitoring, the outlet of the UV detector was connected to a well-type NaI(Tl) detector from EG&G (Mu¨nchen, Germany). Preparative RP-HPLC was performed with the same system using a YMC-Pack J’sphere H80 (4 µm, 150 × 20 mm) column. Synthesis of the Fmoc-Protected Sugar Amino Acids (Fmoc-SAA-OH) (Figure 1). 2,3,4,6-Tetra-Oacetylgalactosyl-β-D-pyranosyl Cyanide (Ac4GalCN). Synthesis of the protected pyranosyl cyanide was carried out using a slightly modified protocol compared with literature (23). Briefly, to a solution of 42.8 g of penta-Oacetylgalacto-β-D-pyranose (110 mmol) in 200 mL of nitromethane were consecutively added 15 mL of trimethylsilyl cyanide (110 mmol) and 3 mL of boron trifluoride etherate (0.33 mmol, 0.3 equiv). After stirring at room temperature for 1 h, an additional 10 mL of trimethylsilyl cyanide (55 mmol, 0.5 equiv) and 2 mL of boron trifluoride etherate (22 mmol, 0.15 equiv) were added. After stirring for a further 30 min, the solvent was evaporated and the residue was dissolved in acetyl acetate (EtOAc). The solution was consecutively washed with a saturated NaHCO3 solution, H2O, and brine and dried over Na2SO4. Evaporation of the solvent and crystallization from EtOAc yielded 32.4 g (83%). Rf (hexane/EtOAc 2:1) ) 0.3; ESI-MS: m/z ) 710 (2M HCN + Na+), 737 (2M + Na+); Analysis (C15H19NO9) C, H, N; calcd C 50.42, H 5.36, N 3.92; found C 50.39, H 5.14, N 4.05; 1H NMR (CDCl3, 250 MHz): 5.5 (t, 1H, H2); 5.42 (dd, 1H, 3J3,4 ) 3.3 Hz, 3J4,5 ) 1.1 Hz, H4); 4.95 (dd, 1H, 3J2,3) 10.1 Hz, 3J3,4 ) 3.3 Hz, H3); 4.15 (d, 1H, 3J1,2 ) 10.2 Hz, H1), 4.08 (d, 2H, 3J5,6 ) 6.3 Hz, H6 und H6′); 3.93 (td, 1H, 3J4,5 ) 1.1 Hz, 3J5.6 ) 6.3 Hz, H5); 2.121.92 (4s, 12H, acetyl-CH3); 13C NMR (DMSO-d6, 62.5 MHz): 169.9, 169.8, 169.3, 169.1 (Acetyl-CO); 115.6 (CN); 74.5 (C1); 69.7 (C2); 67.3 (C3); 65.9 (C4); 65.0 (C5); 61.6 (C6); 20.5, 20.3, 20.24, 20.19 (acetyl-CH3) β-D-Galactopyranosylmethylamine (GalCH2NH2). GalCH2NH2 was synthesized according to BeMiller et al. (24) using a modified workup procedure. Briefly, to a suspension of 20 g of LiAlH4 (527 mmol, 3.8 equiv) in 100 mL of dry tetrahydrofuran (THF) was slowly added a saturated solution of 50 g of Ac4GalCN (140 mmol) in dry THF. After completion of addition, the resulting suspension was stirred for 3 h at RT. Ethanol was added dropwise
Synthesis of [18F]Galacto-RGD
until no further hydrogen evolution was detectable. Subsequently, 100 mL concentrated NH3 solution was slowly added, and the suspension was stirred for 1.5 h. The suspension was centrifuged, and the remaining solid was washed five times with water. The following steps can be accomplished without further purification although the solid contains large amounts of inorganic salts. ESI-MS: m/z ) 194.2 (M + H+); 200.1 (M + Li+) N-(Fluorenylmethyloxycarbonyl)-β-D-galactopyranosylmethylamine. To a solution of 4.7 g of GalCH2NH2 (24 mmol) in 70 mL of a 10% NaHCO3 solution was added a solution of 6.93 g FmocCl (26.8 mmol, 1.1 equiv) in 40 mL of THF. After stirring for 1.5 h at RT, the pH of the solution was adjusted to 2 using cationic ion-exchange resin (Amberlyst 15, H+ form). The resin was filtrated off and washed thoroughly with THF. The organic solvents were removed in vacuo and the remaining emulsion was extracted twice with hexane and five times with EtOAc. The combined EtOAc-fractions were evaporated to dryness and coevaporated with toluene. The resulting crude product (9.6 g) can be used without further purification. Rf (chloroform/methanol 9:1) ) 0.45; ESI-MS: m/z ) 416.2 (M + H+) Fmoc-SAA-OH. The reaction was accomplished using a procedure published by de Nooy et al. (25). For the workup, the pH of the reaction mixture was adjusted to 1.5 using cationic ion-exchange resin (Amberlyst 15, H+ form), then the resin was filtered off and washed thoroughly with water. The filtrate was saturated with NH4Cl and extracted five times with EtOAc. The collected organic layers were evaporated, then coevaporated twice with toluene. The product was purified by crystallization from methanol. Nine grams of N-(fluorenylmethyloxycarbonyl)-β-D-galactopyranosyl-methylamine yield 6.7 g (72%) of Fmoc-SAA-OH as white crystalls. Rf (chloroform/ methanol 9:1) ) 0.15; ESI-MS: m/z ) 429.9 (M + H+); 452.2 (M + Na+); 663.6 (3M + K+ + Na+/2); 858.8 (2M + H+); 880.9 (2M + Na+); 897.0 (2M + K+); 1309.6 (3M + Na+); 1326.7 (3M + K+); 1H NMR (DMSO-d6, 500 MHz): 7.87 (d, 2H, 3JHH ) 7.5 Hz, Fmoc-CH); 7.69 (d, 2H, 3JHH ) 7.5 Hz, Fmoc-CH); 7.4 (t, 2H, 3JHH ) 7.9 Hz, FmocCH); 7.3 (m, 2H, Fmoc-CH); 7.23 (t, br, 1H, NH); 5.04.65 (m, 3H, OH); 4.32 (d, 2H, 3JHH ) 6.7 Hz, FmocCHCH2), 4.24 (t, 1H, 3JHH ) 6.6 Hz, Fmoc-CHCH2); 4.05 (s, 1H, H2); 3.97 (s, br, 1H, H3); 3.55 (dd, 1H, 3JHH ) 4.4 Hz, 3JHH ) 12.7 Hz, H7); 3.38 (m, 1H, H4); 3.30 (m, 1H, H5); 3.19 (m, 1H, H6); 3.05 (m, 1H, H7′); 13C NMR (DMSO-d6, 125 MHz): 170.2 (COOH); 156.2 (Fmoc-CO); 143.9-120.1 (arom-C); 78.2 (C6); 77.1 (C2); 73.8 (C4); 70.2 (C3); 67.9 (Fmoc-CH2); 65.4 (C5); 46.8 (Fmoc-CH); 42.2 (C7). Glycopeptide Synthesis (Figure 2). Cyclo(-Arg(Pbf)Gly-Asp(OtBu)-D-Phe-Lys-). Loading of the TCP-resin, synthesis of the peptide, subsequent cyclization, and selective deprotection of the lysine moiety were carried out following the protocols described elsewhere (20, 21, 26). Side chain protection was 2,2,4,6,7-pentamethyldihydrobenzofuran-5-sulfonyl (Pbf) for arginine, benzyloxycarbonyl (Z) for lysine, and tert-butyl (tBu) for aspartic acid. ESI-MS: (M + H)+ ) 912; RP-HPLC: tR ) 13.8 min, K′ ) 7.6 (30-80% MeCN; 30 min). Cyclo(-Arg(Pbf)-Gly-Asp(OtBu)- D -Phe-Lys(FmocSAA)-). After dissolving 0.08 mmol cyclo(-Arg(Pbf)-GlyAsp(OtBu)-D-Phe-Lys-) and 0.24 mmol Fmoc-SAA-OH in 6 mL dimethylformamide (DMF), 0.24 mmol HATU and 0.24 mmol HOAt were added. The pH was adjusted to 8 using N-ethyldiisopropylamine. After stirring the solution for 1 h at ambient temperature, the solvent was reduced in vacuo, the residue was triturated with water and the
Bioconjugate Chem., Vol. 15, No. 1, 2004 63
Figure 2. Schematic presentation of the synthesis of the 18Flabeled glycosylated RGD-peptide. The protected linear peptide was assembled on a trityl chloride resin. (A) Removal from the resin. (B) Cyclization of the side chain protected peptide. (C) Selective deprotection of the lysine -amino function. (D) Conjugation with Fmoc-SAA-OH. (E) Complete deprotection of the glycopeptide. (F) Labeling with [18F]NFP. Table 1. Analytical Data for the Labeling Precursor c(RGDfK(SAA)) (0-80% MeCN in 20 min) and the Reference Compound [19F]Galacto-RGD (0-50% MeCN in 30 min) peptide
formula
MW HPLC-MS tR [g/mol] (M + H)+ [min] K′
c(RGDfK(SAA)) C34H52N10O12 792.85 [19F]Galacto-RGD C37H55N10O13F 866.90
794 868
10.8 5.0 23.5 6.8
crude peptide was isolated by centrifugation. ESI-MS: (M + H)+ ) 1323; RP-HPLC: tR ) 33.0 min, K′ ) 11.7 (0-80% MeCN; 30 min). Cyclo(-Arg(Pbf)-Gly-Asp(OtBu)-D-Phe-Lys(SAA)-). The Fmoc-protected glycopeptide was dissolved in 5 mL of 20% piperidine in DMF and stirred for 20 min at ambient temperature. After concentration of the solvent in vacuo, the glycopeptide was precipitated with diethyl ether. The crude product was isolated by centrifugation and washed 3 times with diethyl ether. ESI-MS: (M + H)+ ) 1101; RP-HPLC: tR ) 16.7 min, K′ ) 6.0 (30-80% MeCN; 30 min). Cyclo(-Arg(Pbf)-Gly-Asp(OtBu)- D -Phe-Lys(SAA(FP))-). After dissolving 0.01 mmol cyclo(-Arg(Pbf)-GlyAsp(OtBu)-D-Phe-Lys(SAA)-) and 0.02 mmol 2-fluoropropionate in 1 mL of DMF, 0.02 mmol of HATU, and 0.02 mmol of HOAt were added. N-Ethyldiisopropylamine was used to adjust the pH to 8.5. After stirring the solution for 1 h at ambient temperature, the solvent was reduced in vacuo, and the residue was triturated with water. The crude product was isolated by centrifugation and dried in vacuo. ESI-MS: (M + H)+ ) 1174; RP-HPLC: tR ) 19.2 min, K′ ) 9.1 (30-80% MeCN; 30 min). Removal of the Side Chain Protection Groups of the Glycopeptides. Glycopeptides were treated with 20 mL of a solution of 95% trifluoroacetic acid (TFA), 2.5% water, and 2.5% triisobutylsilane for 24 h at ambient temperature. The mixture was filtered if necessary, evaporated in vacuo, triturated with diethyl ether, filtered, and washed several times with diethyl ether. The crude, cyclic glycopeptides were purified by RPHPLC. Analytical data are given in Table 1. [18F]Fluoride Production. No carrier added [18F]fluoride was produced on a RDS-112 cyclotron (Siemens/
64 Bioconjugate Chem., Vol. 15, No. 1, 2004
Figure 3. Schematic structure of [18F]Galacto-RGD. Labeling with [18F]NFP was carried out via acylation of the aminomethyl group at the C1-position of the sugar moiety. (This figure is reproduced with permission from the American Association for Cancer Research, Inc.; see ref 22).
