Highly Efficient Technetium-99m Labeling ... - ACS Publications

Scintigraphic images were collected using a YAP-camera, a γ-camera optimized to perform experiments on small animals with a spatial resolution of 1.0...
0 downloads 0 Views 481KB Size
Download PDF
1046

Bioconjugate Chem. 2004, 15, 1046−1054

Highly Efficient Technetium-99m Labeling Procedure Based on the Conjugation of N-[N-(3-Diphenylphosphinopropionyl)glycyl]cysteine Ligand with Poly(ethylene glycol) Roberta Visentin,† Gianfranco Pasut, Francesco Maria Veronese,* and Ulderico Mazzi* Department of Pharmaceutical Sciences, University of Padova, 35131 Padova, Italy. Received April 28, 2004; Revised Manuscript Received June 30, 2004

The PN2S N-(N-(3-diphenylphosphinopropionyl)glycyl)cysteine ligand was conjugated to methoxypoly(ethylene glycol)-amino (mPEG-NH2) 5 and 20 kDa to yield PN2S(Trt)-PEG5000 1 and PN2S(Trt)PEG20000 2, and then detritylated to PN2S-PEG5000 4 and PN2S-PEG20000 5. When an acidic solution of 99mTcO4- is added to 4 or 5 in solid form, a quantitative yield in a single labeled species, 99mTclabeled PN2S-PEG5000 9 and 99mTc-labeled PN2S-PEG20000 10, respectively, is obtained. The reaction occurs in less than 15 min at room temperature for 4 and 35 °C for 5. This labeling procedure avoids the use of an external reducing agent, and it is based on the amphiphilic properties of PN2S-PEGs. Once in water, 4 and 5 self-assemble in micelles, which catalyze the metal reduction by means of an electron pair transfer from the phosphorus to technetium. The [99mTcO]3+ species is then coordinated, and at micelle level, both the (P)ON2S and the PN2S coordinations are possible, as demonstrated by reacting 99mTc-gluconate and ReOCl3(PPh3)2 with 4 and 5 and with the oxidized analogous (P)ON2SPEG5000 6. Compounds 9 and 10 exhibited a high stability both in vitro and in vivo. Biodistribution studies in mice also indicated that PN2S linking and 99mTc labeling do not modify PEG behavior in water and in vivo since the polymer dictates the fate of the conjugate.

INTRODUCTION

In the last decades, a large number of techniques have been developed to label biomolecules with 99mtechnetium and 186/188rhenium (1, 2). In the most popular approach, namely, the bifunctional approach, the radionuclide is bound to the targeting molecule by means of a bifunctional chelating agent (BFCA) (3, 4). The BFCA comprises a donor atom set to strongly coordinate the radiometal and a functional group to be covalently attached to the targeting molecule, either directly or through a linker. Among the available BFCAs, N-(N-(3-diphenylphosphinopropionyl)glycyl)-S-tritylcysteine, bearing a PN2S set, was previously labeled with 99mTc leading to the pentacoordinated 99mTcO[PN2S]-OH (5). The labeled compound exhibited high stability in human plasma and in vivo, although the biodistribution was characterized by high hepatobiliary excretion, due to the lipophilicity of the complex. This suggested the need for chemical modification of the ligand to achieve a pharmacokinetics that favors renal clearance. Attempts to overcome the problems related to radiopharmaceutical in vivo behavior may include the use of polymeric linkers acting as pharmacokinetic modifiers. Among the available polymers, poly(ethylene glycol) (PEG) is the most successfully used in pharmaceutical science (6, 7). Thanks to its unique characteristics, like absence of immunogenicity and toxicity and high solubility either in aqueous or in organic solvents, PEG is extensively used for conjugation to proteins, peptides, and small drugs. It is able to transfer its properties to the * To whom correspondence should be addressed. Prof. U. Mazzi: Tel +39 049 827 5342; fax +39 049 827 5366; e-mail [email protected]. Prof. F. M. Veronese: Tel +39 049 827 5694; fax +39 049 827 5366; e-mail [email protected]. † E-mail address: [email protected].

conjugated molecules, and the products usually become less antigenic, less immunogenic, and more soluble with respect to the parent molecules. PEGylation also prevents recognition by antibodies and degradation by proteolitic enzymes and decreases renal clearance of small watersoluble molecules, prolonging the half-life of biomolecules. More recently, it was demonstrated that PEG gives to drugs passive tumor targeting by enhanced permeability and retention (EPR) effect (8). These considerations prompted us to synthesize a novel class of PEG conjugates bearing the PN2S chelating agent (PN2S-PEGs) and to investigate the coordination with 99m technetium and cold rhenium. During this study, a new labeling method based on the amphiphilic character of PN2S-PEGs and avoiding the use of an external reducing agent was discovered. In vivo studies are also reported to evaluate the change to the biodistribution of 99m Tc-complex afforded by PEGylation. EXPERIMENTAL SECTION

Materials. All chemicals were reagent grade and used without further purification. Amino acids and reagents were purchased from Novabiochem (Laufelfingen, Switzerland) and Fluka (Sigma-Aldrich, Milan, Italy); 3-diphenylphosphinopropionic acid succinimide ester was provided from Argus Spechem S.a.s. (Prato, Italy); mPEG-NH21 (MW 5 and 20 kDa) were purchased from Nektar Co. (Huntsville, AL). Rhenium was purchased from SigmaAldrich (Milan, Italy) as KReO4, and the precursor ReOCl3(PPh3)2 was synthesized according to the literature procedure (9). The synthesis of the N-[N-(3-diphenylphosphinopropionyl)glycyl]-S-trityl-L-cysteine (PN2S(Trt)-OH) ligand was performed according to the previously reported procedure (10). The syntheses of ligands

10.1021/bc049896d CCC: $27.50 © 2004 American Chemical Society Published on Web 08/19/2004

