⁶⁴Cu-Labeled 2-(Diphenylphosphoryl)ethyldiphenylphosphonium Cations as Highly Selective Tumor Imaging Agents

Abstract

Radiolabeled organic cations, such as triphenylphosphonium (TPP), represents a new class of radiotracers for imaging cancers and the transport function of multidrug resistance P-glycoproteins (particularly MDR1 Pgp) by single photon emission computed tomography (SPECT) or positron emission tomography (PET). This report presents the synthesis and biological evaluation of ⁶⁴Cu-labeled 2-(diphenylphosphoryl)ethyldiphenylphosphonium (TPEP) cations as novel PET radiotracers for tumor imaging. Biodistribution studies were performed using the athymic nude mice bearing subcutaneous U87MG human glioma xenografts to explore the impact of linkers, bifunctional chelators (BFCs) and chelates on biodistribution characteristics of the ⁶⁴Cu-labeled TPEP cations. Metabolism studies were carried out using normal athymic nude mice to determine the metabolic stability of four ⁶⁴Cu radiotracers. It was found that most ⁶⁴Cu radiotracers described in this study have significant advantages over ^99mTc-Sestamibi for their high tumor/heart and tumor/muscle ratios. Both BFCs and linkers have significant impact on biological properties of ⁶⁴Cu-labeled TPEP cations. For example, ⁶⁴Cu(DO3A-xy-TPEP) has much lower liver uptake and better tumor/liver ratios than ⁶⁴Cu(DO3A-xy-TPP), suggesting that TPEP is a better mitochondrion-targeting molecule than TPP. Replacing DO3A with DO2A results in ⁶⁴Cu(DO2A-xy-TPEP)⁺, which has a lower tumor uptake than ⁶⁴Cu(DO3A-xy-TPEP). Substitution of DO3A with NOTA-Bn leads to a significant decrease in tumor uptake for ⁶⁴Cu(NOTA-Bn-xy-TPEP). The use of DOTA-Bn to replace DO3A has little impact on the tumor uptake; but the tumor/liver ratio of ⁶⁴Cu(DOTA-Bn-xy-TPEP)^- is not as good as that of ⁶⁴Cu(DO3A-xy-TPEP), probably due to the aromatic benzene ring in DOTA-Bn. Addition of an extra acetamido group in ⁶⁴Cu(DOTA-xy-TPEP) results in a lower liver uptake; but tumor/liver ratios of ⁶⁴Cu(DOTA-xy-TPEP) and ⁶⁴Cu(DO3A-xy-TPEP) are well comparable due to a faster tumor washout of ⁶⁴Cu(DOTA-xy-TPEP). Substitution of xylene with the PEG₂ linker also leads to a significant reduction in both tumor and liver uptake. MicroPET imaging studies on ⁶⁴Cu(DO3A-xy-TPEP) in athymic nude mice bearing U87MG glioma xenografts showed that the tumor was clearly visualized as early as 1 h postinjection with very high T/B contrast. There was very little metabolite (<2%) detectable in the urine and feces samples for ⁶⁴Cu(DO3A-xy-TPEP), ⁶⁴Cu(DOTA-Bn-xy-TPEP)^- and ⁶⁴Cu(NOTA-Bn-xy-TPEP). Considering both tumor uptake and T/B ratios (particularly tumor/heart, tumor/liver and tumor/muscle), it was concluded that ⁶⁴Cu(DO3A-xy-TPEP) is a promising PET radiotracer for imaging the MDR-negative tumors.

INTRODUCTION

Alteration in the mitochondrial potential (Δψ_m) is an important characteristic of cancer (1-4). It has been demonstrated that the mitochondrial potential in carcinoma cells is significantly higher than that in normal epithelial cells (5-9). For example, the difference in Δψ_m between the colon carcinoma cell line CX-1 and the control green monkey kidney epithelial cell line CV-1 was approximately 60 mV (163 mV in tumor cells versus 104 mV in normal cells). The observation that the enhanced mitochondrial potential is prevalent in tumor cell phenotype provides the conceptual basis for development of the mitochondrion-targeting pharmaceuticals and imaging probes (1-3, 10-13).