CTI, Knoxville, TN) by irradiation of a water target using an 11 MeV proton beam on isotopically enriched [18O]water (Rotem, Beersheva, Israel). At end of bombardment, the enriched water containing [18F]fluoride was passed through an anion-exchange resin (Dowex, AG1 × 8, 40 mg). The [18F]fluoride was eluted from the resin using 200 to 300 µL of a 5.7 mg/mL aqueous K2CO3‚1.5H2O solution and collected in a 5 mL conical vial closed with a septum containing 10 mg of Kryptofix 222 in 500 µL of dry MeCN (acetonitrile for DNA synthesis). The resulting solution was concentrated to dryness at 100 °C in vacuo. The azeotropic drying was repeated at least twice with 500 µL portions of acetonitrile. Synthesis of 4-Nitrophenyl-2-[18F]fluoropropionate. Synthesis of 2-[18F]fluoropropionate was prepared according to literature (27-29). Synthesis of [18F]Galacto-RGD. The glycopeptide (0.01-5.0 mg) was dissolved in 200 µL of dimethyl sulfoxide (DMSO) and added to the polypropylene vial containing 4-nitrophenyl-2-[18F]fluoropropionate (185370 MBq). The reaction mixture was allowed to remain for different times at different temperatures (for details see Figure 4). Purification of the [18F]Galacto-RGD (Figure 3) was performed using RP-HPLC with an acetonitrile/water gradient containing 0.1% trifluoroacetic acid (10-50% acetonitrile in 30 min; flow 10 mL/min, tR ) 11.6 min, K′ ) 5.1). For animal experiments the solvent was removed in vacuo, and the residue was dissolved in phosphate-buffered saline (PBS) pH 7.4. Octanol/Water Partition Coefficient. Approximately 2.5 kBq [18F]Galacto-RGD in 10 µL PBS was diluted with 490 µL of PBS and added to 500 µL of octanol in an Eppendorf tube. After vigorous mixing of the suspension for 3 min at room temperature, the caps were centrifuged (10 300 g; 6 min) and 100 µL aliquots of both layers were counted in a gamma counter. The experiment was repeated six times. Metabolic Stability. Tumor Xenografts. Metabolic stability of [18F]Galacto-RGD was evaluated in mice using a xenotransplanted human melanoma model. It has previously been demonstrated that this tumor model shows high expression of the Rvβ3 integrin (20, 30). Human M21 melanoma cells (30) were cultured in a humidified atmosphere with 5% CO2. The cell culture medium was RPMI 1640 (Seromed Biochrom, Berlin,
Haubner et al.