99mTc

labeling of PEGylated PN2S ligands

and complexes were carried out under an argon atmosphere using solvents degassed and tested for peroxides before use. Solvents for syntheses and analysis and HPLC eluents were purchased from Fluka (Sigma-Aldrich, Milan, Italy) and Carlo Erba Reagenti (Div. Antibioticos, Milan, Italy); MilliQ water was obtained from a Millipore system (Millipore, Vimodrone, Milan, Italy). CDCl3 for 1H NMR, 13 C NMR, and 31P NMR analysis was purchased from Sigma-Aldrich (Milan, Italy). Na99mTcO4 was eluted from a commercial Drytec Sorin 99 Mo/99mTc generator (Nycomed Amersham Sorin, Saluggia, Vercelli, Italy). Caution! 99Tc is a weak β--emitter (E ) 0.292 MeV, t1/2 ) 1.12 × 105 years). All manipulations were carried out in laboratories approved for lowlevel radioactivity use. Handling milligram amounts of 99 Tc does not present a serious health hazard since common laboratory glassware provides adequate shielding and all work is performed in approved and monitored hoods and gloveboxes. Bremsstrahlung is not a significant problem due to the low energy of the β- particles; however, proper radiation safety procedures must be followed at all times, and particular care should be taken when handling solid samples. Analytical Methods. 1H NMR spectra were obtained on a Bruker 300 MHz spectrometer, and chemical shifts (ppm) were referenced to tetramethylsilane (TMS). 31P NMR were recorded on a Bruker AC-200 spectrometer using 85% aqueous H3PO4 as external reference. Multiplicities were reported as s (singlet), d (doublet), dd (doublet of doublets), t (triplet), q (quadruplet), m (multiplet), and b (broad signal). RP-HPLC analysis was performed on a LKB-Bromma system (Pharmacia, Cologno Monzese, Italy) comprising a binary LKB 2249 programmable gradient pumps controlled by a gradient controller 680 automated equipped with a LKB-Bromma 2140 tunable absorbance detector (Pharmacia, Cologno Monzese, Italy) set at 215 nm, a Beckman model 170 radioisotope detector system (Beckman, Milano, Italy), and a data module integrating recorder. The UV chromatograms were elaborated by using the WAVESCAN EG software (Pharmacia, Cologno Monzese, Italy), and the UV-RA (radioactive) chromatograms were elaborated by using the PE Nelson 900 series interface software (Pharmacia, Cologno Monzese, Italy). For analytical injections, a Hamilton PRP-1 column (4.1 mm × 250 mm, Alltech Italia, Sedriano, Milan, Italy) was used with a 20 µL injection loop and equipped with a Hamilton PRP-1 precolumn (Alltech Italia, Sedriano, Milan, Italy). The elution method consisted of a linear gradient (0-3 min, 20% solvent B; 3-28 min, 20-60% solvent B; 28-30 min, 60-100% solvent B; 30-33 100% solvent B; 33-35 min, 100-20% solvent B) using a mobile phase of 0.1% TFA in water (solvent A) and 0.05% TFA in MeCN (solvent B) at a flow rate of 1 mL min-1. Mass spectra were obtained on a Voyager-DE matrixassisted laser desorption ionization time-of-flight (MALDI1 Abbreviations: CH Cl , dichloromethane; CHCl , chloro2 2 3 form; DCC, N,N′-dicyclohexylcarbodiimide; EtOAc, ethyl acetate; EtOH, ethanol; Et2O, diethyl ether; KReO4, potassium perrhenate; mPEG-NH2, methoxy-PEG-amino; MeCN, acetonitrile; Nagluconate, sodium gluconate; mPEG5000-NH2, methoxy-PEGamino MW 5 kDa; mPEG20000-NH2, methoxy-PEG-amino MW 20 kDa; Na99mTcO4-, sodium pertechnetate; PBS, phosphate buffer solution; NHS, N-hydroxysuccinimide; ReOCl3(PPh3)2, trans-oxotrichloro bistriphenylphosphine rhenium(V); saline, physiological solution (0.9% NaCl); SnCl2, tin(II) chloride; TEA, triethylamine; TFA, trifluoroacetic acid.

Bioconjugate Chem., Vol. 15, No. 5, 2004 1047

TOF) mass spectrometry system (PerSeptive Biosystems Inc, Framingham, MA). Data for 2 ns pulses of the 337 nm nitrogen laser were averaged for each spectrum in a linear mode, and a positive ion TOF detection was performed using an accelerating voltage of 25 kV. Synapinic acid was used as matrix. Labeling efficiency and radiochemical purity (RCP) were assessed by reverse-phase high-performance liquid chromatography (RP-HPLC) and thin-layer chromatography (TLC). The RP-HPLC peak of the labeled compound was recovered, and the activity of the collected fraction was measured with a dose calibrator M2361 Messelektronik (Dresden, Germany). RCP was expressed as the percent of injected activity. TLC was performed on Merck silica F254s plates. Ten microliters of the sample was applied at the application point 1 cm from the bottom and eluted over a distance of 7 cm from the bottom. Plates were then cut 4 cm above the application point and the activity of each part was measured using the abovementioned dose calibrator. The absence of free pertechnetate (99mTcO4-) in 99mTc-gluconate solution was assessed using acetone as eluent (99mTcO4- RF ) 1; 99mTcgluconate RF ) 0; 99mTc-colloid RF ) 0), whereas determination of reduced-hydrolyzed 99mTc (colloid) in the same solution was performed using saline (99mTc-gluconate RF ) 0.9-1; 99mTc-colloid RF ) 0). The percentage of colloid (C) was calculated with the following formula: C ) [(ATot - Agluc)/ATot] × 100. Radiochemical purity of 99m Tc-labeled PEGylated compounds was determined by eluting the TLC with saline (RF ) 0.5) and calculated with the following formula: RCP ) [(ATot - Agluc)/ATot] × 100 - C. Alternatively, the absence of 99mTc colloid was verified by subsequent analysis of the labeling mixture via HPLC to ensure a quantitative elution of activity from the column. PN2S(Trt)-PEG5000 (1). To a chilled solution (4 °C) of PN2S(Trt)-OH (198 mg, 0.30 mmol) in CH2Cl2 (10 mL) were added NHS (51 mg, 0.45 mmol) and DCC (185 mg, 0.90 mmol). The mixture was warmed to room temperature and stirred for 5 h under an argon atmosphere. Then mPEG5000-NH2 (1.00 g, 0.20 mmol) previously dissolved in CH2Cl2 (4 mL) and TEA (41 µL, 0.30 mmol) were added. The reaction was left to proceed for 10 h at room temperature. The mixture was then filtered, and the solution was dropped into EtOAc (150 mL) and cooled at 0 °C for 3 h. On cooling, the product precipitated, and the excess of PN2S(Trt)-OH remained in solution. The product collected on a funnel was dried under vacuum and purified by recrystallization from EtOAc. Yield: 90% (1.01 g, 0.18 mmol). 1H NMR (CDCl3): 2.24-2.85 (m, P(CH2)2, Cys-βCH2), 3.38 (s, OCH3, PEG), 3.64 (bs, (OCH2CH2)n, PEG), 4.25 (m, Cys-RCH), 7.24-7.52 (m, ArH). 31P NMR (CDCl3): -14.49. MALDI-TOF MS: bell-shaped spectrum centered at 5701 Da (theory 5642 Da). HPLC: tR ) 24.73 min. PN2S(Trt)-PEG20000 (2). The same procedure was followed as for the synthesis of compound 1, except for the following changes: PN2S(Trt)-OH (49 mg, 0.07 mmol); NHS (13 mg, 0.11 mmol); DCC (45 mg, 0.22 mmol); TEA (9 µL, 0.07 mmol); mPEG20000-NH2 (1.00 g, 0.05 mmol). Yield: 91% (0.94 g, 0.04 mmol).1H NMR (CDCl3): 2.322.87 (m, P(CH2)2, Cys-βCH2), 3.42 (s, OCH3, PEG), 3.71 (bs, (OCH2CH2)n, PEG), 7.20-7.47 (m, ArH). 31P NMR (CDCl3): -14.74. MALDI-TOF MS: bell-shaped spectrum centered at 20 986 Da (theory 20 642 Da). HPLC: tR ) 23.48 min. (P)ON2S(Trt)-PEG5000 (3). The procedure for the synthesis of compound 1 was followed, except for the following changes: (P)ON2S(Trt)-OH (203 mg, 0.30 mmol).