Recently, we reported several ⁶⁴Cu-labeled triphenylphosphonium (TPP) cations as radiotracers for tumor imaging by positron emission tomography (PET) in athymic nude mice bearing U87MG human glioma xenografts (14-16). We found that that ⁶⁴Cu(DO3A-xy-TPP) and ⁶⁴Cu(DO2A-xy-TPP)⁺ have relatively high tumor uptake with long tumor retention. The most striking difference between the ⁶⁴Cu-labeled TPP cations and ^99mTc-Sestamibi, the most successful radiopharmaceutical currently available for both tumor and myocardial perfusion imaging, is that they all have much lower the heart uptake (<0.6% ID/g) than ^99mTc-Sestamibi (∼18 % ID/g) at >30 min postinjection (p.i.). Their tumor/heart ratios are >40x better than that of ^99mTc-Sestamibi at 120 min p.i. The muscle uptake of ⁶⁴Cu(DO3A-xy-TPP) was undetectable at >30 min p.i. while ^99mTc-Sestamibi has very high muscle uptake (∼5 %ID/g) over the 2 h study period. Despite of their charge difference, ⁶⁴Cu(DO3A-xy-TPP) and ⁶⁴Cu(DO2A-xy-TPP)⁺ share very similar lipophilicity, tumor uptake, tumor/heart and tumor/lung ratios. However, the positive charge in ⁶⁴Cu(DO2A-xy-TPP)⁺ results in a dramatic reduction in its liver uptake. As a result, its tumor/liver ratio significantly better than that of ⁶⁴Cu(DO3A-xy-TPP) (16). Results from in vitro assays show that ⁶⁴Cu(DO3A-xy-TPP) is able to localize in mitochondria of glioma cells. MicroPET imaging data show that the tumor could be visualized as early as 30 min p.i. in the tumor-bearing mice administered with ⁶⁴Cu(DO3A-xy-TPP) (14). However, its high liver uptake remains a significant challenge for ⁶⁴Cu(DO3A-xy-TPP) to be clinically useful as a PET radiotracer.

To further improve the tumor uptake and tumor/background (T/B) ratios, we prepared several 2-(diphenylphosphoryl)ethyldiphenylphosphonium (TPEP) conjugates (Figure 1) and their ⁶⁴Cu complexes. We are particularly interested TPEP cation because the phosphoryl group (P=O) that may help improve the radiotracer hydrophilicity and excretion kinetics from the liver. Biodistribution and imaging studies were performed using athymic nude mice bearing subcutaneous U87MG glioma xenografts. The U87MG glioma cell line was chosen because it has no expression of multidrug resistance P-glycoprotein (particularly MDR1 Pgp) (17-20). This tumor-bearing animal model would allow us to evaluate the intrinsic tumor-targeting capability of ⁶⁴Cu radiotracers and their excretion kinetics from non-cancerous organs, such as heart, liver, lungs and muscle. The main objective of this study is to explore the impact of linkers, bifunctional chelators (BFCs) and ⁶⁴Cu-chelates on biodistribution characteristics and excretion kinetics of the ⁶⁴Cu-labeled TPEP cations.

Figure 1.

TPEP conjugates for preparation of the ⁶⁴Cu TPEP cations useful as radiotracers for imaging tumors by PET. The TPEP moiety is used as the “mitochondrion-targeting biomolecule” to carry ⁶⁴Cu into tumor cells that have higher mitochondrial potential than normal cells. DO3A, DO2A, DOTA and NOTA are used as BFCs for ⁶⁴Cu chelation. Different linkers are used to modify pharmacokinetics and improve T/B ratios of ⁶⁴Cu radiotracers.

EXPERIMENTAL

Materials and Instruments

Chemicals were purchased from Sigma/Aldrich (St. Louis, MO) as received. DO3A(OBu-t)₃ (1,4,7,10-tetraazacyclododecane-4,7,10-tris(t-butyl acetate)), DO2A(OBu-t)₂ (1,4,7,10-tetraazacyclododecane-4,7-bis(t-butyl acetate)), DOTA(OBu-t)₃-NHS (1,4,7,10-tetraazacyclododecane-1-(N-hydroxysuccinimide acetate)-4,7,10-tris(t-butyl acetate)) p-SCN-Bn-DOTA(2-(p-isothiocyanobenzyl)-1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid) and p-SCN-Bn-NOTA (2-(p-isothiocyanobenzyl)-1,4,7-triaazacyclononane-1,4,7-triacetic acid) were purchased from Macrocyclics Inc. (Dallas, TX). NMR (¹H, ¹³C and ³¹P) data were obtained using a Bruker DRX 300 MHz FT NMR spectrometer. Chemical shifts are reported as δ in ppm relative to TMS. ESI mass spectral data were collected using positive mode on a Finnigan LCQ classic mass spectrometer, School of Pharmacy, Purdue University. Elemental analysis was performed by Dr. H. Daniel Lee using a Perkin-Elmer Series III analyser, Department of Chemistry, Purdue University. ⁶⁴Cu was produced using a CS-15 biomedical cyclotron at Washington University School of Medicine by the ⁶⁴Ni(p,n)⁶⁴Cu nuclear reaction.