Germany) supplemented with 10% fetal calf serum and gentamycine. Tumor xenografts were obtained by subcutaneous injection 5 × 106 cells. Mice bearing tumors weighing 300-400 mg were used for the studies. Metabolic Stability. Nude mice bearing tumor xenografts of human melanoma M21 were intravenously injected with 3.7 MBq of [18F]Galacto-RGD. The animals were sacrificed and dissected 120 min after tracer injection. Blood, liver, kidney, and tumor were removed. Blood was immediately centrifuged for 5 min at 1200 g. Organs were frozen with liquid nitrogen, homogenized using a Mikro Dismembranator II (B. Braun, Melsungen, Germany), suspended in 1 mL of PBS pH 7.4, and centrifuged for 3 min at 15 600g. After removal of the supernatants, the pellets were washed with 500 µL of PBS. For each sample supernatants of both centrifugation steps were combined and passed through Sep-Pak C18 cartridges. The cartridges were washed with 2 mL of water and eluted with 2 mL of acetonitrile containing 0.1% TFA. Before analysis with RP-HPLC, the acetonitrile was removed in vacuo, and the residues were dissolved in 500 µL of PBS. Aliquots of 350 µL were injected onto a RPHPLC (Sykam, Gilching, Germany; 150 × 10 mm Multosphere 100 RP-18, 5 µm column, CS Chromatographie Service GmbH, Langerwehe, Germany) using a flow of 3 mL/min (gradient: 10-50% acetonitrile, 20 min). Radioactivity was monitored using a flow scintillation analyzer (Radiomatic 500 TR Series, Packard Canberra Company, Meriden, CT). HPLC analysis was carried out in duplicate and extraction efficiency determination in triplicate. To verify if the peak eluting approximately 1.5 min before [18F]Galacto-RGD (designated as metabolite 2) is a metabolite of [18F]Galacto-RGD and not [18F]GalactoRGD in a different protonation status, both peaks were sampled and aliquots were reinjected onto the HPLC. In addition, coinjection of both fractionated samples was carried out. Dynamic PET Study and Dose Estimates. Anesthetized New Zealand white rabbits underwent dynamic PET studies using a Simens ECAT HR+ scanner (Siemens/CTI, Knoxville, Tennessee). Tracer distribution was followed for 60 min after injection of 18.5 MBq of [18F]Galacto-RGD. Residence times were calculated by monoexponential extrapolation of the time activity curves for the various organ systems. According to the residence times radiation doses were calculated for an adult female patient using the MIRDOSE 3.0 program. RESULTS
Radiosynthesis. 18F-Fluorination of Galacto-RGD was carried out using 4-nitrophenyl-2-[18F]fluoropropionate. Synthesis of the prosthetic group follows the protocol previously published (27-29) and resulted in a radiochemical yield of 45.2 ( 10.6% (n ) 20) with an average reaction time of 126 ( 10 min. As shown in Figure 3, radiolabeling of the glycopeptide was performed via acylation of the amino methyl group at the C1-position of the sugar moiety. This led to a 18F-labeled glycopeptide with an even more hydrophilic character than found for radioiodinated Gluco-RGD (21) (log P([125I]Gluco-RGD) ) -2.45 vs log P([18F]Galacto-RGD) ) -3.17). The dependence of the radiochemical acylation yield on peptide concentration, temperature, and reaction time was investigated (for details see Figure 4). The maximum yield for the acylation step was approximately 85%. This can be reached by either using 0.5 mg of peptide at 70 °C within 10 min, 1 mg of peptide at 45 °C within 10 min, or 1 mg of peptide at 70 °C within 2.5 min. At least
Synthesis of [18F]Galacto-RGD
Bioconjugate Chem., Vol. 15, No. 1, 2004 65 Table 2. Extraction Efficiency and Data from the HPLC Analysis of the Soluble Fraction of the Organ Homogenates and the Blood Samples 120 min after Tracer Injection blood
liver
kidneys
tumor
Extraction Efficiency [%] (n ) 3) unsoluble fraction 25.7 ( 3.8 34.1 ( 6.7 5.3 ( 3.1 19.8 ( 5.9 nonretained fractiona 7.8 ( 2.3 5.8 ( 0.1 1.7 ( 0.7 5.4 ( 1.1 soluble fractionb 66.6 ( 2.3 60.2 ( 6.9 93.0 ( 3.7 74.8 ( 6.6 blood met 1 met 2 intact tracer
liver
kidneys
HPLC Analysis [%] (n ) 2) n.d.c 2.4 ( 0.4 3.5 ( 1.3 14.4 ( 5.1 21.9 ( 1.8 27.8 ( 1.3 85.6 ( 5.2 75.8 ( 2.2 68.7 ( 2.5
tumor 0.7d 12.3 ( 1.7 87.4 ( 2.2
a Amount of activity, which could not be fixed on the C-18 cartridge. b Amount of activity, which was eluted from the C-18 cartridge using acetonitrile with 0.1% trifluoroacetic acid. c Not detectable. d Detectable only in one sample.
Figure 4. Dependence of the radiochemical acylation yield from (A) peptide concentration (70 °C, 10 min), (B) reaction temperature (6.3 mM peptide, 10 min), and (C) reaction time (6.3 mM peptide, 70 °C).
