PNP = Diphosphane Ligand - ACS Publications - American Chemical

Jan 19, 2008 - Physiology, Medical Faculty Carl Gustav Carus, Technical University Dresden, 01307 Dresden, Germany, Institute of. Radiopharmacy ...
1 downloads 0 Views 721KB Size
450

Bioconjugate Chem. 2008, 19, 450–460

Labeling of Fatty Acid Ligands with the Strong Electrophilic Metal Fragment [99mTc(N)(PNP)]2+ (PNP ) Diphosphane Ligand) Emiliano Cazzola,† Elisa Benini,† Micol Pasquali,† Peter Mirtschink,‡ Martin Walther,§ Hans-Jurgen Pietzsch,‡ Licia Uccelli,† Alessandra Boschi,† Cristina Bolzati,4 and Adriano Duatti*,† Laboratory of Nuclear Medicine, Department of Radiological Sciences, University of Ferrara, 44100 Ferrara, Italy, Institute of Physiology, Medical Faculty Carl Gustav Carus, Technical University Dresden, 01307 Dresden, Germany, Institute of Radiopharmacy, Forschungszentrum Dresden-Rossendorf, PF 510119, 01314 Dresden, Germany, and ICIS-CNR, Corso Stati Uniti, 4, 35127 Padova, Italy. Received June 25, 2007; Revised Manuscript Received November 28, 2007

The electrophilic metal fragment [99mTc(N)(PNP)]2+ (PNP ) diphosphane ligand) has been employed for the labeling of fatty acid chains of different lengths. To provide a site-specific group for the attachment of the metallic moiety, the fatty acid derivatives were functionalized by appending a bis-mercapto or, alternatively, a dithiocarbamato π-donor chelating systems to one terminus of the carbon chain to yield both dianionic and monoanionic bifunctional ligands (L). The resulting complexes, [99mTc(N)(PNP)(L)]0/+, exhibited the usual asymmetrical structure in which a TctN group was surrounded by two different bidentate chelating ligands. Dianionic ligands gave rise to neutral complexes, while monoanionic ligands afforded monocationic species. Biodistribution studies were carried out in rats. An isolated perfused rat heart model was employed to assess how strucural changes in the radiolabeled fatty acid compound affect the myocardial first pass extraction. Results showed that only monocationic complexes accumulated in myocardium to a significant extent. Conversely, neutral complexes were not efficiently retained into the heart region and rapidly washed out. In isolated perfused rat heart experiments, monocationic complexes exhibited a behavior similar to that of the monocationic flow tracers 99m Tc-MIBI and 99mTc-DBODC with almost identical extraction values, a result that could be attributed to the presence of the monopositive charge. Instead, a slightly lower myocardial extraction was found for neutral complexes. Comparison of the observed kinetic behavior of neutral complexes in the isolated perfused rat heart model with that of the myocardial metabolic tracer [123I]IPPA revealed that the introduction of the metallic moiety partially hampers recognition of the labeled fatty acids by cardiac enzymes, and consequently, their behavior did not completely reflect myocardial metabolism.

INTRODUCTION The search for a Tc-99m labeled fatty acid derivative has been actively pursued in the past years in the attempt to find a suitable alternative to radiopharmaceuticals labeled with positron emitter radionuclides or iodine radioisotopes for monitoring heart metabolism (1, 2). A Tc-99m imaging agent sensitive to alterations in fatty acid metabolism in myocardium would be highly desirable because of the ideal nuclear properties of this radionuclide and its easy availability through the 99Mo/99mTc generator system. A number of different Tc-99m complexes incorporating fatty acid chains of different lengths have been reported, but none of them were shown to have potential use in clinical practice (3, 9). An ideal Tc-99m agent for heart imaging should possess some key features that are essential for visualizing the cardiac region. For instance, one of the most critical requirements is a rapid and quantitative washout of activity from nontarget organs surrounding the heart, such as blood, lungs, and liver, to avoid interference with the activity localized into the myocardium. In particular, liver uptake always constitutes a rather difficult problem, which may become even more important when the radioactive agent is employed for targeting some subtle metabolic process. Indeed, a common drawback plaguing the * [email protected]. † University of Ferrara. ‡ Technical University Dresden. § Institute of Radiopharmacy. 4 ICIS-CNR.

biodistribution behavior of Tc-99m tracers for imaging fatty acid metabolism proposed so far is attributable to high and persistent liver uptake. A few years ago, we discovered that a large class of Tc-99m complexes, having the general structure sketched in Figure 1b, are easily extracted by the myocardium, but rapidly and quantitatively washed out from nontarget organs, in particular, from the liver (10, 12). These complexes can be thought to be composed essentially by two parts: (a) a metalcontaining fragment, {[99mTc(N)(PNP)]2+}, formed by the technetium atom coordinated to a terminal nitride group (N3-) and a bidentate diphosphane ligand (PNP), and (b) a bidentate ligand having a suitable set, (X,Y), of coordinating donor atoms. In this representation, the metallic portion [99mTc(N)(PNP)]2+ can be viewed as a strong electrophilic molecular fragment capable of selectively interacting with bidentate electron-rich ligands (XY) through the overlap of its empty π-antibonding orbitals with the filled π-orbitals of the donor ligand (13, 14). The favorable properties for cardiac imaging of this new category of Tc-99m complexes were particularly apparent in the study of the biological behavior of the complex [99mTc(N)(PNP5)(DBODC)]+ {PNP5 ) bis[(dimethoxypropylphosphanyl)ethyl]ethoxyethylamine; DBODC ) N,N′-bis(ethoxyethyl)dithiocarbamato}. It was found that this complex rapidly accumulates in myocardium of various animal species and in humans, and washout from nontarget organs was extremely fast and quantitative. In particular, heart/liver ratio was 1 order of magnitude higher than those observed for commercial Tc-99m heart imaging agents (15, 16).

10.1021/bc7002378 CCC: $40.75  2008 American Chemical Society Published on Web 01/19/2008

Labeling of Fatty Acid Ligands

Bioconjugate Chem., Vol. 19, No. 2, 2008 451 Table 1. Cromatographics Results (Rt, min)

TLC (Rf) complex syn, anti-[M(N)(PNP3)(5)] syn, anti-[M(N)(PNP3)(9)] [M(N)(PNP3)(10)]+ [M(N)(PNP3)(11)]+ [M(N)(PNP5)(10)]+ [M(N)(PNP5)(11)]+

99m

99m

Re

Tc a

d

0.33, 0.43 14.50, 15.23 0.25, 0.360a 15.28, 15.85d 0.42b 0.31b 0.37c 0.27c 14.80d

Tc

15.37, 16.02d 16.13, 16.68d 14.60d 15.18d 14.97d 15.34d

a C18, CH3CN/0.1-M Et3N (pH )3 adjusted with 1 M H3PO4) (70:30). b SiO2, ETOH/CH3Cl/toluene/0.5 M NH4OAc (l.5:3:3:0.5). c C18, saline/CH3OH/THF/HOAc (glacial) (2:8:1:1). d Reversed-phase C18 column, gradient: A ) MeOH, B ) phosphate buffer (0.01 M, pH 7.4), t ) 0 min, A ) 50%, t ) 15 min, A ) 100%.

Figure 1. (a) The ligands utilized in this study and (b) general structure of asymmetrical nitride Tc(V) complexes.

Based on these findings, we speculated that, presumably, the [99mTc(N)(PNP)]2+ basic molecular motif characterizing this class of complexes might be a key factor favoring the fast liver elimination. This consideration prompted us to exploit this chemistry for designing new Tc-99m radiopharmaceuticals for in vivo monitoring of cardiac metabolism, which should incorporate the [99mTc(N)(PNP)]2+ building block together with a suitable bidentate ligand bearing a lateral fatty acid chain. In our strategy, the combination of these two molecular pieces would allow the development of myocardial tracers targeting fatty-acid-heart metabolism while concomitantly showing a fast liver clearance. This paper describes the synthesis of new bifunctional ligands (L) derived from fatty acids having different lengths in their carbon backbone and their reactions with the metal synthon [99mTc(N)(PNP)]2+ to afford new asymmetrical complexes of the type [99mTc(N)(PNP)(L)]0/+. The chemical characterization and full biological evaluation in animals of these new Tc-99m agents will be also reported.

EXPERIMENTAL SECTION Materials. Technetium-99m was eluted from a 99Mo/99mTc generator provided by Amersham Sorin (Saluggia, Italy). The diphosphane ligands bis[(dimethoxypropylphosphanyl)ethyl]methoxyethylamine (PNP3), bis[(dimethoxypropylphosphanyl)ethyl]ethoxyethylamine (PNP5), and bis[(diphenylphosphanyl)ethyl]methylamine (PNP7) (Figure 1a) were purchased from Argus Chemicals (Florence, Italy) or prepared according to literature procedures (17). All commercially available compounds and solvents were reagent grade and were used without further purification. Labeling of p-iodophenylpentadecanoic acid (EMKA-CHEMIE, Markgröningen, Germany) with 123I was performed