HPLC Methods

Method 1 used a LabAlliance semi-prep HPLC system equipped with a UV/vis detector (λ = 254 nm) and Zorbax C₁₈ semi-prep column (9.4 mm × 250 mm, 100 Å pore size). The flow rate was 2.5 mL/min. The mobile phase was isocratic with 90% solvent A (0.1% acetic acid in water) and 10% solvent B (0.1% acetic acid in acetonitrile) at 0 - 5 min, followed by a gradient mobile phase going from 90% solvent A and 10% solvent B at 5 min to 60% solvent A and 40% solvent B at 20 min. Method 2 used the LabAlliance HPLC system equipped with a UV/vis detector (λ = 254 nm), a β-ram IN-US detector and Vydac C₁₈ column (4.6 mm × 250 mm, 300 Å pore size). The flow rate was 1 mL/min with the mobile phase being isocratic with 90% solvent A (10 mM ammonium acetate) and 10% solvent B (acetonitrile) at 0 - 5 min, followed by a gradient mobile phase going from 10% B at 5 min to 60% B at 20 min.

(4-(Bromomethyl)benzyl)(2-(diphenylphosphoryl)ethyl)diphenylphosphonium Bromide (TPEP-xy-Br)

To a hot solution of α,α’-dibromo-p-xylene (528 mg, 2 mmol) in toluene (30 mL) was added slowly bis(diphenylphosphino)ethane monooxide (828 mg, 2 mmol). The mixture was heated to reflux for 20 h. The white solid was filtered, washed with toluene (20 mL) and diethyl ether (40 mL), and then dried under vacuum to give the expected intermediate product: 4-bromomethylbenzyl(diphenylphosphinoethane)phosphonium bromide (TPEP-xy-Br). The yield was 1.6 g (93 %). ¹H NMR (CDCl₃, chemical shift δ in ppm relative to TMS): 2.61(m, 2H, O=PCH₂); 3.05(m, 2H, PCH₂); 4.59 (s, 2H, CH₂Br); 4.75 (d, 2H, PCH₂, J_PH = 13 Hz); 6.83 (d, 4H, C₆H₄); 7.52 - 7.91 (m, 20H, C₆H₅). ESI-MS: m/z = 598.5 for [M⁺ + H] (calcd. 598 for C₃₄H₃₂OP₂Br⁺). TPEP-xy-Br was isolated as the bromide salt, and was used for the next step reaction without further purification.

(2-(Diphenylphosphoryl)ethyl)diphenyl(4-((4,7,10-tris(carboxymethyl)-1,4,7,10-tetraazacyclododecan-1-yl)methyl)benzyl)phosphonium Acetate (DO3A-xy-TPEP)

To a solution of TPEP-xy-Br (67.8 mg, 0.1 mmol) and DO3A(OBu-t)₃ (51.5 mg, 0.1 mmol) in dry DMF (3 mL) was added triethylamine (0.07 mL, 0.5 mmol). The reaction mixture was stirred at 50 °C overnight. After complete removal of volatiles, the residue was dissolved in 12 N HCl (2 mL). The resulting solution was stirred at room temperature for 30 min. Volatiles were completely removed under vacuum. The residue was then dissolved in water (3 mL), was then subjected to HPLC purification (Method 1). The fractions at ∼17.3 min were collected, combined, and lyophilized to give a white powder. The yield was 30.1 mg (33 %). ¹H NMR (D₂O, chemical shift δ in ppm relative to TMS): 1.89 (s, 3H, CH₃COO^-); 2.80 - 3.7 (m, 26H); 3.67 (s, 2H); 4.29 (d, 2H, PCH₂, J_PH = 13 Hz); 6.71(d, 2H, C₆H₄); 7.18 (d, 2H, C₆H₄); 7.38 - 7.67 (m, 20H, C₆H₅). ESI-MS: m/z = 863.2 for [M + H]⁺ (calcd. 863 for [C₄₈H₅₇N₄O₇P₂]⁺). Anal. Calcd. for C₄₈H₅₇N₄O₇P₂·CH₃CO₂·5H₂O: C, 59.28; H, 6.96; N, 5.53. Found: C, 59.48; H, 6.85; N, 5.71.