76% can be reached using 0.5 mg Galacto-RGD at 45 °C within 2.5 min. Final RP-HPLC with a semipreparative column allows preparation of [18F]Galacto-RGD with radiochemical purities >98% and with specific activities ranging from 40 to 100 TBq/mmol. The total radiochemical yield was 29.5 ( 5.1% (EOB) with a total reaction time of 200 ( 18 min including final HPLC purification (calculation based on 1.0 mg of peptide, 70 °C, and 10 min reaction time (n ) 6)). Typically, starting with 2200 MBq of [18F]F-, 185 MBq of [18F]Galacto-RGD was prepared in a synthesis time compatible with the halflife of 18F-fluorine. Metabolism of [18F]Galacto-RGD. The metabolic stability of [18F]Galacto-RGD was determined in mouse blood and liver, kidney, and tumor homogenates 2 h after tracer injection. For all organs extraction efficiency was between 60% and 93% (Table 2). The lowest extraction efficiency was found for the liver homogenates and highest was found for the kidneys. Between 2 and 8% of the total activity could not be fixed on the C-18 cartridges, which can correspond with very hydrophilic metabolites or protein bound activity. HPLC analysis of the soluble fraction of the different samples (Figure 5) allowed detection of two radiolabeled metabolites in tumor, liver, and kidney homogenates. The retention times of metabolite 1 (met 1) and metabolite 2 (met 2) were ∼4.0 min and ∼9.5 min, respectively. The retention time for [18F]Galacto-RGD was ∼11.0 min. In blood samples only one metabolite, corresponding with met 2 in the other samples, was found. Detailed HPLCanalysis of met 2 and [18F]Galacto-RGD showed that both
peaks represent two different radiolabeled species and not [18F]Galacto-RGD in different protonated forms. The average fraction of intact tracer was between 70% and 87% (Table 2). The average amount of met 1 was between 1% and 4% and of met 2 between 12% and 28%. Highest metabolic stability of [18F]Galacto-RGD was found in tumor (87% intact tracer) and blood (86% intact tracer). PET Study and Dose Estimates. The dynamic PET study of two New Zealand white rabbits showed fast renal elimination of the tracer with low activity accumulation in almost all organs of the rabbit (Figure 6). One minute after injection of 18.5 MBq of [18F]GalactoRGD, highest activity concentration was found in heart, kidneys, liver, and large blood vessels. At 15 min pi, most activity could be detected in kidneys and at 60 min pi in the urinary bladder. Calculation of time activity curves demonstrated that there is no tracer retention in any of the investigated organs (Figure 7). The expected radiation dose to an adult female patient after injection of [18F]Galacto-RGD has been estimated to be of 2.2 × 10-2 mGy/MBq. Only slightly different results were obtained from radiation dose calculations based on the biodistribution data in mice (biodistribution data can be found in (22)). An effective radiation dose of 3.8 × 10-2 mGy/MBq was calculated on the basis of the mouse data. The highest radiation doses were found for the urinary bladder wall and the small intestine (1.5 × 10-1 mGy/MBq and 2.2 × 10-1 mGy/MBq, respectively). In rabbits the radiation dose for the gallbladder wall and the small intestine were lower than in the mouse model, probably reflecting slower excretion kinetics in larger animals. Table 3 shows the expected radiation dose for the individual organ systems. DISCUSSION
It has been shown that sugar derivatives can improve several pharmacological properties of peptides such as bioavailability (31), increased stability toward enzymatic degradation (32), and solubility under physiological conditions (33). Recently, radiolabeled octreotide and octreotate derivatives conjugated with glucose, maltose or maltotriose have been described (29, 34-36). These glycosylated tracers showed reduced liver and increased tumor uptake resulting in excellent tumor-to-background ratios which allows imaging of SSTR-positive tumors with high contrast. We used a similar approach to improve the pharmacokinetic behavior of our first generation compounds and conjugated sugar amino acids with cyclo(-Arg-Gly-AspD-Tyr-Lys-) (21). An advantage of sugar amino acids,
66 Bioconjugate Chem., Vol. 15, No. 1, 2004
Haubner et al.
Figure 5. Representative HPLC profiles of the reference compound, and the soluble fraction of blood samples, tumor, kidney, and liver homogenates collected 120 min after tracer injection.
Figure 6. Dynamic PET study of a New Zealand white rabbit over 60 min after injection of [18F]Galacto-RGD. The image 1 min after tracer injection shows blood-pool activity in the heart, liver, right kidney, and large blood vessels. In the image 15 min pi there is accumulation of the tracer in the right kidney and clearance of blood-pool activity. Furthermore, activity is found in the bladder. At 60 min pi the tracer has accumulated in the bladder. Liver and right kidney are only faintly visible. Please note that some parts of the left side of the rabbit including the left kidney were not in the field of view of the PET scanner.