through Cu(I)-assisted nucleophilic exchange to give the final [123I]IPPA radiocompound as described previously (18). Instruments. FT IR spectra were recorded on a Nicolet 510P Fourier transform spectrometer, in the range 4000-400 cm-1 and in KBr mixture using a Spectra-Tec diffuse reflectance collector accessory. Elemental analyses (C, H, N, S) were performed on a Carlo Erba 1106 elemental analyzer. 1H, 13C, and 31P NMR spectra were collected on a Brucker 400 instrument using SiMe4 as internal reference (1H, 13C) and 85% aqueous H3PO4 as external reference (31P). Chromatography. The radiochemical purity (RCP) of the final Tc-99m compounds were determined by thin-layer chromatography (TLC) on SiO2 plates and reversed-phase C18 plates (Merck). The activity on the plates were located and measured by a Cyclone instrument (Packard). High-performance liquid chromatography (HPLC) was performed on a Beckman System Gold instrument equipped with a radioisotope detector (model 170). HPLC analysis was carried out using a reversed-phase ODS precolumn (45 × 4.6 mm) and a reversed-phase ODS column (250 × 16 mm) (Beckman) at flow rate of 1 mL/min. Chromatographic results are reported in Table 1. Synthesis of ω-[(2,3-Dimercaptopropanoyl)amino] Fatty Acid Ligands. In the following the symbol, H2n (where n ) 5, 9) stands for the protonated dimercapto fatty acid ligand, while n represents the resulting deprotonated dianionic ligand. When applied, the standard workup procedure for product isolation involved quenching of the reaction mixture in aqueous solution followed by extraction thoroughly with chloroform and washing of the extracts with saturated saline. The combined extracts were then dried over MgSO4, filtered, and the solvent removed under reduced pressure. Preparation of 11-{[2,3-Bis(acetylthio)propanoyl]amino}undecanoic Acid (H25). The synthesis of this ligand was conducted according to the reaction scheme illustrated in Figure 2. Methyl-11-[(2,3-dibromopropanoyl)amino]undecanoate (3). Dibromopropionic acid (1) (1.00 g, 4.31 mmol) and DCC (890 mg, 4.31 mmol) were dissolved in 80 mL of iced chloroform (0 °C). Another solution, prepared separately by dissolving the ω-amino fatty acid ester hydrochloride 2 (1.10 g, 4.37 mmol) and ethyldiisopropylamine (738 µL, 4.31 mmol) in 20 mL of chloroform, was added dropwise to the first solution under stirring. The reaction mixture was stirred overnight without further cooling and then weakly acidified (pH ∼ 3.5) with a few drops 2 M hydrobromic acid. Finally, it was filtered off and evaporated under reduced pressure to remove the solvent. After completing the standard workup procedure for product isolation, purification was carried out by column chromatography (silicagel, CHCl3/Et2O/HOAc (150/10/0.15)] yielding 1.41 g (76%) of a colorless solid (3). 1 H NMR (δ ppm, CDCl3). 6.28 (br, 1H, NH), 4.46 (dd, 3J1 ) 4.5 Hz, 3J2 ) 8.3 Hz, 1H, CHBr), 3.97 (dd, 3J1 ) 8.4 Hz,

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

Cazzola et al.

Figure 2. Reaction diagram for the synthesis of the ligand H25.

J2 ) 10.2 Hz, 1H, ½CH2Br), 3.82 (dd, 3J1 ) 4.5 Hz, 2J2 ) 10.3 Hz, 1H, ½CH2Br2), 3.66 (s, 3H, OCH3), 3.31 (q, 3J ) 6.7 Hz, 2H, CH2N), 2.29 (t, 3J ) 7.5 Hz, 2H, CH2COO), 1.65–1.49 (m, 4H, 2CH2), 1.38–1.22 (m, 12H, 6CH2). 13 C NMR (δ ppm, CDCl3) 174.4, COO, 165.9, CONH, 51.4, OCH3, 46.4, CHBr, 40.3, CH2N, 34.1, CH2COO, 32.3, CH2Br, 29.3, 29.2(2C), 29.1(2C), 29.0, 26.7, 24.9. IR (cm-1, KBr). 3289, 3076 ν(NH), 1727 ν(CdO, ester), 1654 ν(CdO, amide), 1558 δ(NH), 1178 ν(C-O), 594 ν(C-Br). Elemental analysis for C15H27Br2NO3 (mm ) 429.19). Calcd: C, 41.9, H, 6.34, N, 3.26. Found: C, 42.00, H, 6.08, N, 3.27. Methyl-11-{[2,3-bis(acetylthio)propanoyl]amino}undecanoate (4). Triethylamine (785 µL, 5.63 mmol) was added to a solution of thioacetic acid (440 mg, 5.78 mmol) in acetone (25 mL) at –78 °C. To the resulting mixture, a solution of the dibromide derivative 3 (1.20 g, 2.80 mmol) in acetone (15 mL) was further added dropwise. After slowly warming up to room temperature, the mixture was stirred for 16 h. Then, the solvent was removed under reduced pressure, and the crude product was obtained by applying the standard workup procedure. Purification, carried out by column chromatography [silicagel, mobile phase, n-hexane/Et2O/HOAc (100:100:0.2)] yielded 738 mg of the protected fatty acid ligand 4 as a colorless solid (yield, 63%). 1 H NMR (δ ppm, CDCl3). 6.14 (br, 1H, NH), 4.08 (dd, 3J1 ) 6.3 Hz, 3J2 ) 8.7 Hz, 1H, CHS), 3.65 (s, 3H, OCH3), 3.36 (dd, 3J1 ) 8.7 Hz, 2J2 ) 13.7 Hz, 1H, ½CH2S), 3.23 (dd, 3J1 ) 6.1 Hz, 2J2 ) 13.7 Hz, 1H, ½CH2S), 3.25–3.16 (m, 2H, CH2N), 2.38 (s, 3H, CCH3), 2.33 (s, 3H, CCH3), 2.29 (t, 3J ) 7.6 Hz, 2H, CH2COO), 1.65–1.55 (m, 2H, CH2), 1.50–1.40 (m, 2H, CH2), 1.33–1.21 (m, 12H, 6CH2). 13 C NMR (δ ppm, CDCl3). 195.7 (COS), 195.2 (COS), 174.3 (COO), 169.3 (CONH), 51.4 (OCH3), 45.5 (CHS), 39.7 (CH2N), 34.1 (CH2COO), 30.4 (CCH3), 30.3 (CCH3), 29.4, 29.3(2C), 29.2, 29.1(2C), 29.0, 26.7, 24.9. IR (cm-1, KBr). 3278, 3094 ν(NH), 1739 ν(CdO, ester), 1692 ν(CdO, thioester), 1647 ν(CdO, amide), 1564 δ(NH), 1176 ν(C-O). Elemental analysis for C19H33NO5S2 (mm, 419.61). Calcd: C, 54.39, H, 7.93, N, 3.34, S, 15.28. Found: C, 54.35, H, 8.02, N, 3.31, S, 15.28. 11-[(2,3-Dimercaptopropanoyl)amino]undecanoic Acid (H25). Hydrochloric acid (1.5 M, 3 mL) was added to a solution of 4 (500 mg, 1.19 mmol) in glacial acetic acid (6 mL), and the resulting mixture was heated for 45 min at 100 °C. After cooling to room temperature, the reaction mixture was diluted with 75 mL of water and the crude product was extracted according to the standard workup procedure. Purification was carried out by column chromatography [silicagel, mobile phase, CHCl3/EtOAc/HOAc (80:20:1)] yielding the fatty acid ligand H25 as a colorless solid (298 mg, yield, 78%). 1 H NMR (δ ppm, CDCl3). 6.48 (br, 1H, NH), 3.44 (ddd, 3J1 ) 4.7 Hz, 3J2 ) 6.6 Hz, 3J3 ) 10.6 Hz, 1H, CHSH), 3.28 (dq, 2

J1 ) 1.4 Hz, 3J2 ) 7.2 Hz, 2H, CH2N), 3.15 (ddd, 3J1 ) 4.8 Hz, 3J2 ) 8.6 Hz, 2J3 ) 13.7 Hz, 1H, ½CH2SH), 2.95 (ddd, 3J1 ) 6.8 Hz, 3J2 ) 9.6 Hz, 2J3 ) 13.8 Hz, 1H, ½CH2SH), 2.42 (d, 3J ) 10.6 Hz, 1H, CHSH), 2.34 (t, 3J ) 7.5 Hz, 2H, CH2COO), 1.68 (dd, 3J1 ) 8.5 Hz, 3J2 ) 9.6 Hz, 1H, CH2SH), 1.67–1.58 (m, 2H, CH2), 1.57–1.48 (m, 2H, CH2), 1.38–1.22 (m, 12H, 6CH2). 13 C NMR (δ ppm, CDCl3). 179.2 (COO), 170.4 (CONH), 46.0 (CHS), 40.3 (CH2N), 34.1 (CH2COO), 30.7 (CH2S), 29.6, 29.5, 29.4, 29.3 (2C), 29.1, 27.0, 24.9. IR (cm-1, KBr). 3300, 3092 ν(NH), 2547 ν(SH), 1713 ν(CdO, acid), 1641 ν(CdO, amide), 1558 δ(NH). Elemental analysis for C14H27NO3S2 (mm, 321.51). Calcd: C, 52.30, H, 8.46, N, 4.36, S, 19.95. Found: C, 52.33, H, 8.41, N, 4.31, S, 19.91. ESI MS (m/z). 344 (100%) (C14H27NO3S2Na+). Preparation of 12-{[2,3-Bis(acetylthio)propanoyl]amino}dodecanoic Acid (H29). The synthesis of this ligand was conducted according to the reaction scheme illustrated in Figure 3. 2,3-Bis(acetylthio)propionic Acid (7). Triethylamine (3.65 mL, 26.2 mmol) and dibrompropionic acid (1) (3.00 g, 12.9 mmol), dissolved in acetone (10 mL), were added to a solution of thioacetic acid (2.05 g, 26.9 mmol) in acetone (20 mL), at –78 °C. After warming up slowly to room temperature, the mixture was stirred for 16 h, and then filtered off to eliminate precipitated ammonium salt. Finally, the solvent was removed under reduced pressure, and the resulting crude product was purified by column chromatography [silicagel, mobile phase, n-hexane/Et2O/HOAc (60:40:1)] to afford a highly viscous, light yellow oil (6). Yield, 52% (1.49 g). 1 H NMR (δ ppm, CDCl3). 4.34 (t, 3J ) 7.0 Hz, 1H, CHS), 3.42 (dd, 3J1 ) 7.8 Hz, 2J2 ) 13.9 Hz, 1H, ½CH2S), 3.28 (dd, 3 J1 ) 6.3 Hz, 2J2 ) 14.1 Hz, 1H, ½CH2S), 2.40 (s, 3H, CH3), 2.35 (s, 3H, CH3). 13 C NMR (δ, CDCl3). 194.5 (COS), 192.8 (COS), 175.1 (COO), 45.1 (CHS), 30.4 (CH3), 30.1 (CH3), 29.9 (CH2S). Methyl-12-{[2,3-bis(acetylthio)propanoyl]amino}dodecanoate (8). To a solution of 6 (795 mg, 3.58 mmol) in dry tetrahydrofuran (25 mL), 1-hydroxybenzotriazole (535 mg, 3.96 mmol) was first added at –10 °C, then followed by a solution of DCC (775 mg, 3.76 mmol) in tetrahydrofuran (5 mL). The resulting mixture was stirred for 30 min at -10 °C. A solution composed of the ω-amino-fatty-acid-ester hydrochloride 7 (1.00 g, 3.76 mmol) and ethyl diisopropylamine (625 µL, 3.65 mmol) dissolved in chloroform (5 mL) was added slowly to the reaction mixture. After warming up slowly to room temperature, the mixture was further stirred for 16 h, and then the solvent was removed under reduced pressure. Purification of the crude product, carried out by column chromatography [silicagel, mobile phase, n-hexane/Et2O/HOAc (100:100:0.2)], yielded 1.18 g of the protected fatty acid 8 as a colorless solid. (Yield, 76%). 3