1048 Bioconjugate Chem., Vol. 15, No. 5, 2004

Yield: 94% (1.06 g, 0.18 mmol). 1H NMR (CDCl3): 2.482.90 (m, P(CH2)2, Cys-βCH2), 3.37 (s, OCH3, PEG), 3.68 (bs, (OCH2CH2)n, PEG), 4.32 (m, Cys-RCH), 7.17-7.88 (m, ArH). 31P NMR (CDCl3): 33.46. MALDI-TOF MS: bell-shaped spectrum centered at 5700 Da (theory 5658 Da). HPLC: tR ) 24.75 min. PN2S-PEG5000 (4). Compound 1 (100 mg, 1.7 × 10-2 mmol) was dissolved in CH2Cl2 (1 mL) and TFA (7 mL). Triethylsilane was added until the solution became colorless, and the final mixture was stirred for 45 min. TFA was removed under vacuum, and the resulting residue was crystallized as a white powder upon addition of cold EtOAc. Yield: 92% (88 mg, 1.6 × 10-2 mmol). 1H NMR (CDCl3): 2.31-2.89 (m, P(CH2)2, Cys-βCH2), 3.25 (s, OCH3, PEG), 3.68 (bs, (OCH2CH2)n, PEG), 4.56 (m, Cys-RCH), 7.38-7.64 (m, ArH). 31P NMR (CDCl3): -15.02. HPLC: tR ) 19.53 min. PN2S-PEG20000 (5). The procedure was followed as in the synthesis of compound 4, except for the following changes: compound 2 (100 mg, 0.5 × 10-2 mmol); TFA (15 mL). Yield: 90% (89 mg, 0.4 × 10-2 mmol). 1H NMR (CDCl3): 2.27-2.78 (m, P(CH2)2, Cys-βCH2), 3.36 (s, OCH3 PEG), 3.72 (bs, (OCH2CH2)n, PEG), 7.35-7.51 (m, ArH). 31P NMR (CDCl3): -14.91. HPLC: tR ) 20.02 min. (P)ON2S-PEG5000 (6). The same procedure for the synthesis of compound 4 was followed, except for the following change: compound 3 (100 mg, 1.7 × 10-2 mmol). Yield: 89% (85 mg, 1.5 × 10-2 mmol). 31P NMR (CDCl3): 36.76. HPLC: tR ) 19.77 min. ReO[PN2S]-PEG5000 (7). To a solution of 4 (250 mg, 4.6 × 10-2 mmol) in degassed CH2Cl2 (5 mL) was added a solution of ReOCl3(PPh3)2 (38 mg, 4.4 × 10-2 mmol) in the same solvent (3 mL). TEA was added to the mixture to pH 9, resulting in a reddish-brown solution. The final complex was crystallized by adding Et2O to the solution previously concentrated under vacuum. HPLC: tR ) 21.34 min. UV-vis (nm, CHCl3): 507. 1H NMR (CDCl3): 2.28-2.99 (m, P(CH2)2), 3.23 (s, OCH3, PEG), 3.71 (bs, (OCH2CH2)n, PEG), 4.30 (dd, 3J ) 11.1, Cys-βCH, anti isomer), 4.51 (d, 3J ) 20.4, Gly-CH, anti isomer), 4.71 (d, 3J ) 21.0, Gly-CH, syn isomer), 4.82 (d, 3J ) 20.4, Gly-CH, anti isomer), 4.88 (d, 2J ) 21.0, Gly-CH, syn isomer), 5.25 (d, Cys-RCH, anti isomer), 5.66 (dd, 3J ) 8.4, Cys-RCH, syn isomer), 7.40-7.74 (m, ArH). ReO[(P)ON2S]-PEG5000 (8). To a solution of 6 (50 mg, 0.9 × 10-2 mmol) in degassed CH2Cl2 (3 mL) was added a solution of ReOCl3(PPh3)2 (7 mg, 0.9 × 10-2 mmol) in the same solvent (2 mL). TEA was added to the mixture to pH 7, resulting in an orange solution. The final complex was crystallized by adding Et2O to the solution previously concentrated under vacuum. 99m Tc-labeled PN2S-PEG5000 (9). 99mTcO4- solution freshly eluted from generator was acidified to pH 2 (0.1 M HCl). One hundred microliters of the solution (1-2 mCi) was added to compound 4 (2 mg, 3.7 × 10-4 mmol) in a 1.5 mL eppendorf vial, and the final mixture was kept at room temperature for 10-15 min. Labeling yield > 95%. HPLC: tR ) 21.19 min. 99m Tc-labeled PN2S-PEG20000 (10). The same procedure was followed as in the synthesis of compound 9, except for the following changes: compound 5 (8 mg, 3.9 × 10-4 mmol); incubation at 37 °C for 10-15 min. Labeling yield > 95%. HPLC: tR ) 21.91 min. 99m Tc-gluconate. 99mTc-gluconate was prepared by adding 100 µL of freshly eluted 99mTcO4- (5-10 mCi) to 10 µL of 0.01 M Na-gluconate (saline) and 1 µL of 0.1 M SnCl2 (0.1 M HCl). The product was purified by Sep-pak chromatography. HPLC: tR ) 2.05 min.