which are sugar derivatives with an amino and a carboxylic function (37), is that they can easily be used with standard peptide synthesis protocols. Moreover, due to the carbon-carbon bond at the anomeric center, the
resulting glycopeptides are expected to be very stable toward metabolic degradation. For the synthesis of the first radioiodinated glycopeptide, a sugar amino acid based on glucose was used. The resulting radioiodinated
Synthesis of [18F]Galacto-RGD
Bioconjugate Chem., Vol. 15, No. 1, 2004 67
Figure 7. Time activity curves derived from the dynamic PET study shown in Figure 6. Table 3. Estimated Radiation Doses for a Female Adult Patient after Intravenous Injection of [18F]Galacto-RGD, as Derived from Melanoma-Bearing Nude Mice and New Zealand White Rabbits radiation dose [mGy/MBq] target organ adrenals brain breast gallbladder wall large intestine small intestine stomach heart wall kidneys liver lungs muscle ovaries pancreas red marrow bone skin spleen thymus thyroid urinary bladder wall uterus total body effective dose
mouse data
rabbit data
10-3
1.59 × 10-2 1.42 × 10-2 1.24 × 10-2 1.72 × 10-2 2.78 × 10-2 1.85 × 10-2 1.60 × 10-2 1.58 × 10-2 2.56 × 10-2 1.63 × 10-2 1.00 × 10-2 8.60 × 10-3 1.92 × 10-2 1.68 × 10-2 1.41 × 10-2 1.51 × 10-2 1.14 × 10-2 1.55 × 10-2 1.51 × 10-2 1.39 × 10-2 1.04 × 10-1 2.22 × 10-2 4.47 × 10-2 2.16 × 10-2
8.87 × 1.29 × 10-4 8.05 × 10-4 2.21 × 10-2 8.19 × 10-2 2.19 × 10-1 1.06 × 10-2 2.11 × 10-3 2.04 × 10-2 1.03 × 10-2 1.69 × 10-3 6.40 × 10-3 3.98 × 10-2 8.34 × 10-3 8.79 × 10-3 4.31 × 10-3 2.17 × 10-3 6.92 × 10-3 4.66 × 10-3 3.13 × 10-4 1.50 × 10-1 3.47 × 10-2 9.09 × 10-3 3.78 × 10-2
Gluco-RGD showed promising improved pharmacokinetics and increased activity retention in the tumor (21). For 18 F-labeling with prosthetic groups, another sugar amino acid based on galactose was introduced (22) allowing acylation at the amino function at the C1-position of the sugar moiety. This Fmoc-protected sugar amino acid (Fmoc-SAA-OH, Figure 1) can be synthesized in four steps starting from pentaacetyl protected galactose. The reaction starts with the transformation of the anomeric acetyl into a cyanide group by reaction of the protected galactose with trimethylsilyl cyanide under Lewis acidic conditions. The nitrile group is then reduced to an aminomethylene unit using lithiumaluminum hydride and the newly formed amino group is subsequently Fmoc-protected via standard protocols. In the last step the primary hydroxyl group is oxidized using a TEMPO-catalyzed oxidation which yields the carboxylic acid moiety that enables fast and efficient glycosylation of the partially protected cyclo(-Arg-Gly-Asp-D-Phe-Lys-) via the -amino function of the lysine. The final deprotection of the peptide side chain functions, as well as the primary amino function at the
sugar moiety, offers a RGD-containing glycopeptide which can be labeled with 18F-fluorine by prosthetic groups, such as succinimidyl-4-[18F]fluorobenzoate ([18F]SFB) (38) and 4-nitrophenyl-2[18F]-fluoropropionate ([18F]NFP) (27). To minimize the negative effect of hydrophobic prosthetic groups, such as [18F]SFB, on the optimized hydrophilicity of sugar-conjugated tracer, [18F]NFP was chosen for labeling. Acylation of the glycopeptide with this prosthetic group resulted in the radiolabeled compound with high radiochemical purity and radiochemical yield. By using a peptide concentration of 0.32 mM, radiochemical acylation yields of approximately 76% can already be reached at 45 °C within 2.5 min. Moreover, it is not necessary to add any additional base during the acylation step when the amino function of the sugar is in the deprotonated form. This can be carried out by adding 1 equiv sodium carbonate and subsequent removal of the solvent in vacuo. Although the synthesis of [18F]NFP is a time-consuming three step procedure necessary to be improved, the discussed synthesis route allows the production of [18F]Galacto-RGD in amounts sufficient for clinical trials. In murine tumor models it has already been demonstrated that [18F]Galacto-RGD shows receptor specific accumulation in the tumor with good tumor-to-background ratios and allows noninvasive determination of Rvβ3-expression using a small animal scanner (22). The dynamic PET study using a New Zealand white rabbit confirmed the data from biodistribution studies with tumor-bearing mice indicating fast predominantly renal excretion of the tracer. The radiation dose estimates using the MIRDOSE program resulted in values very similar to that found for [18F]FDG, indicating low radiation doses for patients. The studies concerning the metabolic stability showed that there remains between 5 and 35% of the injected activity in the unsoluble fraction, which could be due to coprecipitation, or especially for the blood sample, also due to some binding to cells. However, this was not verified