Labeling of Fatty Acid Ligands

Bioconjugate Chem., Vol. 19, No. 2, 2008 453

Figure 3. Reaction diagram for the synthesis of the ligand H29. 1

H NMR (δ ppm, CDCl3). 6.14 (br, 1H, NH), 4.08 (dd, 3J1 ) 6.1 Hz, 3J2 ) 8.7 Hz, 1H, CHS), 3.65 (s, 3H, OCH3), 3.36 (dd, 3J1 ) 8.7 Hz, 2J2 ) 13.7 Hz, 1H, ½CH2S), 3.23 (dd, 3J1 ) 6.2 Hz, 2J2 ) 13.6 Hz, 1H, ½CH2S), 3.25–3.16 (m, 2H, CH2N), 2.38 (s, 3H, CCH3), 2.33 (s, 3H, CCH3), 2.29 (t, 3J ) 7.6 Hz, 2H, CH2COO), 1.66–1.55 (m, 2H, CH2), 1.50–1.40 (m, 2H, CH2), 1.34–1.20 (m, 14H, 7CH2). 13 C NMR (δ ppm, CDCl3). 195.7 (COS), 195.2 (COS), 174.3 (COO), 169.3 (CONH), 51.4 (OCH3), 45.5 (CHS), 39.8 (CH2N), 34.1 (CH2COO), 30.4 (CCH3), 30.3 (CCH3), 29.4 (2C), 29.3 (2C), 29.2, 29.1 (2C), 29.0, 26.7, 24.9. IR (cm-1, KBr). 3293, 3091 ν(NH), 1736 ν(CdO, ester), 1715, 1703 ν(CdO, thioester), 1641 ν(CdO, amid), 1561 δ(NH), 1173 ν(C-O) cm-1. Elemental analysis for C20H35NO5S2 (mm, 433.63). Calcd: C, 55.40, H, 8.14, N, 3.23, S, 14.79. Found: C, 55.22, H, 8.09, N, 3.21, S, 15.16. 12-{[2,3-Bis(acetylthio)propanoyl]amino}dodecanoic Acid (H29). Conversion of 8 (800 mg, 1.84 mmol), carried out as described above for the ligand H25 afforded H29 as a colorless solid. Yield, 80% (496 mg). 1 H NMR (δ ppm, CDCl3). 6.50 (br, 1H, NH), 3.45 (ddd, 3J1 ) 4.7 Hz, 3J2 ) 6.7 Hz, 3J3 ) 10.6 Hz, 1H, CHSH), 3.28 (dq, 3 J1 ) 1.1 Hz, 3J2 ) 7.1 Hz, 2H, CH2N), 3.15 (ddd, 3J1 ) 4.8 Hz, 3J2 ) 8.4 Hz, 2J3 ) 13.8 Hz, 1H, ½CH2SH), 2.95 (ddd, 3J1 ) 6.9 Hz, 3J2 ) 9.6 Hz, 2J3 ) 13.7 Hz, 1H, ½CH2SH), 2.42 (d, 3J ) 10.6 Hz, 1H, CHSH), 2.34 (t, 3J ) 7.5 Hz, 2H, CH2COO), 1.68 (dd, 3J1 ) 8.4 Hz, 3J2 ) 9.5 Hz, 1H, CH2SH), 1.66–1.58 (m, 2H, CH2), 1.57–1.48 (m, 2H, CH2), 1.38–1.22 (m, 14H, 7CH2). 13 C NMR (δ ppm, CDCl3). 179.2 (COO), 170.2 (CONH), 45.7 (CHS), 40.1 (CH2N), 33.9 (CH2COO), 30.4 (CH2S), 29.4, 29.3 (3C), 29.1 (2C), 28.9, 26.8, 24.6. IR (cm-1, KBr). 3304, 3091 ν(NH), 2552 ν(SH), 1699 ν(CdO, acid), 1643 ν(CdO, amid), 1559 δ(NH). Elemental analysis for C15H29NO3S2 (mm, 335.53).Calcd: C, 53.70, H, 8.71, N, 4.17, S, 19.11. Found:C, 53.90, H, 8.47, N, 4.03, S, 18.83. ESI MS (m/z). 359 (100%) (C15H31NO3S2Na+). Synthesis of ω-[(Dithiocarbamato)amino] Fatty Acid Ligands. In the following, the symbol Xn• (where X ) Ba, 2K and n• ) 10•, 11•) stands for the salt of the doubly deprotonated dithiocarbamato fatty acid ligand as represented by n•. The monoanionic form of n•, involving only a deprotonated sulfur atom at the dithiocarbamato group, is represented by the symbol n (10, 11). The preparation of dithiocarbamato derivatives of fatty acid ligands was performed according to reactions reported in Figure 4.

Figure 4. Reaction diagram for the synthesis of the ligands Ba10• and K211•.

Barium 11-(carbodithioatoamino)undecanoate (Ba10•). Barium hydroxide (1.85 g, 5.88 mmol) was dissolved in 100 mL of water and, to the resulting solution, 11-aminoundecanoic acid (1.00 g, 4.9 mmol) and acetone (50 mL) were added under stirring. The reaction mixture was cooled at 0 °C, and carbon disulfide (0.50 mL, 6.37 mmol) was then added dropwise under stirring. The reaction was kept at 0 °C and stirred overnight. Precipitation of a colorless solid was observed, which was filtered off. Additional precipitate was collected after evaporation of the solvent under reduced pressure. The combined solids were then washed with CH2Cl2, ethanol, and diethyl ether to afford 1.1 g of Ba10• as a colorless solid. (Yield, 80%). 1 H NMR (δ ppm, D2O). 3.76 (t, 3J ) 7.1 Hz, 2H, CH2COO), 3.68 (br, 1H, NH), 2.45 (t, 3J ) 7.4 Hz, CH2N), 1.81 (br, 2H, CH2CH2COO), 1.80 (br, 2H, CH2CH2NH), 1.57 (br, 6H, CH2CH2CH2). 13 C NMR (δ ppm, D2O). 209.7 (CS2), 184.6 (COO), 48.4 (CH2N), 40.7 (CH2COO), 37.8, 31.9, 28.9, 28.8, 28.8, 28.7, 28.7, 27.8 (CH2). IR (cm-1, KBr). 3280, 3070 ν(NH), 1706 ν(CdO, acid), 1501 δ(NH), 1471 ν(CNS, dithiocarbamate). Elemental analysis for C12H21NO2S2Ba (mm, 412.76). Found: C, 51.95, H, 8.36, N, 5.05, S, 23.11. Found: C, 52.04, H, 8.41, N, 5.31, S, 22.91. ESI MS (m/z). 301 (100%) (C12H21NO2S2 Na+). ES- MS (m/z). 277 (100%) (C12H21NO2S2). Dipotassium 12-(carbodithioatoamino)dodecanoate (K211•). Potassium hydroxide (0.65 g, 10.2 mmol) was dissolved in 70 mL of water. To this solution, 12-aminoundecanoic acid (1.00 g, 4.64 mmol) and acetone (30 mL) were added under stirring. The solution was cooled at 0 °C, and carbon disulfide (0.5 mL, 6.37 mmol) was added dropwise under stirring. The resulting yellow solution was stirred overnight at 0 °C and then evaporated under reduced pressure to give a pale yellow solid. After washing with CH2Cl2, ethanol, and diethyl ether, K211• (1.0 g) was collected as a colorless solid. (Yield, 75%). 1 H NMR (δ ppm, D2O): 3.75 (t, 3J ) 7.1 Hz, 2H, CH2COO), 3.66 (br, 1H, NH), 2.41 (t, 3J ) 7.4 Hz, CH2N), 1.85 (br, 2H,

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

Cazzola et al.

Figure 5. Reaction diagram for the synthesis of rhenium complexes with fatty acid ligands.