Visentin et al. 99m

TcO[PN2S]-PEG5000 (11). To a sample of compound 4 (2 mg, 3.7 × 10-4 mmol) was added a solution of purified 99mTc-gluconate (50 µL, 1-2 mCi) diluted with absolute EtOH (250 µL). The pH was increased to 9-10 by addition of 0.1 M NaOH, and the mixture was kept at 37 °C for 45 min. Labeling yield > 95%. HPLC: tR ) 20.96. 99m TcO[PN2S]-PEG20000 (12). The labeling of compound 5 (8 mg, 3.9 × 10-4 mmol) was performed following the same procedure described for the synthesis of compound 11. Labeling yield > 95%. HPLC: tR ) 21.45. 99m TcO[(P)ON2S]-PEG5000 (13). The labeling of compound 6 (2 mg, 3.7 × 10-4 mmol) was performed following the same procedure described for the synthesis of compound 11. Labeling yield > 95%. HPLC: tR ) 20.67. Light Scattering Measurements. The tendency of 4 and 5 to self-associate in micellar aggregates in aqueous solutions was evaluated by dynamic light scattering (DLS) experiments performed at 25 °C using a Nicomp 170 computing autocorrelator (Pacific Scientific, Silver Spring, MD) equipped with a Spectra Physics Stabilite 2017 (Spectra Physics, Mountain View, CA; polarized argon laser 496 nm). A solid sample of the compound was directly dissolved in PBS (pH 4, 7.4, and 10) or in saline (pH 2, 0.1 M HCl) to a final concentration of 20 mg/mL for 4 and 80 mg/mL for 5. DLS analysis was also employed to assess the presence of micellar aggregates in the labeling mixtures containing compounds 9 and 10. Critical Micellar Concentration. Critical micellar concentration of 4 and 5 was evaluated by fluorescence experiments using pyrene as fluorescence probe following a literature procedure (11). An amount of PEGylated compound was added to a volumetric flask, and degassed saline (pH 2, 0.1 M HCl) was added to give a final concentration of 20.0 g/L. The solution was diluted to yield stock solutions at different conjugate concentrations from 20.0 to 0.05 g/L. A known amount of pyrene in acetone was added to each of a series of 10 mL volumetric flasks, and the acetone was removed by evaporation. The amount of pyrene was chosen to give a final concentration of 6 × 10-7 M, close to the saturation solubility of pyrene in water at 22 °C, or 1.2 × 10-7 M. To each flask, a stock solution (10 mL) of PEGylated compound was then added. The closed flasks were heated for 3 h at 65 °C to equilibrate pyrene and micelles and subsequently allowed to cool overnight in the dark at room temperature. For measurements, 3 mL of each solution was placed in a quartz cuvette, and the fluorescence (λex ) 339 nm) and excitation (λem ) 390 nm) spectra were recorded on a Shimatzu RF-540 spectrofluorophotometer at 25 °C and accumulated with an integration time of 1s/0.5 nm. Specific Activity Determination. To 99mTcO4- solution freshly eluted from generator (100 µL, 1-2 mCi) was added a standard solution of 99TcO4- (99Tc is the carrier) in saline (9 µL). The final mixture was acidified to pH 2 (0.1 M HCl) and added to 4 (2 mg, 3.7 × 10-4 mmol) in a 1.5 mL eppendorf vial. The solution was kept at room temperature for 10-15 min. The experiment was performed using 3.95 × 10-2, 3.95 × 10-3, and 3.95 × 10-4 M 99TcO4- solutions to have a 99Tc/PN2S-PEG5000 molar ratio of 1:1, 1:10, and 1:100. The specific activity was determined with the following formula: specific activity ) A (MBq)/PN2S-PEG5000 (µmol), where A is the activity when no free 99mTcO4- is detectable in the labeling mixture assuming the whole Tc (99mTc and 99Tc) present in the solution as 99mTc, since 99mTc and 99Tc are chemically indistinguishable. The amount of generator eluate 99m Tc being negligible, A can be calculated by using the following formula: A ) [(nTcNAV)/Nt] × 37, where nTc )

99mTc

labeling of PEGylated PN2S ligands

Tc moles, NAV ) Avogadro’s number (6.022 × 1023), Nt ) number of 99mTc atoms that correspond to 1 mCi of activity (1.16 × 1012), and 37 is the conversion factor between mCi and MBq (1 mCi ) 37 MBq). In Plasma Stability. The stability of 9 and 10 was evaluated in human plasma. Aliquots of the labeling mixture (30 µL, 500 µCi) were diluted with freshly prepared human plasma from healthy donors (250 µL) and incubated at 37 °C for various times (1, 2, 4, and 6 h). After incubation, protein fractions in the samples were precipitated with MeCN/EtOH (1/1, 50 µL) and centrifuged (4 °C, 6000 rpm). After repeated precipitation and centrifugation, the supernatant was analyzed by RPHPLC. Data are expressed as percentage of activity corresponding to intact complex and are means ( SD of five experiments. The stability was assessed in the presence and in the absence of micelles. To reach a final concentration of 4 and 5 in the plasma sample under the critical micellar concentration, the labeling mixture was conveniently diluted before incubation. In Vivo Biodistribution. Compounds 9 and 10 were studied in vivo. All animal experiments were carried out in accordance with national regulations. Male Swiss mice (25-28 g) were from our department. Animals were anaesthetized by intraperitoneal injection of 100 µL of a solution containing tiletamine hydrochloride/zolazepam hydrochloride (1/1, 40 mg/kg) and xilazine (2 mg/kg). The biodistribution was evaluated by intravenous administration of labeled compounds via the tail vein (n ) 3). Before injection, the labeling mixture was neutralized (1 M NaOH) and diluted with saline to a final concentration in PEGylated compound of 5 and 20 mg/mL for 4 and 5, respectively. The biodistribution was evaluated by administration of about 700 µCi/100 µL. Scintigraphic images were collected using a YAP-camera, a γ-camera optimized to perform experiments on small animals with a spatial resolution of 1.0-1.2 and a 40 × 40 mm2 field of view (12). Acquisition of images started 5 min postinjection and images of 1 min were acquired for a total time of 40 min and 3 h after administration of 9 and 10, respectively. Collected images were elaborated in 10 min images using the Scion Image program. After scintigraphic analysis, tissues and organs were excised from the sacrificed animals and weighed, and the radioactivity counts were determined with a dose calibrator. The blood was withdrawn from the heart through a syringe immediately after sacrifice and counted. Data are expressed as percentage of injected dose (% ID) and of injected dose per gram of tissue (% ID/g). The activity in whole blood was estimated assuming a blood volume of 6.5% of the total body weight.

Bioconjugate Chem., Vol. 15, No. 5, 2004 1049

99

RESULTS AND DISCUSSION

Synthesis of PEGylated Ligands. The PEGylation of the PN2S(Trt)-OH ligand and of the oxidized analogous (P)ON2S(Trt)-OH was achieved at the level of the cysteine carboxylic group, following DCC/NHS activation in dichloromethane to couple to mPEG-NH2 5 and 20 kDa. PN2S(Trt)-PEG5000 1, PN2S(Trt)-PEG20000 2, and (P)ON2S(Trt)PEG5000 3 (Figure 1) were obtained in high yield (>90%). The conjugates showed high solubility in water where both PN2S(Trt)-OH and (P)ON2S(Trt)-OH and detritylated derivatives were insoluble. The purity and the identity of the products were assessed by RP-HPLC, MALDI-TOF spectrometry, 1H NMR, and 31P NMR. Laser desorption mass spectrometry gave a bell-shaped spectrum centered at 5701 Da for 1, at 20 986 Da for 2, and at 5700 Da for 3 with a family of adjacent peaks at

Figure 1. Structures of PEGylated PN2S (1) and (P)ON2S (3) ligands and structures of detritylated derivatives (4 and 6) used for 99mTc-labeling and Re-coordination studies.