(4-((4,10-Bis(carboxymethyl)-1,4,7,10-tetraazacyclododecan-1-yl)methyl)benzyl)(2-(diphenylphosphoryl)ethyl)diphenylphosphonium Acetate (DO2A-xy-TPEP)

To a solution of TPEP-xy-Br (67.8 mg, 0.1 mmol) and DO2A(OBu-t)₂ (80 mg, 0.2 mmol) in dry DMF (3 mL) was added triethylamine (0.07 mL, 0.5 mmol). The mixture was stirred at room temperature overnight. After complete removal of volatiles, the residue was dissolved in 12 N HCl (3 mL). The solution was stirred at room temperature for 30 min. Volatiles were removed completely on a rotary evaporator. The residue was then dissolved in water (3 mL), and the product was separated from the mixture by HPLC purification (Method 1). The fractions at ∼11.8 min were collected, combined and lyophilized to give a white powder. The yield was 28.5 mg (33 %). ¹H NMR (in CDCl₃): 1.89 (s, 3H, CH₃COO^-); 2.42 (m, 4H); 2.70 - 3.01 (m, 20H); 3.40 (s, NCH₂); 4.85 (d, 2H, PCH₂, J_PH = 13 Hz); 6.78 (d, 2H, C₆H₄); 7.30 (d, 2H, C₆H₄); 7.47 - 7.83 (m, 20H, C₆H₅). ESI-MS: m/z = 805.4 for [M + H]⁺ (805 calcd. for C₄₆H₅₅N₄O₅P₂⁺). Anal. Calcd. for C₄₆H₅₅N₄O₅P₂·CH₃CO₂·6.5H₂O: C, 58.71; H, 7.29; N, 5.71. Found: C, 58.50; H, 6.89; N, 5.77.

(2-(Diphenylphosphoryl)ethyl)diphenyl(2-(2-(2-(tosyloxy)ethoxy)ethoxy)ethyl) phosphonium Tosylate (TPEP-PEG₂-OTs)

TsO-PEG2-OTs was prepared using the literature procedure.¹⁶ TPEP mono-oxide (104 mg, 0.25 mmol) and TsO-PEG₂-OTs (230 mg, 0.5 mmol) were suspended in dry toluene (10 mL). The mixture was heated to reflux for 20 h. Volatiles were removed under vacuum. The residue was dissolved in 50% DMF/H₂O mixture. The resulting solution was subject to HPLC purification (Method 1). The fractions at ∼16.5 min were collected and lyophilized to give a white solid. The yield was 65.5 mg (30%). ¹H NMR (in CDCl₃): 2.33 (s, 3 H, CH₃), 2.43 (s, 3 H, CH₃), 2.75 (m, 2H, O=PCH₂), 3.17-3.71 (m, 12 H), 4.09 (t, 2H, PCH₂), 7.12 (d, 2 H, J = 7.8 Hz), 7.32-7.78 (m, 26 H). ESI-MS: m/z = 701.17 for [M]⁺ (calcd. 701 for [C₃₉H₄₃O₆P₂S]⁺).

(2-(Diphenylphosphoryl)ethyl)diphenyl(2-(2-(2-(4,7,10-tris(carboxymethyl)-1,4,7,10-tetraazacyclododecan-1-yl)ethoxy)ethoxy)ethyl)phosphonium Tosylate (DO3A-PEG2-TPEP)

TPEP-PEG₂-OTs (24 mg, 0.027 mmol) and DO3A(t-Bu)₃ (14 mg, 0.027 mmol) were dissolved in dry DMF (3 mL). After addition of Et₃N (30 μL, 0.20 mmol), the solution was stirred at room temperature for 20 h. After removal of volatiles, the residue was dissolved in 12 N HCl (3 mL). The solution was stirred for 2 h. After complete removal of volatiles under vacuum, the residue was dissolved in water (3 mL). The resulting solution was subjected to HPLC purification (Method 1). The fractions at ∼10 min were collected, combined and lyophilized to give a white powder. The yield was 8.5 mg (30%). ¹H NMR (in CDCl₃): 2.21 (s, 3H, CH₃), 2.39-3.63 (m, 38 H), 7.19 (d, 2H), 7.35-7.72 (m, 22 H). ³¹P NMR: 32.41 (d, J = 54.4 Hz), 42.84 (d, J = 54.4 Hz). ESI-MS: m/z = 875.01 for [M]⁺ (calcd. 875 for [C₄₆H₆₁N₄O₉P₂]⁺). Anal. Calcd. for C₄₆H₆₁N₄O₉P₂·CH₃C₆H₄SO₃·H₂O: C, 59.76; H, 6.62; N, 5.71. Found: C, 59.48; H, 6.85; N, 5.62.