CH2CH2COO), 1.82 (br, 2H, CH2CH2NH), 1.55 (br, 7H, CH2CH2CH2). 13 C NMR (δ ppm, D2O). 209.7 (CS2), 184.5 (COO), 48.4 (CH2NH), 37.7 (CH2COO), 29.2, 28.8, 28.7, 28.6, 27.8, 27.6, 26.3, 26.2 (CH2). IR (KBr). 3193, 2998 ν(NH), 1702 ν(CdO, acid), 1518 ν(NH), 1470 δ(CN) cm-1. Elemental analysis for C13H23NO2S2K2 (327.64). C, 53.57, H, 8.65, N, 4.81, S, 22.00. Found: C, 53.90, H, 8.47, N, 4.83, S, 21.83. ESI MS (m/z). MS (ES-) 290 (100%) (C13H24NO2S2). Synthesis of Rhenium Complexes with Fatty Acid Ligands. Rhenium complexes with fatty acid ligands were prepared at the macroscopic concentration level (>10-6 mol dm-3) starting from the precursor complex [Re(N)(PNP7)Cl2] and following the reaction scheme described in Figure 5. The precursor complex [Re(N)(PNP7)Cl2] was obtained according to procedures reported previously (13, 17). Re(N)(PNP7)(L) (L ) 5, 9; PNP ) PNP7). A 2-fold molar excess of the appropriate dimercapto fatty acid ligand was added to 75 mg of the precursor complex [Re(N)(PNP7)Cl2] suspended in a mixture of chloroform (10 mL) and ethanol (10 mL) containing triethylamine (55.0 µL, 395.0 µmol). The reaction mixture was heated under reflux and the course of the reaction monitored by TLC [silicagel, CHCl3/EtOAc (1:1)]. Approximately after 2 h, the conversion of the precursor complex was almost complete. Then, the mixture was neutralized by 2 M citric acid, and the solvent removed under reduced pressures. The crude precipitate was washed with CH2Cl2, EtOH, and Et2O. The resulting complexes were further purified through steps, a and b, carried out by column chromatography [(a) silicagel, mobile phase, CHCl3/MeOH (10:1), (b) silicagel, CHCl3/MeOH/ acetone (20:1:1)]. After purification, the resulting product was washed with diethyl ether. [Re(N)(PNP7)(5)]. Yield, 72% (72 mg). Light-yellow solid collected as syn/anti stereoisomeric mixture. 1 H NMR (δ ppm, CDCl3). 7.95–7.82 (m, 8H, CHar), 7.80 (t, 3 J ) 5.6 Hz, 1H, NHsyn), 7.46–7.20 (m, 20H, CHar), 7.07–6.97 (m, 4H, CHar), 6.96–6.86 (m, 8H, CHar), 6.36 (t, 3J ) 5.8 Hz, 1H, NHanti), 3.78 (dd, J1 ) 3.1 Hz, J2 ) 6.0 Hz, 1H, CHSsyn), 3.70 (dd, J1 ) 4.8 Hz, J2 ) 11.8 Hz, 1H, CHSanti), 3.41–3.21 (m, 4H, ½CH2Ssyn, ½CH2Santi, CH2NHsyn), 3.17 (q, 3J ) 6.6 Hz, 2H, CH2NHanti), 3.13–2.52 (m, 17H, 4CH2N, 4CH2P,

½CH2Ssyn), 2.52–2.43 (m, 1H, ½CH2Santi), 2.34 (t, 3J ) 6.9 Hz, 2H, CH2COOanti), 2.32 (t, 3J ) 6.9 Hz, 2H, CH2COOsyn), 2.10 (s, 3H, NCH3anti), 2.01 (s, 3H, NCH3syn), 1.67–1.56 (m, 6H, 2CH2syn, CH2anti), 1.47–1.18 (m, 26H, 6CH2syn, 7CH2anti). 13 C NMR (δ ppm, CDCl3). 177.3 (2C, 2COO), 174.8 (CONHsyn), 174.6 (CONHanti), 135–127 (48Car), 60.6 (d, 3JCP ) 13.0 Hz, CHSanti), 55.5 (d, 3JCP ) 13.0 Hz, CHSsyn), 53.8 (CH2NCH2anti), 53.6 (3C, CH2NCH2,anti, CH2NCH2syn), 46.6 (NCH3anti), 45.9 (NCH3syn), 42.9 (d, 3JCP ) 9.9 Hz, CH2Ssyn), 39.7 (CH2NHsyn), 39.4 (CH2NHanti), 38.9 (d, 3JCP ) 6.9 Hz, CH2Santi), 33.9 (CH2COOsyn), 33.8 (CH2COOanti), 29.3, 29.2, 29.1, 29.0 (4C), 28.8 (4C), 28.7, 26.7, 26.6, 25.5 (d, 1JCP ) 33.6 Hz, CH2Psyn), 25.3 (d, 1JCP ) 32.8 Hz, CH2Panti), 24.9 (d, 1 JCP ) 32.0 Hz, CH2Panti), 24.7 (2C), 24.0 (d, 1JCP ) 32.0 Hz, CH2Psyn). 31 P NMR (δ ppm, CDCl3). 17.0 (d, 2JPP ) 8.5 Hz, syn), 16.5 2 (d, JPP ) 8.6 Hz, anti), 14.7 (d, 2JPP ) 7.3 Hz, syn), 14.5 (d, 2 JPP ) 8.6 Hz, anti). IR (cm-1, KBr). 1723 ν(CdO, acid), 1627 ν(CdO, amide), 1531 δ(NH), 1436 ν(P-Ph), 1100 ν(Re-P), 1051 ν(RetN) cm-1. Elemental analysis for C43H56N3O3P2ReS2 (mm, 975.22). Calcd: C, 52.96, H, 5.79, N, 4.31, S, 6.58. Found: C, 52.70, H, 5.89, N, 4.21, S, 6.49. [Re(N)(PNP7)(9)]. Yield, 68% (67 mg). Light-yellow solid collected as a syn/anti stereoisomeric mixture. 1 H NMR (δ ppm, CDCl3). 7.95–7.81 (m, 8H, CHar), 7.80 (t, 3 J ) 5.8 Hz, 1H, NHsyn), 7.46–7.20 (m, 20H, CHar), 7.08–6.97 (m, 4H, CHar), 6.96–6.86 (m, 8H, CHar), 6.39 (t, 3J ) 5.6 Hz, 1H, NHanti), 3.78 (dd, J1 ) 3.0 Hz, J2 ) 6.0 Hz, 1H, CHSsyn), 3.70 (dd, J1 ) 4.9 Hz, J2 ) 12.2 Hz, 1H, CHSanti), 3.42–3.21 (m, 4H, ½CH2Ssyn, ½CH2Santi, CH2NHsyn), 3.17 (q, 3J ) 6.6 Hz, 2H, CH2NHanti), 3.14–2.53 (m, 17H, 4CH2N, 4CH2P, ½CH2Ssyn), 2.53–2.44 (m, 1H, ½CH2Santi), 2.34 (t, 3J ) 7.4 Hz, 2H, CH2COOanti), 2.32 (t, 3J ) 7.3 Hz, 2H, CH2COOsyn), 2.11 (s, 3H, NCH3,anti), 2.02 (s, 3H, NCH3syn), 1.68–1.56 (m, 6H, 2CH2,syn, CH2anti), 1.47–1.20 (m, 30H, 7CH2syn, 8CH2anti). 13 C NMR (δ ppm, CDCl3). 177.1 (2C, 2COO), 174.9 (CONHsyn), 174.6 (CONHanti), 135–127 (48Car), 60.5 (d, 3JCP ) 13.7 Hz, CHSanti), 55.5 (d, 3JCP ) 12.2 Hz, CHSsyn), 53.8 (CH2NCH2anti), 53.6 (3C; CH2NCH2anti, CH2NCH2syn), 46.6 (NCH3anti), 45.9 (NCH3syn), 42.9 (d, 3JCP ) 9.9 Hz, CH2Ssyn), 39.8 (CH2NHsyn), 39.5 (CH2NHanti), 38.9 (d, 3JCP ) 6.1 Hz, CH2Santi), 33.8 (CH2COOsyn), 33.7 (CH2COOanti), 29.4, 29.2,