44 ( 1 Da (the ethylene oxide unit) due to the polydispersity of PEG. The 31P NMR spectra showed a singlet at -14.49 ppm for 1 and at -14.74 ppm for 2, characteristic of a phosphine phosphorus, and at 33.46 ppm for 3, confirming the presence of completely oxidized phosphorus. The 1H NMR signals of the PN2S or (P)ON2S set protons were detectable despite the broadness of PEG methylene signal, the ligand proton resonances being upshielded or deshielded with respect to PEG chemical shifts. The RP-HPLC chromatogram, as revealed at 215 nm, showed a single homogeneous peak at 24.73, 23.48, and 24.75 min for 1, 2 and 3, respectively. To perform the complexation reactions with 99mtechnetium and 185/187rhenium, the conjugates were detritylated with trifluoroacetic acid and triethylsilane (13) to yield PN2S-PEG5000 4, PN2S-PEG20000 5, and (P)ON2SPEG5000 6. The products were crystallized from degassed ethyl acetate to avoid the oxidation of phosphorus or the dimerization of cysteine. The purity of compounds was assessed by RP-HPLC, showing a single homogeneous peak at 19.49, 20.02, and 19.77 min for 4, 5, and 6, respectively. Direct Labeling of PN2S-PEGs. The use of phosphine ligands as reducing agents for 99mTcO4- is reported in the literature (14, 15), but the reaction usually requires harsh conditions in terms of time and heating. Instead, it was observed that when an acidic solution of 99mTcO4(pH 2) is added to compound 4 or 5 in solid form, a quantitative yield in a single labeled species, 9 and 10, respectively, is obtained, and the reaction occurs in less than 15 min at room temperature for 4 and 35 °C for 5. The chromatograms depicted in Figure 2 are related to the labeling of 4 10 min after TcO4- addition. The peak revealed by UV at 19.49 min, the same of 4 before labeling, indicates the stability of PEGylated PN2S ligand under reaction conditions. The unique peak in the RA chromatogram at 21.19 min indicates the radiochemical purity of the labeled compound 9 and the absence of free 99m TcO4- (tR ≈ 3 min). The higher retention time observed for 9 is in agreement with the increased lipophilicity of the labeled species due to metal coordination. Compound 9 is not assessable by UV detector since 99mTc concentration in the labeling mixture is in the range of picomoles.

1050 Bioconjugate Chem., Vol. 15, No. 5, 2004

Visentin et al.

Figure 2. Direct labeling of PN2S-PEG5000 4. RP-HPLC analysis of the labeling mixture 10 min after addition of TcO4(pH 2) to solid 4 at room temperature is presented. The peak in the UV chromatogram is related to compound 4. The RA chromatogram indicates the presence of a single labeled species 9 and the absence of free TcO4- (tR ≈ 3 min). Table 1. Dynamic Light Scattering Analysis (25 °C, λ ) 496 nm) of PN2S-PEG5000 4 (20 mg/mL) and PN2S-PEG20000 5 (80 mg/mL) in Saline (pH 2, 0.1 M HCl), PBS (pH 4, 7.4, and 10), and the Labeling Mixture (pH 2) Leading to 99mTc-Labeled PN S-PEG 99mTc-Labeled 2 5000 9 and PN2S-PEG20000 10, Respectivelya pH

4

4b

5

5b

2 4 7.4 10

316.7 325.2 318.8 391.6

125.7

2854.4 3029.1 2757.6 2591.9

2057.3

a Values represent the mean diameter (nm) of micellar aggregates. b Labeling mixture.

Similar chromatograms were obtained when the direct labeling of compound 5 was performed. It was observed that no labeling takes place if 99mTcO4is added to the unconjugated PN2S-OH ligand or to a mixture of the ligand and mPEG-NH2 non-covalently bound. Furthermore, the presence of phosphorus as phosphine and the integrity of the PN2S set are essential, since the labeling does not occur when the conjugate is oxidized to 6 or dimerized through a disulfide bridge. It was also found that the mixing conditions of TcO4with the PEGylated ligand are very critical: the reaction is faster when the TcO4- solution is added to PN2S-PEG in solid state than when the two reagents are solubilized separately and mixed. In this case, the labeling was still incomplete after 1 day. A reasonable explanation of all these results is that the reduction and coordination of the metal take place thanks to a micellar catalysis due to self-aggregation of PN2S-PEGs in water. In fact, the conjugates have an amphiphilic character, the PN2S moiety being highly lipophilic and the polymeric PEG chain being highly hydrophilic, and it is well-known that micelles can catalyze redox reactions (16). Light scattering measurements indicated that 4 and 5 effectively aggregate in water solutions to yield micelles (Table 1) with a cmc in the range 1-0.5 g/L. The cmc was determined in acidic saline to resemble the labeling reaction conditions using pyrene as a fluorescent probe in the presence of increasing concentrations of PN2SPEG conjugate (11). The characteristic feature of pyrene excitation spectra, namely, the (0,0) band shift from 334 to 337 nm upon probe partition into the inner core of

Figure 3. Determination of critical micellar concentration of PN2S-PEG5000 4 using physically entrapped pyrene as a fluorescent probe. The excitation spectra of pyrene, monitored at 390 nm in saline (pH 2) at 25 °C in the presence of increasing concentrations of 4, show the shift in the (0,0) band as pyrene partitions between aqueous and micellar environments.

micelles, was considered to be indicative of cmc achievement (Figure 3). PN2S-PEG micelles were found to easily disassemble by dilution. By dilution of the concentrated solutions below the cmc just before fluorescence analysis, the shift of the (0,0) band from 337 to 334 nm was observed, in agreement with a rapid disaggregation of micelles. Micelles were also present in the labeling mixtures (Table 1), and the micellar state of the conjugate is necessary since no technetium labeling was obtained below the cmc. On the other hand, when 99mTcO4- is added to previously solubilized PN2S-PEG, the reaction occurs but at much lower rate since the PN2S moiety, forming the inner lipophilic core of the already shaped micelles, is not available for the reduction and coordination of the metal. However, since micelles are in fluxional equilibrium with monomers, 99mTcO4- can still be loaded into micelles where the redox reaction rapidly takes place, but this internalization process is slow and difficult to complete. Furthermore, the catalytic effect mediated by micelles explains the efficiency of the labeling reaction, quantitative in less than 15 min, and the possibility to perform the reaction at acid pH. This pH, although favoring the redox reaction, is not as convenient as the alkaline usually required for the deprotonation of peptide ligand amides (17). Finally, the metal is not simply entrapped into micelles as 99mTcO4- but is strictly bound to PN2S-PEG by coordination followed by reduction, since 99mTcO4- is not released through micelle dissociation below the cmc, as detected under HPLC conditions. In conclusion, we may state that when 99mTcO4- is added to the solid 4 or 5, the conjugate dissolves, associating into micelles, while the anion 99mTcO4- binds to the phosphine phosphorus through an electrostatic interaction, since at acidic pH the phosphorus is present as phosphonium, as depicted in Figure 4.