in more detail in this study. Nevertheless, between 65 and 90% of the tracer in the soluble fraction remains intact during the time frame essential for 18F-PET studies. Especially in tumor and blood, even 2 h after tracer injection more than 85% of the activity found correspond with the intact tracer, indicating that the resulting signal mainly reflects tracer accumulation. In conclusion, glycosylation of RGD-peptides using a sugar amino acid is easy to accomplish via standard peptide chemistry protocols. Subsequently, radiolabeling with [18F]NFP results in [18F]Galacto-RGD in high radiochemical yields and radiochemical purity. Introduction of the sugar moiety in combination with this prosthetic group leads to a hydrophilic tracer with fast predominantly renal excretion and suitable metabolic stability in vivo which allows noninvasive determination of Rvβ3expression in different murine tumor models. The route described enables production of a standard amount of [18F]Galacto-RGD sufficient for patient studies, starting with approximately 2.2 GBq of [18F]fluoride. The estimated radiation dose is in the same range as found for [18F]FDG. ACKNOWLEDGMENT
This work was supported by a grant from the Sander Foundation (grant no. 96.017.3). Wolfgang Linke, Claudia Bodenstein, Gitti Dzewas, and Burghard Cordes are acknowledged for excellent technical assistance and Amy R. Claus for carefully reading the manuscript. We thank
68 Bioconjugate Chem., Vol. 15, No. 1, 2004
the RDS-cyclotron team, and especially Michael Herz, for providing the 18F-fluorine and initial training at the hot cells. We acknowledge David A. Cheresh, The Scripps Research Institute, La Jolla, CA, for supplying us with M21 melanoma cells. The rabbit was kindly provided by Ivo Buschmann and Emanuelle K. Chorianopoulos, University of Freiburg, Germany. LITERATURE CITED (1) Cox, D., Aoki, T., Seki, j., Motoyama, Y., and Yoshida, K. (1994) The pharmacology of integrins. Med. Res. Rev. 14, 192-228. (2) Hynes, R. O. (1992) Integrins: versatility, modulation, and signaling in cell adhesion. Cell 69, 11-25. (3) Teitelbaum, S. L. (2000) Bone resorption by osteoclasts. Science 289, 1504-1508. (4) Bishop, G. G., McPherson, J. A., Sanders, J. M., Hesselbacher, S. E., Feldman, M. J., McNamara, C. A., Gimple, L. W., Powers, E. R., Mousa, S. A., and Sarembock, I. J. (2001) Selective alpha(v)beta(3)-receptor blockade reduces macrophage infiltration and restenosis after balloon angioplasty in the atherosclerotic rabbit. Circulation 103, 1906-1911. (5) Albelda, S. M., Smith, C. W., and Ward, P. A. (1994) Adhesion molecules and inflammatory injury. FASEB J. 8, 504-512. (6) Storgard, C. M., Stupack, D. G., Jonczyk, A., Goodman, S. L., Fox, R. I., and Cheresh, D. A. (1999) Decreased angiogenesis and arthritic disease in rabbits treated with an alpha(v)beta(3) antagonist. J. Clin. Invest. 103, 47-54. (7) Clezardin, P. (1998) Recent insights into the role of integrins in cancer metastasis. Cell Mol. Life Sci. 54, 541-548. (8) Eliceiri, B. P., and Cheresh, D. A. (1999) The role of alpha(v) integrins during angiogenesis: insights into potential mechanisms of action and clinical development. J. Clin. Invest. 103, 1227-1230. (9) Ranieri, G., and Gasparini, G. (2001) Angiogenesis and angiogenesis inhibitors: a new potential anticancer therapeutic strategy. Curr. Drug Targets Immune Endocr. Metabol. Disord. 1, 241-253. (10) Kerr, J. S., Slee, A. M., and Mousa, S. A. (2002) The alpha(v) integrin antagonists as novel anticancer agents: an update. Expert Opin Investig Drugs 11, 1765-1774. (11) Aumailley, M., Gurrath, M., Muller, G., Calvete, J., Timpl, R., and Kessler, H. (1991) Arg-Gly-Asp constrained within cyclic pentapeptides. Strong and selective inhibitors of cell adhesion to vitronectin and laminin fragment P1. FEBS Lett. 291, 50-54. (12) Gurrath, M., Muller, G., Kessler, H., Aumailley, M., and Timpl, R. (1992) Conformation/activity studies of rationally designed potent anti- adhesive RGD peptides. Eur. J. Biochem. 210, 911-921. (13) Ruoslahti, E., and Pierschbacher, M. D. (1987) New perspectives in cell adhesion: RGD and integrins. Science 238, 491-497. (14) Brooks, P. C., Montgomery, A. M., Rosenfeld, M., Reisfeld, R. A., Hu, T., Klier, G., and Cheresh, D. A. (1994) Integrin alpha(v)beta(3) antagonists promote tumor regression by inducing apoptosis of angiogenic blood vessels. Cell 79, 11571164. (15) Brower, V. (1999) Tumor angiogenesissnew drugs on the block. Nat. Biotechnol. 17, 963-968. (16) Dechantsreiter, M. A., Planker, E., Matha, B., Lohof, E., Holzemann, G., Jonczyk, A., Goodman, S. L., and Kessler, H. (1999) N-Methylated cyclic RGD peptides as highly active and selective alpha(V)beta(3) integrin antagonists. J. Med. Chem. 42, 3033-3040. (17) Felding-Habermann, B., Mueller, B. M., Romerdahl, C. A., and Cheresh, D. A. (1992) Involvement of integrin alpha(v) gene expression in human melanoma tumorigenicity. J. Clin. Invest. 89, 2018-2022. (18) Gasparini, G., Brooks, P. C., Biganzoli, E., Vermeulen, P. B., Bonoldi, E., Dirix, L. Y., Ranieri, G., Miceli, R., and Cheresh, D. A. (1998) Vascular integrin alpha(v)beta(3): a