Labeling of Fatty Acid Ligands

29.1 (2C), 29.0 (2C), 28.9 (3C), 28.8 (2C), 28.7 (3C), 26.7, 26.5, 25.5 (d, 1JCP ) 32.8 Hz, CH2Psyn), 25.4 (d, 1JCP ) 33.6 Hz, CH2Panti), 24.9 (d, 1JCP ) 32.0 Hz, CH2Panti), 24.6 (2C), 24.0 (d, 1JCP ) 32.0 Hz, CH2Psyn). 31 P NMR (δ ppm, CDCl3). 17.0 (d, 2JPP ) 8.5 Hz, syn), 16.5 (d, 2JPP ) 8.6 Hz, anti), 14.7 (d, 2JPP ) 8.6 Hz, syn), 14.5 (d, 2 JPP ) 8.5 Hz, anti). IR (cm-1, KBr). 1724 ν(CdO, acid), 1630 ν(CdO, amide), 1529 δ(NH), 1435 ν(P-Ph), 1099 ν(Re-P), 1051, ν(RetN). Elemental analysis for C44H58N3O3P2ReS2 (mm, 989.25). Calcd: C, 53.42, H, 5.91, N, 4.25, S, 6.48. Found: C, 53.00, H, 6.01, N, 4.1, S, 6.33. [Re(N)(PNP7)(11)]+. 23 mg (70.2 mmol) of the dithiocarbamate fatty acid ligand K211• were added to a chloroform (10 mL)/ethanol (10 mL) mixture containing 150.0 mg (195.7 µmol) of [Re(N)(PNP7)Cl2]. This mixture was refluxed for 1 h and then neutralized by addition of 2 M citric acid. After removal of the solvent under reduced pressure, the crude product was isolated by applying the standard workup procedure. The resulting light yellow solid was dissolved in acetone followed by addition of an ethanolic solution of [BF4]Na (2.0 mg in 10 mL). The final product was recovered as BF4 salt by slow evaporation of the solvent. Yield, 77% (80 mg). 1 H NMR (δ ppm, CDCl3). 7.95–7.80 (m, 8H, CHar), 7.46–7.2 0 (m, 20H, CHar), 7.06–6.98 (m, 4H, CHar), 6.96–6.86 (m, 8H, CHar), 4.21 (br, 1H, NH), 3.73 (t, 3J ) 7.1 Hz, 2H, CH2COO), 2.88 (t, 3J ) 7.4 Hz, CH2N), 2.08 (s, 3H, NCH3), 2.01 (br, 2H, CH2CH2NH), 1.85 (br, 2H, CH2CH2COO), 1.61 (br, 7H, CH2CH2CH2). 13 C NMR (δ ppm, CDCl3). 25.5 (d, 1JCP ) 32.4 Hz, CH2P), 24.9 (d, 1JCP ) 32.0 Hz, CH2P), 24.7 (2C), 24.0 (d, 1JCP ) 32.0 Hz, CH2P). 13 C NMR (δ ppm, D2O). 317.7 (CS2), 184.5 (COO), 58.7 (CH2NH), 38.3 (CH2COO), 30.2, 29.6, 29.0, 28.8, 27.1, 26.5, 26.0, 25.5 (CH2). 31 P NMR (δ ppm, CDCl3). 13.22 (d, 2JPP ) 8.7 Hz), 12.5 (d, 2 JPP ) 8.9 Hz). IR (cm-1, KBr). 1725 ν(CdO, acid), 1632 ν(CdO, amide), 1524 δ(NH), 1099 ν(Re-P), 1051, ν(RetN) cm-1. Elemental analysis for C42H55N3O2P2ReS2BF4 (mm, 1033.01). Calcd: C, 48.83, H, 5.37, N, 4.07, S, 6.21. Found: C, 48.22, H, 5.30, N, 3.98, S, 6.03. Preparation of 99mTc-Complexes. A three-step procedure was employed for the preparation of 99mTc-complexes. In all preparations, the first two steps were carried out as follows. Freshly generator eluted [99mTcO4]Na (0.250 mL, 50.0 MBq to 3.0 GBq) was added to a vial containing 5.0 mg of succinic dihydrazide (SDH) and 0.1 mg of SnCl2 suspended in a mixture of 0.1 mL of saline and 1.0 mL of ethanol. The reaction solution was kept at room temperature for 30 min. Then, 1.0 mg of the appropriate diphosphane ligand (PNP ) PNP3, PNP5), dissolved in 0.2 mL of ethanol was further added to the reaction vial, which was left at room temperature for an additional 30 min to afford the precursor metal fragment [99mTc(N)(PNP)]2+. The third step was conducted according to the procedure detailed below for each category of 99mTc-nitride complexes. [99mTc(N)(PNP)(L)] (L ) 5, 9, PNP ) PNP3). The intermediate precursor [99mTc(N)(PNP)]2+ was almost immediately converted into the final neutral complex [99mTc(N)(PNP)(L)] by adding an ethanolic solution of the appropriate dithiol ligand (H25, H29) (1.0 mg in 0.2 mL) to the reaction vial heated at 50 °C. Chromatographic characterization revealed the formation of an isomeric pair (a and b) of the same complex. Total yield determined by HPLC chromatography was approximately 95% (a, 45%; b, 50%). A representative HPLC chromatogram for this class of complexes is reported in Figure 6a.

Bioconjugate Chem., Vol. 19, No. 2, 2008 455

[99mTc(N)(PNP)(L′)]+ (L′ ) 10, 11, PNP ) PNP3, PNP5). Conversion of the precursor [99mTc(N)(PNP)]2+ into the final monocationic complex [99mTc(N)(PNP)(L′)]+ was similarly obtained by adding an ethanolic solution of the appropriate dithiocarbamate ligand (Ba10•, K211•) (1.0 mg in 0.2 mL) to the reaction vial heated at 60 °C. Yield, 96% (determined by HPLC). A representative HPLC chromatogram for this class of complexes is reported in Figure 6b. Purification. To avoid interference of excess of reagents and free ligands on the biological evaluation of the new 99mTccomplexes, the following purification procedures were applied. [99mTc(N)(PNP)(L)] Complexes. The reaction mixture was diluted with 8.0 mL of water and loaded onto a C18 SepPak cartridge (Waters) previously activated with ethanol (95%, 5.0 mL) and water (5.0 mL). The cartridge was rinsed with water (20.0 mL) and ethanol (60%, 3.0 mL). Finally, the desired 99m Tc-complex was recovered by repeated elution with ethanol (0.4 mL followed by 3 × 1.0 mL). The first fraction was not utilized for biological studies. The remaining fractions were combined, and the total radiochemical purity (RCP) (measured by HPLC) was approximately 100% (a, 48%; b, 52%). [99mTc(N)(PNP)(L′)]+ Complexes. The reaction mixture was diluted with 8.0 mL of water and loaded onto a C18 SepPak cartridge (Waters) previously activated with ethanol (95%, 5.0 mL) and water (5.0 mL). The cartridge was rinsed with water (20 mL) and an ethanolic solution containing [N(n-Bu)4]Br (10% w/w, 1.0 mL). The final 99mTc-complex was eluted by repeated elution with the same [N(n-Bu)4]Br solution (1 × 0.4 mL followed by 3 × 1.0 mL). The first fraction was not utilized for biological studies. The remaining fractions were combined, and RCP (measured by HPLC) was approximately 100%. Serum Stability. The selected 99mTc-complex (100 µL), purified as described above, were added to a propylene test tube (5.0 mL) containing rat serum (900 µL) or, alternatively, saline (900 µL). The resulting mixture was incubated at 37 °C for 2 h. RCP changes in time were monitored by TLC at 15, 30, 60, and 120 min. No significant changes of RCP were observed in saline. Conversely, a slow decrease of RCP occurred in rat serum (RCP, 82% at 15 min and 45% at 120 min after incubation). In Vitro Reaction with Glutathione (GSH) and Cysteine. An aliquot of a stock aqueous solution of GSH (50.0 µL, 10.0 mM) was added to a propylene test tube (5.0 mL) containing a phosphate buffer (250.0 µL, 0.2 M, pH ) 7.4), water (100.0 µL), and the appropriate 99mTc-complex (100.0 µL) purified as described above. The mixture was incubated at 37 °C for 2 h. For the blank, an equal volume of water was added in place of GSH solution. Aliquots of the resulting mixtures were withdrawn at 15, 30, 60, and 120 min after incubation, and analyzed by TLC chromatography. The complexes were found to be inert toward substitution by GSH. An identical procedure was employed with aqueous solutions of cysteine hydrochloride (10.0 mM and 1.0 mM). 99mTccomplexes were found to be stable toward transchelation by cysteine. Biological Evaluation. Animal experiments were performed according to the animal welfare regulations of the Italian and German local authorities. Biodistribution Studies. Female Sprague–Dawley rats, weighing 200-250 g, were housed for 1 week under a 12 h light/12 h dark cycle with free access to food and water for subsequent biodistribution studies. Animals were fasted before experiments for 12 h and then anesthetized with an intramuscular injection of a mixture of ketamine (80 mg kg-1) and xilazine (19 mg kg-1). After purification, the selected 99mTc-complex was diluted with a physiological solution containing 5% human serum albumin (HSA). The final solution was less than 5% ethanol

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

Cazzola et al.

Figure 6. Representative HPLC chromatograms of complexes (a) [99mTc(N)(PNP3)(9)] and (b) [99mTc(N)(PNP5)(10)]+.

content. Radioactivity (100 µL, 300-370 kBq) was administered through the jugular vein. The animals (n ) 3) were sacrificed by dislocation at 5, 30, and 60 min time intervals postinjection. The blood was withdrawn from the heart through a syringe immediately after sacrifice and counted for the radioactivity. Organs were excised, rinsed with saline, weighed, and radioactivity determined with a NaI well counter. The percent injected dose per gram (%ID g-1) for each organ and blood was calculated and results are reported in Tables 2 and 3. Isolated Perfused Rat Heart Experiments. Female Wistar rats (250-350 g) were obtained from Charles River Laboratories, Sulzfeld, Germany. Anesthesia was performed in two steps as follows. After resting for a short time in a chamber flushed with diethyl ether, deep anesthesia was achieved by intraperitoneal (i.p.) injection of 2–3 mL of urethane (15%). In addition, 1000 U/kg i.p. heparin was injected before opening the peritoneal cavity. Hearts were rapidly removed during continuous cooling and mounted vertically via the aorta on a perfusion cannula (this operation required less than 1 min after removal). Perfusion was immediately started according to the Langendorff technique with a modified KHB containing NaCl (116.0 mmol L-1), KCl (4.6 mmol L-1), MgSO4 (1.2 mmol L-1), KH2PO4 (1.2 mmol L-1),

NaHCO3 (25.0 mmol L-1), CaCl2 (2.5 mmol L-1), glucose (11.0 mmol L-1), and pyruvate (2.0 mmol L-1). The buffer was continuously bubbled with a mixture of 95% O2/5% CO2 (pH, 7.38-7.43), equilibrated with fatty-acid-free bovine serum albumin (BSA 0.1%, 14.9 µmol L-1), and maintained at 37 °C in a water bath. Total coronary flow was measured using an ultrasonic transit-time flow-meter (T206 Transonic Systems Inc., Ithaca, NY) inserted in the arterial perfusion line and kept constant at 10 mL/min. Coronary perfusion pressure (CPP) was measured through a pressure transducer (Gould Statham, USA) connected to the perfusion cannula. Heart rate was calculated manually from a CPP printout. “Arterial” and “venous” perfusate samples collected anaerobically were analyzed for pO2, pCO2, pH, and HCO3– (AVL 990S, AVL Scientific Corporation, Roswell, GA) before each intervention. The experimental procedure was performed as already described (7, 19). In brief, Tc-99m labeled fatty acids were diluted in a ratio of 1:2 with KHB enriched with 6% BSA and left to equilibrate at room temperature for 30 min. After a period of 30 min to achieve hemodynamic stabilization of the beating heart, the selected 99mTc-labeled fatty acid complex was infused proximal to the perfusion cannula in a 1:1000 ratio. Starting with the infusion, the venous effluent perfusate was completely