99mTc

labeling of PEGylated PN2S ligands

Bioconjugate Chem., Vol. 15, No. 5, 2004 1051

Figure 4. Direct labeling of PEGylated PN2S ligands. A drawing of the mechanism of PN2S-PEG micelle aggregation with the two possible coordination species is shown.

The Coulombic interaction favors the metal accumulation in the micelle environment where the reduction reaction occurs by means of an electron pair transfer from the phosphine phosphorus to technetium to give the [99mTcO]3+ species. It is presumed that the ligand acts as a formal two-electron reducing agent in the presence of TcVIIO4- and that, upon reduction of TcVII to TcV, the PIII center is oxidized to the corresponding phosphine oxide, PV. The reduction is then followed by the coordination. The [99mTcO]3+ core could be coordinated by the same ligand molecule involved in the redox reaction by means of the new (P)ON2S set, or alternatively, it could be coordinated through the PN2S set of a second ligand molecule not involved in the reduction and included in the same micelle. Transchelation Labeling of PN2S-PEGs. To evaluate the above-reported coordination hypotheses, both 4 and 5 and the oxidized analogous 6 were employed in the labeling reaction with 99mTc-gluconate. In this case, the reaction involves the reduction of 99mTcO4- to [99mTcO]3+ by Sn(II) in the presence of sodium gluconate to form the intermediate 99mTc-gluconate, which later exchanges with the ligand of interest to give the final complex. Preliminary studies indicated the need to purify 99mTcgluconate from technetium-hydrolyzed species leading to unspecific absorption to PEG chain. On the other hand, no interaction between 99mTc-gluconate and mPEG-NH2 was observed. The labeling reaction carried out by the addition of 99m Tc-gluconate to 4 or 5 dissolved in water or in the solid state at a final concentration above the cmc was found to proceed but with a slow rate, since it was still incomplete after 4 h. This indicated that micelles interfere with the exchange ligand reaction, since the PN2S moiety, which forms the lipophilic core, is less available for the coordination. This was not the case for the reaction carried out in water/EtOH 1:5 (v/v) to avoid micelle formation, a condition that yielded in 35-45 min to a single homogeneous peak detected by HPLC at 20.85 and 21.45 min for 4 and 5, respectively. The peaks were attributed to 99mTcO[PN2S]-PEG5000 11 and 99mTcO[PN2S]PEG20000 12 obtained from the transchelation between 99m Tc-gluconate and the PN2S set. The reaction performed using compound 6 led to analogous results with a single labeled species detected at 20.67 min. In this case, the [99mTcO]3+ core is supposed to be coordinated by the (P)ON2S set with the phosphine oxide oxygen bound to the metal center instead of the phosphine phosphorus, leading to the 99mTcO[(P)ON2S]-PEG5000 complex 13.

99mTc

reduction-coordination mediated by

These results confirmed that at micelle level both coordinations may take place. Rhenium Coordination Studies. To further confirm the two coordination possibilities, PEGylated PN2S and (P)ON2S were reacted with rhenium. In analogy to a previously reported method for the synthesis of the ReO[PN2S]-OMe complex (10), the reaction of 4 was performed in dichloromethane using ReOCl3(PPh3)2 as rheniumV oxo starting material. As observed with the PN2SOMe ligand, the solution, initially orange-green, turned to reddish-brown by triethylamine addition to reach alkalinity. This demonstrated that PN2S-PEG5000 coordinates [ReO]3+ leading to the pentacoordinated complex ReO[PN2S]-PEG5000 7. The 1H NMR analysis indicated that both the syn and the anti ReO[PN2S]-PEG5000 complexes are formed, although the HPLC did not separate the two isomers. In the range 4-6 ppm, the signals related to cysteine (Cys) and glycine (Gly) protons were clearly assigned and found to be close to those detected in the syn and anti ReO[PN2S]-OMe spectra. The Cys R protons were detected at 5.25 and 5.66 ppm for anti and syn isomer, respectively. As for the ReO[PN2S]-OMe anti isomer, the splitting pattern of anti isomer Cys R proton appeared as a doublet, indicating that it was at a nearly right angle to one of the β protons, thereby giving no coupling constant. On the other hand, syn isomer Cys R proton appeared as a doublet of doublets indicating that the torsion angles between the Cys R and β protons were different from those in the anti isomer. The Gly R proton signals were detected as two doublets (4.51 and 4.82 ppm, anti isomer; 4.71 and 4.88 ppm, syn isomer), with a geminal coupling of 20.4 and 21.0 Hz for anti and syn isomer, respectively. The β protons of the Cys residue were overlapped to the broad signal of PEG methylene, except for one detected as a doublet at 4.30 ppm and related to anti isomer. Different from ReO[PN2S]-OMe isomers, always obtained in a 1:1 molar ratio, PN2S-PEG coordination reaction led to a preference for the anti isomer (70-75%). It is reasonable that the presence of the PEG chain favors, by steric hindrance, one of the diastereomers. The ReOCl3(PPh3)2 precursor was reacted also with oxidized 6, and the ReO[(P)ON2S]-PEG5000 8 complex was obtained and found to be stable both in the solid state and in solution. The yield was significantly lower than that achieved with 7, and unfortunately, we did not succeed to remove the large amount of unreacted PN2SPEG5 by diethyl ether crystallization. This made a good characterization of the product impossible, since in the 1 H NMR spectrum the strong intensity of PEG OCH3 and

1052 Bioconjugate Chem., Vol. 15, No. 5, 2004

Visentin et al.

Figure 6. In human plasma stability of 99mTc-labeled PN2SPEG5000 9 after incubation at 37 °C at a concentration above (a) and below (b) the PN2S-PEG5000 cmc, compared with the stability of unPEGylated 99mTcO[PN2S]-OMe (c) and 99mTcO[PN2S]-OH (d). Values represent the means ( SD (n ) 5) of the percentage of intact labeled compound.