Haubner et al. new prognostic indicator in breast cancer. Clin. Cancer Res. 4, 2625-2634. (19) Haubner, R., Wester, H. J., Weber, W. A., and Schwaiger, M. (2003) Radiotracer-based strategies to image angiogenesis. Quat. J. Nucl. Med. 47, 189-199. (20) Haubner, R., Wester, H. J., Reuning, U., SenekowitschSchmidtke, R., Diefenbach, B., Kessler, H., Stocklin, G., and Schwaiger, M. (1999) Radiolabeled alpha(v)beta(3) integrin antagonists: a new class of tracers for tumor targeting. J. Nucl. Med. 40, 1061-1071. (21) Haubner, R., Wester, H. J., Burkhart, F., SenekowitschSchmidtke, R., Weber, W., Goodman, S. L., Kessler, H., and Schwaiger, M. (2001) Glycosylated RGD-containing peptides: tracer for tumor targeting and angiogenesis imaging with improved biokinetics. J. Nucl. Med. 42, 326-336. (22) Haubner, R., Wester, H. J., Weber, W. A., Mang, C., Ziegler, S. I., Goodman, S. L., Senekowitsch-Schmidtke, R., Kessler, H., and Schwaiger, M. (2001) Noninvasive imaging of alpha(v)beta(3) integrin expression using 18F-labeled RGD-containing glycopeptide and positron emission tomography. Cancer Res. 61, 1781-1785. (23) de las Heras, F. G., and Fernandez-Resa, P. (1982) Synthesis of Ribosyl and Arabinosyl Cyanides by Reaction of 1-O-Acyl Sugars with Trimethylsilyl Cyanide. J. Chem. Soc., Perkin Trans. 1 903-907. (24) BeMiller, J. N., Yadav, M. P., Kalabokis, V. N., and Myers, R. W. (1990) N-Substituted beta-D-galactopyranosylmethylamines, and C-beta-D-galactopyranosylformamides, and related compounds. Carbohydr Res. 200, 111-126. (25) de Nooy, A. E. J., Besemer, A. C., and van Bekkum, H. (1995) Highly selective nitroxyl radical-mediated oxidation of primary alcohol groups in water soluble glucans. Carbohydr Res. 269, 89-98. (26) Haubner, R., Gratias, R., Diefenbach, B., Goodman, S. L., Jonczyk, A., and Kessler, H. (1996) Structural and Functional Aspects of RGD-Containing Cyclic Pentapeptides as Highly Potent and Selective Integrin alpha(v)beta(3) Antagonists. J. Am. Chem. Soc. 118, 7461-7472. (27) Guhlke, S., Coenen, H. H., and Sto¨cklin, G. (1994) Fluoroacylation Agents Based on Small nca [18F]Fluorocarboxylic Acids. Appl. Radiat. Isot. 45, 715-727. (28) Wester, H. J., Brockmann, J., Ro¨sch, F., Wutz, W., Herzog, H., Smith-Jones, P., Stolz, B., Bruns, C., and Sto¨cklin, G. (1997) PET-pharmacokinetics of 18F-octreotide: A comparison 67Ga-DFO-octreotide and 86Y-DTPA-octreotide. Nucl. Med. Biol. 24, 275-286. (29) Wester, H. J., Schottelius, M., Scheidhauer, K., Meisetschlager, G., Herz, M., Rau, F. C., Reubi, J. C., and Schwaiger, M. (2003) PET imaging of somatostatin receptors: design, synthesis and preclinical evaluation of a novel 18F-labeled, carbohydrated analogue of octreotide. Eur. J. Nucl. Med. Mol. Imaging 30, 117-122. (30) Cheresh, D. A., and Spiro, R. C. (1987) Biosynthetic and functional properties of an Arg-Gly-Asp-directed receptor involved in human melanoma cell attachment to vitronectin, fibrinogen, and von Willebrand factor. J. Biol. Chem. 262, 17703-17711. (31) Albert, R., Marbach, P., Bauer, W., Briner, U., Fricker, G., Bruns, C., and Pless, J. (1993) SDZ CO 611: a highly potent glycated analogue of somatostatin with improved oral activity. Life Sci. 53, 517-525. (32) Marastoni, M., Spisani, S., and Tomatis, R. (1994) Synthesis and biological activity of d-glycopyranosyl peptide T derivatives. Arzneimittelforschung/Drug Res. 44, 984-987. (33) Michael, K., Wittmann, V., Ko¨nig, W., Sandow, J., and Kessler, H. (1996) S- and C-glycopeptide derivatives of an LH-RH agonist. Int. J. Pept. Protein Res. 48, 59-70. (34) Leisner, M., Kessler, H., Schwaiger, M., and Wester, H. J. (1999) Synthesis of Na-D-Phe1-Amadori Derivatives of Tyr3Octreotide: Precursor for 123I-/18F-Labeled SSTR-Binding SPECT/PET Tracers with Improved Biodistribution. J. Labelled Compd. Radiopharm. 42 (Suppl.), S549-S551. (35) Schottelius, M., Wester, H. J., Reubi, J. C., SenekowitschSchmidtke, R., and Schwaiger, M. (2002) Improvement of
Synthesis of [18F]Galacto-RGD Pharmacokinetics of Radioiodinated Tyr3-Octreotide by Conjugation with Carbohydrates. Bioconjugate Chem. 13, 10211030. (36) Wester, H. J., Schottelius, M., Scheidhauer, K., Reubi, J. C., Wolf, I., and Schwaiger, M. (2002) Comparison of radioiodinated TOC, TOCA and Mtr-TOCA: the effect of carbohydration on the pharmacokinetics. Eur. J. Nucl. Med. Mol. Imag. 29, 28-38.
Bioconjugate Chem., Vol. 15, No. 1, 2004 69 (37) Gruner, S. A., Locardi, E., Lohof, E., and Kessler, H. (2002) Carbohydrate-based mimetics in drug design: sugar amino acids and carbohydrate scaffolds. Chem. Rev. 102, 491-514. (38) Wester, H. J., Hamacher, K., and Sto¨cklin, G. (1996) A Comparative Study of N.C.A. Fluorine-18 Labeling of Proteins via Acylation and Photochemical Conjugation. Nucl. Med. Biol. 23, 365-372. BC034170N