Labeling of Fatty Acid Ligands

Bioconjugate Chem., Vol. 19, No. 2, 2008 457

Table 2. Biodistribution of [99mTc(N)PNP3)(L)]0/+ Complexes in Normal Ratsa

Table 3. Biodistribution of [99mTc(N)PNP5)(L)]0/+ Complexes in Normal Ratsa

time after injection (min) organ

5

30

time after injection (min) 60

organ

5

L)5 blood heart lungs liver kidneys intestine

0.30 ( 0.05 0.11 ( 0.03 0.22 ( 0.08 3.10 ( 0.92 0.74 ( 0.45 19.45 ( 2.17

blood heart lungs liver kidneys intestine

0.22 ( 0.08 0.10 ( 0.03 0.13 ( 0.07 9.92 ( 0.25 1.38 ( 0.34 18.77 ( 0.86

blood heart lungs liver kidneys intestine

0.21 ( 0.01 0.37 ( 0.00 0.30 ( 0.04 14.55 ( 0.76 2.07 ( 0.63 8.24 ( 0.50

0.12 ( 0.00 0.03 ( 0.01 0.05 ( 0.01 0.39 ( 0.00 0.17 ( 0.09 14.25 ( 9.53

0.09 ( 0.01 0.02 ( 0.00 0.04 ( 0.01 0.29 ( 0.11 0.08 ( 0.02 21.90 ( 14.99

blood heart lungs liver kidneys intestine

0.34 ( 0.03 0.13 ( 0.02 0.21 ( 0.09 3.26 ( 0.52 0.84 ( 0.25 20.45 ( 2.03

0.06 ( 0.01 0.02 ( 0.00 0.03 ( 0.01 2.38 ( 0.45 0.38 ( 0.23 22.21 ( 10.97

blood heart lungs liver kidneys intestine

0.28 (0.05 0.12 ( 0.04 0.17 ( 0.09 8.72 ( 0.13 1.55 ( 0.22 19.28 ( 0.80

0.09 ( 0.05 0.19 ( 0.01 0.15 ( 0.02 2.90 ( 0.40 0.42 ( 0.05 19.93 ( 0.48

blood heart lungs liver kidneys intestine

0.15 ( 0.01 0.48 ( 0.02 0.34 ( 0.06 6.98 ( 0.76 2.17 ( 0.23 19.88 ( 0.35

L)9 0.08 ( 0.00 0.02 ( 0.00 0.03 ( 0.01 3.49 ( 1.14 0.34 ( 0.07 20.77 ( 3.20

0.23 ( 0.01 1.07 ( 0.02 0.37 ( 0.05 18.28 ( 0.54 3.35 ( 0.13 6.26 ( 0.80

0.04 ( 0.00 0.97 ( 0.04 0.13 ( 0.01 5.77 ( 0.09 0.88 ( 0.06 20.62 ( 3.01

0.07 ( 0.01 0.01 ( 0.00 0.03 ( 0.00 0.23 ( 0.08 0.10 ( 0.02 19.90 ( 12.25

0.06 ( 0.01 0.04 ( 0.00 0.08 ( 0.01 3.69 ( 0.09 0.45( 0.06 23.02 (5.08

0.04 ( 0.01 0.01 ( 0.00 0.02 ( 0.01 2.59 ( 0.25 0.32 ( 0.19 20.96 ( 12.11

0.02 ( 0.00 0.22 ( 0.01 0.16 ( 0.02 3.22 ( 0.09 0.51 ( 0.05 20.78 ( 3.01

0.02 ( 0.05 0.18 ( 0.01 0.10 ( 0.03 2.73 ( 0.17 0.41 ( 0.01 19.07 ( 0.27

L ) 11 0.02 ( 0.00 0.59 ( 0.03 0.10 ( 0.01 1.75 ( 0.16 0.49 ( 0.03 23.70 ( 0.39

a Values are expressed as % injected dose/gram (%ID/g) ( standard deviations (n ) 5).

recovered in successive 15 s collecting periods. Infusion was stopped after 180 s, while continuing the collection of venous effluate for further 60 s to rinse out the residual radiolabeled substance from the coronary vessels, which were not taken up into the extravasal space. Then, perfusion was stopped and hearts instantaneously removed and separated into ventricles and atria. For input calculation (Ain) five 15-s perfusate samples were collected under identical radioactivity infusion and flow settings directly from the perfusion cannula in the absence of a heart. All perfusate and heart samples were measured with a gamma counter (Wallac 1470 Wizard, Perkin-Elmer, Freiburg, Germany). Only experiments in which the calculated amounts of counts per minute were in the expected experimental range ((20%) were included in the final analysis. Measurements were corrected for radionuclide decay. Metabolic Analysis. After intravenous administration of 99mTc fatty acid complexes (50 MBq) into normal rats, the animals were sacrificed at 60 min postinjection, and intestine homogenates were obtained. Analysis was performed by HPLC as detailed before.

RESULTS AND DISCUSSION Neutral and monocationic nitride Tc(V) complexes of the type Tc(N)(PNP)(L)]0/+ (PNP ) PNP3, PNP5; L ) 5, 9, 10, 11) were successfully obtained by reacting the strong electrophilic precursor [99mTc(N)(PNP)]2+ with bidentate chelating ligands bearing lateral fatty acid chains of different lengths (Figure 1a). These bifunctional ligands were prepared by appending a dimercapto or, alternatively, a dithiocarbamato chelating group to one terminus of a fatty acid chain containing eleven or twelve carbon atoms following the reaction schemes illustrated in Figures 2–4. Dimercapto derivatives (5, 9) coordinated to the [99mTc(N)(PNP)]2+ metal fragment as dianionic bidentate 99m

0.10 ( 0.02 0.02 ( 0.00 0.04 ( 0.01 0.41 ( 0.06 0.27 ( 0.07 16.15 ( 6.50

L ) 10

L ) 11 blood heart lungs liver kidneys intestine

60

L)9

L ) 10 0.06 ( 0.00 0.23 ( 0.03 0.14 ( 0.01 7.22 ( 0.08 0.65 ( 0.04 17.42 ( 2.54

30 L)5

blood heart lungs liver kidneys intestine

0.15 (0.02 0.89 ( 0.00 0.30 ( 0.04 5.31 ( 0.40 2.18 ( 0.09 21.00 ( 0.43

0.03 ( 0.02 0.53 ( 0.10 0.28 ( 0.10 1.03 ( 0.22 0.87 ( 0.10 17.73 ( 3.95

0.02 ( 0.00 0.40 ( 0.00 0.19 ( 0.03 0.30 ( 0.07 0.36 ( 0.06 17.95 ( 8.08

a Values are expressed as % injected dose/gram (%ID/g) ( standard deviations (n ) 5).

ligands, thus affording neutral complexes. Instead, dithiocarbamato derivatives (10, 11) bound to the same fragment as monoanionic bidentate ligand, and consequently, the resulting complexes were monocationic. Structural characterization of Tc99m complexes was carried out by comparison with the corresponding rhenium complexes. These compounds were synthesized starting from the mixed halogeno-phosphino complex [Re(N)(PNP)Cl2]. The reaction scheme utilized for the synthesis of rhenium complexes is illustrated in Figure 5. In particular, rhenium complexes with the diphosphane ligand PNP7 were used as model compounds because of their ability to precipitate in a microcrystalline form that allowed an easier characterization by spectroscopic methods. Conversely, the diphosphane ligands PNP3 and PNP5 usually gave rise to oily compounds, which were difficult to isolate and characterize after chromatographic purification. Analytical and spectroscopic data, conjoined with results obtained previously with other similar dimercapto and dithiocarbamato complexes (12, 20), clearly showed that these complexes possess the usual asymmetrical structure with two different bidentate ligands coordinated to the same MtN (M ) Tc, Re) group as illustrated in Figure 5. Specifically, one diphosphane ligand PNP binds to the MtN core through the two π-donor phosphorus atoms, while one dimercapto (or dithiocarbamato) ligand is linked to the same core through the two π-donor sulfur atoms. At both macroscopic and tracer levels, dimercapto complexes were isolated as a syn/ anti isomeric pair. This isomerism arises from the relative orientation of the pendant fatty acid chain with respect to the MtN group, and is related to the sp3 tetrahedral hybridization of the carbon atom of the ethyl bridge connecting the two sulfur atoms and also linked to the fatty acid chain (Figure 5). When the fatty acid chain is tethered to the planar sp2 carbon atom through the amino nitrogen atom of the dithiocarbamato moiety,

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

Cazzola et al.

Figure 8. Ventricular extractions of the dodecanoic fatty acid complexes [99mTc(N)(PNP3)(11)]+ (A) and [99mTc(N)(PNP5)(9)] (B), and of the metabolic tracer [123I]IPPA (C) and flow tracers 99mTc-MIBI (D) and 99m TcN-DBODC (E) as determined in the same isolated perfused rat heart model.