Figure 5. Direct labeling of PEGylated PN2S ligands: determination of 99mTc-labeled PN2S-PEG5000 specific activity through carrier added (99Tc) experiments. At the 99Tc/PN2S-PEG5000 molar ratio 1:100, the 99mTcO4- peak disappears suggesting that all the pertechnetate present in the labeling mixture (99TcO4and 99mTcO4-) is reduced and coordinated. A 1.9 TBq/µmol specific activity was calculated considering the whole Tc present in the solution as 99mTc.

CO(CH2)2-CO signals at 3.27 and 3.62 ppm, respectively, hampered the detection of the peaks related to coordinated Gly-Cys moiety. On the other hand, the same coordination events, initial green solution, which changed to orange by addition of TEA, were observed by reacting the (P)ON2S-OMe ligand with [ReO]3+ as for (P)ON2SPEG5000 (unpublished results). In agreement with the results obtained with 99mTcgluconate, rhenium-oxo studies confirmed that at micelle level both the PN2S and (P)ON2S coordinations are possible. Specific Activity Determination. The specific activity of 9 was evaluated by means of carrier-added experiments. Different amounts of 99TcO4- were added as carrier to 99m TcO4- solution. The final mixtures were acidified to pH 2 and added to 4 in the solid state. The solutions were kept at room temperature for 10-15 min. The experiment was performed in triplicate using 3.95 × 10-2, 3.95 × 10-3, and 3.95 × 10-4 M 99TcO4- solutions to have a 99Tc/ PN2S-PEG5000 molar ratio of 1:1, 1:10, and 1:100. Figure 5 indicates that, when the molar ratio was in the range 1:1-1:10, the labeling reaction did not occur completely, since 99mTcO4- is still detectable at about 3 min. Instead, at the molar ratio 1:100, the 99mTcO4- peak disappeared, and it was assumed that all of the pertechnetate present in the labeling mixture (99TcO4- and 99mTcO4-) was reduced and coordinated. Based on the fact that 99mTc and 99Tc are chemically indistinguishable, the specific activity was calculated considering the whole Tc present in the solution as 99mTc, yielding the considerably high value of 1.9 TBq/µmol. In Vitro Plasma Stability. The in plasma stability of 9 obtained through the direct labeling procedure was

evaluated by incubation at two concentrations, above and below the PN2S-PEG5 critical micellar concentration. The results depicted in Figure 6 demonstrated that 9 incubated in absence of micelle exhibited a stability analogous to those already observed for unPEGylated 99m TcO[PN2S]-OH and 99mTcO[PN2S]-OMe (5), while in the presence of micelles, the degradation was markedly reduced. As for the unPEGylated labeled compounds, the decomposition of 9 is due to the release of TcO4- and not to byproducts of backbone fragmentation, as demonstrated by RP-HPLC analysis. These experiments demonstrated that PEGylation did not affect the ability of the coordination set to stabilize [TcO]3+ toward oxidation or exchange reactions and that micelles further protect the complex by inclusion in the inner core. In Vivo Biodistribution. The labeled compounds 9 and 10 obtained by direct labeling were studied in vivo to evaluate stability, blood residence time, and biodistribution. In Figure 7, the scintigraphic images related to the their biodistribution are depicted. Scintigraphic images related to 9 were collected for 40 min and showed a rapid and efficient clearance from the bloodstream mainly by the urinary system, also confirmed by ex vivo counting of activity in organs and tissues (Table 2). Those data indicated that the biodistribution of labeled PN2S-PEG5000 is analogous to that reported in the literature for free PEG5000 (18, 19) and that the PEGylation of the PN2S ligand drastically reduces its clearance through the hepatobiliary route (5). For the biodistribution of 10, the image acquisition time was extended to 3 h. A slow clearance from the blood pool was found, in agreement with a high retention of activity in organs and tissues, as confirmed by the ex vivo counting (Table 3). Also in this case, the biodistribution of labeled PN2S-PEG20000 was similar to that reported for free PEG20000 (18, 19), demonstrating that PN2S linking and labeling do not modify the polymer behavior and that the polymer dictates the fate of the conjugate. The observed biodistribution of both 99mTc-PN2S-PEG 5000 and 20000 strictly correlates with that of the unmodified PEG with the same molecular weight, indicating that the excreted species are monomeric. This is also in agreement with the high degree of dilution in the blood pool, which causes micelle disaggregation. The low activity found in the stomach and the recovery of unmodified labeled conjugates in urine further demonstrated the high affinity of the coordination set for technetium and the stability toward degradation, also due to the shielding effect afforded by PEG to enzymatic

99mTc

labeling of PEGylated PN2S ligands

Bioconjugate Chem., Vol. 15, No. 5, 2004 1053

Figure 7. Scintigraphic images related to in mouse biodistribution of 99mTc-labeled PN2S-PEG5000 9 (top) and 99mTc-labeled PN2SPEG20000 10 (bottom) obtained through the direct labeling method. Table 2. Biodistribution in Mice of 99mTc-Labeled PN2S-PEG5000 9 at 40 min Postinjectiona organ

% ID

% ID/g

right kidney left kidney bladder heart bloodb stomach intestines liver lungs spleen

1.84 ( 0.42 1.86 ( 0.34 47.21 ( 4.19 0.42 ( 0.06 13.36 ( 1.02 0.55 ( 0.07 4.73 ( 0.11 3.89 ( 0.48 0.59 ( 0.19 0.21 ( 0.03

8.58 ( 2.46 9.00 ( 1.46 2.52 ( 0.41 6.73 ( 0.27 0.86 ( 0.35 4.38 ( 1.33 2.42 ( 0.36 2.20 ( 0.60 2.91 ( 0.53

a Values represent means ( SD (n ) 3) of the percent of injected dose (% ID) and of injected dose per gram (% ID/g) of tissue. b Total blood was estimated to be 6.5% of the body weight.