Figure 7. Representative HPLC analysis of intestine homogenates 60 min after intravenous administration of [99mTc(N)(PNP3)(9)] in normal rats.

this isomerism cannot be generated, and only a single species was isolated (Figure 5). Representative chromatographic profiles of these two types of fatty acid asymmetrical nitride Tc-99m complexes are reported in Figure 6. Stability studies demonstrated that the Tc-99m complexes reported here are highly stable in solution and in human plasma, and resistant to transchelation by cysteine and glutathione. Since liver washout was considered as a key issue for developing a suitable metabolic myocardial imaging agent, derivatives containing the ligand PNP7, bearing two phenyl groups on each of the two phosphorus atoms, were not further evaluated because of the poor liver washout showed by this category of complexes. Biological evaluation in animals was conducted only for Tc-99m complexes containing the diphosphane ligands PNP3 and PNP5. After purification, the resulting Tc-99m complexes were dissolved in saline containing 5% HSA. Biodistribution data (Tables 2 and 3) confirmed the fast and quantitative liver washout usually exhibited by this type of complexes. Elimination of liver activity into the intestine was relevant for all complexes as well as the rapid washout from lungs and blood. However, some significant heart uptake was found only for monocationic complexes with dithiocarbamato ligands, with neutral complexes being only slightly retained by the myocardium at 5 min post injection, and then completely washed out from this tissue after 30 min from injection. On the contrary, a slow elimination of myocardial activity was observed for monocationic complexes, and approximately 50% of the activity found in the myocardium at 5 min postinjection was still retained after 60 min from administration (Table 3). In view of the facts that most of the radioactivity was excreted via the hepatobiliary pathway and concentration of radioactivity in urine was too low for a reliable metabolic analysis, we focused on metabolites in intestine homogenates. Metabolic stability of both neutral and monocationic Tc-99m fatty acid complexes described here was studied by HPLC analysis of intestine homogenates collected at 60 min postinjection. Results showed that these complexes undergo a significant metabolic transformation in the liver as illustrated in the representative example reported in Figure 7. The monocationic complexes [99mTc(N)(PNP)(L)]+ (PNP ) PNP3; L ) 10, 11) and the neutral complex [99mTc(N)(PNP5)(9)] were evaluated on the isolated heart model. As a reference, the ventricular extractions of the well-established iodinated metabolic cardiac tracer [123I]IPPA, and of the monocationic myocardial flow tracers 99mTc-MIBI and the recently reported

agent 99mTcN-DBODC (15, 16), were determined. As shown in Figure 8, the monocationic dodecanoic fatty acid derivative [99mTc(N)(PNP3)(11)]+ achieved the highest ventricular extraction (11.33 ( 1.79% ID) among the newly synthesized compounds. For the undecanoic fatty acid complex [99mTc(N)(PNP3)(10)]+ only a slightly lower extraction (10.24% ID) was obtained. As compared to the corresponding PNP3 analogue, the neutral PNP5 derivative [99mTc(N)(PNP5)(9)] clearly showed a lower ventricular extraction (8.52 ( 1.3% ID). The highest ventricular extraction was measured for the tracer [123I]IPPA (15.51 ( 1.10% ID) (7). The myocardial extractions of the new agent 99mTcN-DBODC and of 99mTc-MIBI were 11.20 ( 5.19% ID and 11.34 ( 4.33% ID, respectively and, thus, fall in the same range of the monocationic Tc-99m fatty acid complexes reported here. The so-called “metal fragment approach” may be worth consideration as an extension of the original concept of the “bifunctional approach”, which is commonly used for designing new radiolabeled diagnostic probes, when applied to the labeling of bioactive molecules with metallic radioisotopes. The introduction of this more sophisticated chemical method was mostly prompted by the need to obtain high specific activity labeling in preparations of radiotracers for monitoring in vivo receptor density. The key feature of the metal fragment approach lies in the use of metallic functional groups possessing an intrinsically high electronic stability but that display a selective reactivity toward specific classes of coordinating atoms and ligands through substitution reactions. The metallic group [99mTc(N)(PNP)]2+, composed of a TctN core bound to a bidentate diphosphane ligand, nicely represents this category of molecular moieties that can be conveniently employed for the high specific activity labeling of bioactive molecules. It is highly stable toward redox reactions but interacts selectively only with ligands containing soft π-donor atoms as a result of its strong electrophilic character. This means that a biomolecule having a suitable set of π-donor atoms is capable of coordinating readily to this fragment. In the present study, the above strategy was applied to the design and synthesis of a new class of Tc-99m radiopharmaceuticals for monitoring fatty acid metabolism in myocardial tissue. Since similar studies on Tc-99m-labeled fatty-acid complexes with different chelating systems showed that derivatives incorporating a C12 carbon chain exhibited the highest myocardial uptake (7, 9, 20), C11 and C12 chain lengths were selected here as the most appropriate. A possible explanation for this observation could be found by considering the greater sterical size of the chelating system containing the metal atom as compared to iodophenyl substituents. In fact, it was calculated that 99mTc compounds substituted with C11 or C12 fatty acids ligands have approximately the same size of ω-(p-[123I]iodophenyl fatty acids. Furthermore, it should be noted that prolongation

Labeling of Fatty Acid Ligands

Figure 9. Extraction kinetic curves of the dodecanoic fatty acid neutral complex [99mTc(N)(PNP5)(9)] (•) and [123I]IPPA (O).

of the carbon chain length usually does not prevent the competitive fast and high liver uptake of labeled fatty acids. The choice of the specific diphosphane ligand was dictated by the requirement to achieve a rapid and quantitative liver washout necessary to improve target-to-background ratio. All biological data available so far on Tc-99m complexes with PNP3 and PNP5 (10, 11, 14, 16, 20, 22) clearly indicate that these ligands usually impart to the resulting complexes the property of being rapidly washed out from the liver. Biodistribution studies on [99mTc(N)(PNP)(L)]0/+ (PNP ) PNP3, PNP5) complexes reported here sharply confirmed this prediction. Both neutral and monocationic species exhibit a fast liver elimination associated with a rapid washout from blood and lungs. Unfortunately, these favorable properties were counteracted by the relatively low heart uptake of these complexes, which did not exceed 1.07% of injected activity. In particular, only monocationic derivatives showed some significant myocardial accumulation and retention. Conversely, neutral complexes were rapidly washed out from the heart after a first slight accumulation at 5 min postinjection. Generally, dodecanoic fatty acid derivatives showed higher cardiac uptake than undecanoic analogues. The above findings contrasted with results obtained in isolated perfused rat heart experiments. This model provides a suitable method for evaluating the uptake and extraction of radiolabeled compounds without the presence of disturbing effects always occurring in whole animal studies, e.g., effects of stress having an influence on the endocrinological system that regulates metabolic processes. Based on bidistribution results, the two monocationic complexes [99mTc(N)(PNP3)(10)]+ and [99mTc(N)(PNP3)(11)]+, which showed the highest heart accumulation, and the neutral complex [99mTc(N)(PNP5)(9)] were evaluated in the isolated rat heart model. Even though the well-established iodinated fatty acid tracer [123I]IPPA achieved the highest outcome, some of the new Tc-99m fatty acid complexes showed remarkable values of myocardial extraction. In particular, there was almost no difference in the incorporated activity between the monocationic fatty acid complexes [99mTc(N)(PNP3)(11)]+ and [99mTc(N)(PNP3)(10)]+ and the recently discovered flow tracer 99mTcN-DBODC. The similarity between the extraction values of 99mTc-MIBI and 99mTcN-DBODC had been demonstrated previously in animal studies (15, 16). The present study on the isolated heart also offered a simple possibility to investigate myocardial first pass extraction kinetics. This is demonstrated in Figure 9, which reports a more detailed analysis of the uptake and extraction kinetics of the neutral derivative [99mTc(N)(PNP5)(9)] as compared to [123I]IPPA. In the rapid, almost linear initial uptake occurring during the first 30 s of tracer infusion, both radiolabeled compounds showed a

Bioconjugate Chem., Vol. 19, No. 2, 2008 459

similar amount of uptake (12.60 ( 0.56% ID for [99mTc(N)(PNP5)(9)] vs 13.78 ( 1.32% ID for [123I]IPPA). In the subsequent period of tracer infusion, corresponding to the net accumulation by the myocardium including influx, efflux, and intracellular metabolism (23), the extraction kinetic curve of [99mTc(N)(PNP5)(9)] continued to increase but with a perceptible lower slope as compared to [123I]IPPA. This fact points to an accelerated efflux of the evaluated Tc-99m fatty acid analogue, which could be, presumably, attributed to a disturbed recognition by intracellular enzymes of fatty acid transport and metabolism. The observed increased efflux was also evidenced by a faster washout of the radiolabeled 99mTc compound out of the myocardium at the end of infusion. Whereas during the first 60-s period of labeled fatty acid perfusion only 8.35% ID of [123I]IPPA left the infused heart, this value increased up to 7.62% ID for [99mTc(N)(PNP5)(9)]. Thus, at the end of the infusion process, only 33.17% of [123I]IPPA activity washed out from perfused heart, while this washout was 44.97% for [99mTc(N)(PNP5)(9)]. The apparently altered metabolic behavior of 99mTc-labeled fatty acid was also confirmed by metabolic studies performed on intestine samples. These data revealed that these complexes are quantitatively metabolized in the liver. Interestingly, this process did not transform each Tc-99m complex to a single metabolite, but rather to a mixture of undetermined species, a result that is difficult to be attributed to a specific metabolic pathway. The comparative analysis of the kinetic behavior of [99mTc(N)(PNP5)(9)] and [123I]IPPA may allow a tentative interpretation of biodistribution data in rats. As mentioned above, the fast washout of [99mTc(N)(PNP5)(9)] suggests that the resulting radiolabeled fatty acid has partially lost its original biological properties, thus becoming a poor substrate for cardiac enzymatic activity. This may readily explain the in vivo behavior of neutral complexes characterized by some initial uptake followed by a rapid washout. Evidently, this behavior is mostly driven only by passive diffusion as a consequence of the lack of an efficient metabolic trapping. Conversely, the situation appears substantially different for monocationic complexes in which the presence of a positive charge permits their blockage on the mitochondrial membrane of myocites through the action of a well-established mechanism (24, 25). However, further studies are required to assess whether this retention could favor an easier recognition of the labeled fatty acid by cardiac enzymes and promote some metabolic activity.

CONCLUSIONS The design and preparation of a new class of asymmetrical nitride Tc-99m complexes incorporating a lateral fatty acid chain have been successfully carried out. The selective affinity of the strong electrophilic metal fragment [99mTc(N)(PNP)]2+ allowed us to obtain the final radiopharmaceuticals in high specific activity and almost quantitative yields (>95%), thus giving further support to previous studies showing the strong efficiency of this fragment as a labeling moiety for small biomolecules. However, biological evaluation of the resulting complexes in an isolated perfused heart model showed that the tethering of the metal fragment to the fatty acid chain caused some significant perturbation of its original biological properties. In particular, the comparison between the behavior of the metabolic tracer [123I]IPPA and the neutral complex [99mTc(N)(PNP5)(9)], in which the potential effect of the presence of a monopositive charge on the heart uptake mechanism was ruled out, clearly demonstrated that recognition of the labeled fatty acid as a substrate for β-oxidation was partially hampered, presumably as a result of some steric hindrance brought about by the metallic moiety onto the bioactive molecule. Additional studies are

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

required to assess whether the metabolic activity could be restored by increasing the separation between the metal fragment and the fatty acid chain through the insertion of some suitable linker.

ACKNOWLEDGMENT Financial support for this work by Nihon Medi-Physics, Tokyo, Japan is gratefully acknowledged.

LITERATURE CITED (1) Corbett, J. R. (1999) Fatty acid for imaging myocardial metabolism. Semin. Nucl. Med. 29, 237–258. (2) Schelbert, H. R. (2000) PET contributions to understanding normal and abnormal cardiac perfusion and metabolism. Ann. Biomed. Eng. 28, 922–929. (3) Mach, R. H., Kung, H. F., Jungwiwattanaporn, P., and Guo, Y. Z. (1991) Synthesis and bioditribution of a new class of 99mTc-labeled fatty acid analogs for myocardial imaging. Int. J. Appl. Radiat. Instrum. B 18, 215–226. (4) Magata, Y., Kawaguchi, T., Ukon, M., Yamamura, N., Uehara, T., Ogawa, K., Arano, Y., Temma, T., Mukai, T., Tadamura, E., and Saji, H. (2004) A Tc-99m labeled long fatty ac-id derivative for myocardial imaging. Bioconjugate Chem. 15, 389– 393. (5) Chu, T., Zhang, Y., Liu, X., Wang, Y., Hu, S., and Wang, X. (2004) Synthesis and biodis-tribution of 99mTc-carbonyltechnetium-labeled fatty acid. Appl. Radiat. Isot. 60, 845–850. (6) Jung, C. M., Kraus, W., Leibnitz, P., Pietzsch, J., Kropp, J., and Spies, H. (2002) Synthe-ses and first crystal structures of rhenium complexes derived from -functionalized fatty acid as model compounds of technetium tracers for myocardial metabolism imaging. Eur. J. Inorg. Chem. 7, 1219–1225. (7) Walther, M., Jung, C. M., Bergmann, R., Pietzsch, H. J., Rode, K., Fahmy, K., Mirtschink, P., Stehr, S., Heintz, A., Wunderlich, G., Kraus, W., Pietzsch, H. J., Kropp, J., Deußen, A., and Spies, H. (2007) Synthesis and biological evaluation of a new type of 99mTechnetium-labeled fatty acid for myocardial metabolism imaging. Bioconjugate Chem. 18, 216–230. (8) Mirtschink, P., Stehr, S. N., Pietzsch H-J., Bergmann, R., Pietzsch, J., Wunderlich, G., Heintz, A. C., Kropp, J., Spies, H., Kraus, W., Deussen, A. and Walther, M. (2007) Modified ”4 + 1” mixed ligand technetium labeled fatty acids for myocardial imaging: evaluation of myocardial uptake and biodistribution. Bioconjugate Chem. Accepted for publication. (9) Uehara, T., Uemura, T., Hirabayashi, S., Adachi, S., Odaka, K., Akizawa, H., Magata, Y., Irie, T., and Arano, Y. (2007) Technetium-99m labeled long chain fatty acid analogues metabolized by beta oxidation in the heart. J. Med. Chem. 50, 543– 549. (10) Boschi, A., Bolzati, C., Uccelli, L., Duatti, A., Benini, E., Refosco, F., Tisato, F., and Pif-fanelli, A. (2002) A class of asymmetrical nitrido technetium-99m complexes as heart imaging agents with improved biological properties. Nucl. Med. Commun. 23, 689–693. (11) Boschi, A., Uccelli, L., Bolzati, C., Duatti, A., Sabba, N., Moretti, E., Di Domenico, G., Zavattini, G., Refosco, F., and Giganti, M. (2003) Synthesis and biological evaluation of monocationic asymmetric 99mTc-nitride heterocomplexes showing high heart uptake and improved imaging properties. J. Nucl. Med. 44, 806–814. (12) Bolzati, C., Refosco, F., Cagnolini, A., Tisato, F., Boschi, A., Duatti, A., Uccelli, L., Dolmella, A., Marotta, E., and Tubaro, M. (2004) Synthesis, solution-state and solid-state structural characterization of monocationic nitrido heterocomplexes [M(N)(DTC)(PNP)]22 (M ) 99Tc, Re; DTC ) dithiocarbamate; PNP ) heterodiphos-phane). Eur. J. Inorg. Chem. 9, 1902–1913. (13) Bolzati, C., Boschi, A., Uccelli, L., Tisato, F., Refosco, F., Cagnolini, A., Duatti, A., Prakash, S., Bandoli, G., and Vittadini,

Cazzola et al. A. (2002) Chemistry of the strong electrophilic metal fragment [99Tc(N)(PXP)]2+ (PXP ) diphosphine ligand). A novel tool for the se-lective labeling of small molecules. J. Am. Chem. Soc. 124, 11468–11479. (14) Boschi, A., Duatti, A., and Uccelli, L. (2005) Development of technetium-99m and rhe-nium-188 radiopharmaceuticals containing a terminal metal-nitrido multiple bond for diagnosis and therapy. Top. Curr. Chem. 252, 85–101. (15) Hatada, K., Riou, L. M., Ruiz, M., Yamamichi, Y., Duatti, A., Lima, R. L., Goode, A. R., Watson, D. D., Beller, G. A., and Glover, D. K. (2004) 99mTc-N-DBODC5, a new myocardial perfusion imaging agent with rapid liver clearance: comparison with 99mTc-sestamibi and 99mTc-tetrofosmin in rats. J. Nucl. Med. 45, 2095–2101. (16) Hatada, K., Ruiz, M., Riou, L. M., Lima, R. L., Goode, A. R., Watson, D. D., Beller, G. A., and Glover, D. K. (2006) Organ biodistribution and myocardial uptake, washout, and redistribution kinetics of Tc-99m N-DBODC5 when injected during vasodilator stress in canine models of coronary stenoses. J. Nucl. Cardiol. 13, 779–790. (17) Tisato, F., Refosco, F., Porchia, M., Bolzati, C., Bandoli, G., Dolmella, A., Duatti, A., Boschi, A., Jung, C. M., Pietzsch, H-J., and Kraus, W. (2004) The crucial role of the di-phosphine heteroatom X in the stereochemistry and stabilization of the substitution-inert [M(N)(PXP)]2+ m etal fragments (M ) Tc, Re; PXP ) diphosphine ligand). Inorg. Chem. 43, 8617–8625. (18) Mertens, J., Eersels, J., and Vanryckeghem, W. (1987) New high yield Cu(I) assisted 123I radioiodination of 15(p-I-phenyl)9-methyl pentadecanoic acid, a potential myocar-dial tracer. Eur. J. Nucl. Med. 13, 159–60. (19) Heintz, A. C., Jung, C. M., Stehr, S. N., Mirtschink, P., Walther, M., Pietzsch, J., Berg-mann, R., Pietzsch, H-J., Spies, H., Wunderlich, G., Kropp, J., and Deussen, A. (2007) Myocardial uptake and biodistribution of newly designed technetium labeled fatty acid analogues. Nucl. Med. Commun. 28, 637–647. (20) Bolzati, C., Benini, E., Cazzola, E., Jung, C, Tisato, F., Refosco, F., Pietzsch, H. J., Spies, H., Uccelli, L., and Duatti, A. (2004) Synthesis, characterization, and biological evaluation of neutral nitrido technetium(V) mixed ligand complexes containing di-thiolates and aminodiphosphines. A novel system for linking technetium to bio-molecules. Bioconjugate Chem. 15, 628–637. (21) Boschi, A., Uccelli, L., Duatti, A., Bolzati, C., Refosco, F., Tisato, F., Romagnoli, R., Baraldi, P. G., Varani, K., and Borea, P. A. (2003) Asymmetrical nitrido tc-99m hetero-complexes as potential imaging agents for benzodiazepine receptors. Bioconjugate Chem. 14, 1279–1288. (22) Bolzati, C., Mahmood, A., Malago, E., Uccelli, L., Boschi, A., Jones, A. G., Refosco, F., Duatti, A., and Tisato, F. (2003) The [(99m)Tc(N)(PNP)](2+) metal fragment: a techne-tiumnitrido synthon for use with biologically active molecules. The N-(2-methoxyphenyl)piperazyl-cysteine analogues as examples. Bioconjugate Chem. 14, 1231–1242. (23) Stremmel, W., and Berk, P. D. (1986) Hepatocellular influx of [14C]oleate reflects mem-brane transport rather than intracellular metabolism or binding. Proc. Natl. Acad. Sci. U.S.A 83, 3086–3090. (24) Piwnica-Worms, D., Kronauge, J. F., and Chiu, M. L. (1990) Uptake and retention of hexakis (2 methoxyisobutyl isonitrile) technetium(I) in cultured chick myocardial cells. Mitochondrial and plasma membrane potential dependence. Circulation 82, 1826–1838. (25) Piwnica-Worms, D. P., Kronauge, J. F., LeFurgey, A., Backus, M., Hockett, D., Ingram, P., Lieberman, M., Holman, B. L., Jones, A. G., and Davison, A. (1994) Mitochondrial localization and characterization of 99Tc-sestamibi in heart cells by electron probe X-ray microanalysis and 99Tc-NMR spectroscopy. Magn. Reson. Imaging 12, 641–652. BC7002378