Table 3. Biodistribution in Mice of 99mTc-Labeled PN2S-PEG20000 10 at 3 h Postinjectiona organ

% ID

% ID/g

right kidney left kidney bladder heart bloodb stomach intestines liver lungs spleen

1.59 ( 0.12 1.52 ( 0.07 29.93 ( 1.93 0.65 ( 0.19 20.88 ( 1.76 0.84 ( 0.26 8.03 ( 0.89 5.83 ( 1.58 1.17 ( 0.16 0.16 ( 0.05

7.30 ( 0.70 7.28 ( 0.74 3.85 ( 1.03 11.47 ( 1.66 1.84 ( 0.60 7.23 ( 1.77 3.54 ( 0.52 4.32 ( 0.50 1.19 ( 0.26

a Values represent means ( SD (n ) 3) of the percent of injected dose (% ID) and of injected dose per gram (% ID/g) of tissue. b Total blood was estimated to be 6.5% of the body weight.

cleavage of amide bonds or hydrolysis by water often observed in PEG-peptide drugs. CONCLUSIONS

The PEGylation of the PN2S ligand allows a new simple, clean, and efficient 99mTc-labeling procedure and

avoids the use of an external reducing agent. The mechanism of labeling is based on a supramolecular assembly of the PN2S-PEG conjugates in aqueous solution to yield micelles that favor the reduction and coordination of technetium. The labeling does not modify the polymer behavior in water, and PEG may still convey to the labeled conjugates its pharmacokinetic properties, thanks also to the great in vivo stability of the complex. This novel method can therefore represent an useful tool to further improve the potential of technetium-99m in diagnosis and of rhenium-186/188 in radiotherapy through the labeling of targeting biomolecules conjugated to the PN2S set via a bifunctional PEG. Studies are now directed to (i) verify the accumulation of monofunctional PEG conjugates in solid tumors exploiting the EPR effect taking advantage of the high steric hindrance of the molecules and (ii) evaluate the new labeling procedure in bioconjugates with the PN2S set bound to specific targeting peptides using heterobifunctional PEGs. ACKNOWLEDGMENT

The work originated from collaboration within the program COST-B12 WG-5. Gianfranco Pasut was supported by University of Padova Grant CPDG034599 “Progetto Giovani Ricercatori 2003”. The authors thank Dr. Fabrizio Mancin for assistance with light scattering analysis, Dr, Lorena Giovagnini for 31P NMR spectra, Mr. Michele Bello for scintigraphic images elaboration, and Dr. Guido Bordignon for MALDI-TOF spectrometry. LITERATURE CITED (1) Liu, S., and Edwards, S. D. (1999) 99mTc-Labeled small peptides as diagnostic radiopharmaceuticals. Chem. Rev. 99, 2235-2268. (2) Hom, R. K., and Katzenellenbogen, J. A. (1997) Technetium99m-labeled receptor-specific small-molecule radiopharmaceuticals: recent developments and encouraging results. Nucl. Med. Biol. 24, 485-498.

1054 Bioconjugate Chem., Vol. 15, No. 5, 2004 (3) Jurisson, S. S., and Lydon, J. D. (1999) Potential technetium small molecule radiopharmaceuticals. Chem. Rev. 99, 22052218. (4) Fichna, J., and Janecka, A. (2003) Synthesis of targetspecific radiolabeled peptides for diagnostic imaging. Bioconjugate Chem. 14, 3-17. (5) Visentin, R., Giron, M. C., Bello, M., and Mazzi, U. (2004) Technetium-99m labeling of N-[N-(3-Diphenylphosphinopropionyl)glycyl]cysteine (PN2S-OH) and its methyl ester derivative (PN2S-OMe). Nucl. Med. Biol. 31, 655-662. (6) Harris, J. M., and Chess, R. B. (2003) Effect of PEGylation on pharmaceuticals. Nat. Rev. Drug Discoveries 2, 214-221. (7) Pasut, G., Guiotto, A., and Veronese, F. M. (2004) Protein, peptide and non-peptide drug PEGylation for therapeutical application: a review. Expert Opin. Ther. Pat. 14, 859-894. (8) Seymour, L. W., Ulbrich, K., Steyger, P. S., Brereton, M., Subr, V., Strohalm, J., and Duncan, R. (1994) Tumouritropism and anticancer efficacy of polymer-based doxorubicin prodrugs in the treatment of subcutaneous murine B16F10 melanoma. Br. J. Cancer 70, 636-641. (9) Chatt, J., and Rowe, G. A. (1962) Complex compounds of tertiary phosphines and a tertiary arsine with rhenium(V), rhenium(III), and rhenium(II). J. Chem. Soc. 4019-4033. (10) Visentin, R., Rossin, R., Giron, M. C., Dolmella, A., Bandoli, G., and Mazzi, U. (2003) Synthesis and characterization of rhenium(V)-oxo complexes with N-[N-(3-diphenylphosphinopropionyl)glycyl]cysteine methyl ester. X-ray crystal structure of {ReO[Ph2P(CH2)2C(O)-Gly-Cys-OMe,(P,N,N,S)]}. Inorg. Chem. 42, 950-959. (11) Wilhelm, M., Zhao, C., Wang, Y., Xu, R., Winnik, M. A., Mura, J., Riess, G., and Croucher, M. (1991) Poly(styreneethylene oxide) block copolymer micelle formation in water: A fluorescence probe study. Macromolecules 24, 1033-1040.

Visentin et al. (12) Vittori, F., Malatesta, T., and de Notaristefani, F. (1997) The YAP Camera: an accurate gamma camera particularly suitable for new radiopharmaceuticals research. IEEE Trans. Nucl. Sci. 44, 47-53. (13) Pearson, D. A., Blanchette, M., Baker, M. L., and Guindon, C. A. (1989) Trialkylsilanes as scavengers for the trifluoroacetic acid deblocking of protecting groups in peptide synthesis. Tetrahedron Lett. 30, 2739-2742. (14) Smith, C. J., Katti, K. V., Higginbotham, C., and Volkert, W. A. (1997) In vitro and in vivo characterization of novel water-soluble dithio-bisphosphine 99mTc complexes. Nucl. Med. Biol. 24, 685-691. (15) Freiberg, E., Davis, W. M., Nicholson, T., Davison, A., and Jones, A. G. (2002) Reduction of the pertechnetate anion with bidentate phosphine ligands. Inorg. Chem. 41, 5667-5674. (16) Tascioglu, S. (1996) Micellar solutions as reaction media. Tetrahedron 52, 11113-11152. (17) Vanbilloen, H. P., Bormans, G. M., De Roo, M. J., and Verbruggen, A. M. (1995) Complexes of technetium-99m with tetrapeptides, a new class of 99mTc-labeled agents. Nucl. Med. Biol. 22, 325-338. (18) Working, P. K., Newman, M. S., Johnson, J., and Cornacoff, J. B. (1997) Safety of Poly(ethylene glycol) and Poly(ethylene glycol) derivatives. Poly(Ethylene Glycol): Chemistry and Biological Applications (Harris, J. M., and Zalipsky, S., Eds.) pp 44-57, American Chemical Society Symposium Series 680, American Chemical Society, Washington, DC. (19) Yamaoka, T., Tabata, Y., and Ikada, Y. (1994) Distribution and tissue uptake of poly(ethylene glycol) with different molecular weights after intravenous administration to mice. J. Pharm. Sci. 83, 601-606.

BC049896D