(4-((1,3-Dioxoisoindolin-2-yl)methyl)benzyl)(2-(diphenylphosphoryl)ethyl)diphenyl-phosphonium Bromide (TPEP-xy-PA)

TPEP-xy-Br (340 mg, 0.5 mmol) and potassium pthalimide (110 mg, 0.6 mmol) were suspended in acetonitrile (10 mL). The suspension was heated to reflux for 24 h. The solid was filtered and the filtrate was dried under vacuum to give a light-yellow solid (300 mg, 0.40 mmol, 80%). ¹H NMR (in CDCl₃): 2.83 (m, 2 H), 3.14 (m, 2 H), 4.66 (s, 2 H, CH₂N), 5.14 (d, 2 H, PCH₂, J_PH = 14.4 Hz), 6.80 (dd, 2 H, J = 8.1, 2.7 Hz), 6.93 (d, 2 H, J = 8.1 Hz), 7.45-7.90 (m, 20 H), 8.05-8.18 (m, 4 H). ³¹P NMR: 28.9 (d, J = 53.1 Hz), 33.0 (d, J = 53.1 Hz). ESI-MS: m/z = 664.03 for [M]⁺ (calcd. 664 for [C₄₂H₃₆NO₃P₂]⁺).

(4-(Aminomethyl)benzyl)(2-(diphenylphosphoryl)ethyl)diphenylphosphonium (TPEP-xy-NH₂)

The TPEP-xy-PA intermediate (300 mg, 0.40 mmol) and 0.53 mL of aqueous hydrazine (55%, 88 mmol) were added into absolute ethanol (20 mL). The resulting solution was heated to reflux for 20 h. Volatiles were removed under vacuum. The light-yellow solid was dissolved in dichloromethane (20 mL). After filtration, the solvent was evaporated to get light-yellow solid. The yield was 230 mg (92%). ¹H NMR (in CDCl₃): 2.88 (m, 2 H), 3.10 (m, 2 H), 3.72 (s, 2 H, CH₂N), 5.14 (d, 2 H, PCH₂, J_PH = 14.4 Hz), 6.85 (m, 4 H), 7.42-7.85 (m, 16 H), 8.05-8.15 (m, 4 H). ³¹P NMR: 28.8 (d, J = 51.7 Hz), 33.1 (d, J = 51.7 Hz). ESI-MS: m/z = 534.11 for [M]⁺ (calcd. 534 for [C₃₄H₃₄NOP₂]⁺).

(2-(Diphenylphosphoryl)ethyl)diphenyl(4-((2-(4,7,10-tris(carboxymethyl)-1,4,7,10-tetraazacyclododecan-1-yl)acetamido)methyl)benzyl)phosphonium Acetate (DOTA-xy-TPEP)

TPEP-xy-NH₂ (10 mg, 0.01 mmol) and PF₆ salt of DOTA(t-Bu)₃-NHS (6.7 mg, 0.01 mmol) were dissolved in dry DMF (3 mL). Upon addition of triethylamine (44 μL, 0.32 mmol), the resulting mixture was stirred at room temperature for 20 h. After removal of volatiles, the residue was dissolved in the concentrated HCl (3 mL). The solution was then stirred at room temperature for 2 h. Volatiles were completely removed under vacuum, and the residue was dissolved in water (3 mL). The product was separated from the mixture by HPLC (Method 1). The fractions at ∼18 min were collected and lyophilized to give a white powder. The yield was 5.6 mg (56%). ¹H NMR (in D₂O): 1.88 (s, 3H, CH₃COO^-); 2.55-3.35 (m, 26 H), 3.59 (d, 2 H, NCH₂), 4.12 (s, 2 H, CH₂NH), 4.21 (d, 2 H, PCH₂, J = 14.7 Hz), 6.60 (dd, 2 H, J = 7.8 and 2.2 Hz), 6.94 (d, 2 H, J = 7.8), 7.32-7.75 (m, 20 H). ³¹P NMR: 30.64 (d, J = 53.0 Hz), 43.09 (d, J = 53.0 Hz). ESI-MS: m/z = 920.1 for [M]⁺ (calcd. 920 for [C₅₀H₆₀N₅O₈P₂]⁺). Anal. Calcd. for C₅₀H₆₀N₅O₈P₂·CH₃CO₂·2H₂O: C, 61.47; H, 6.65; N, 6.89. Found: C, 61.11; H, 6.53; N, 6.96.

(2-(Diphenylphosphoryl)ethyl)diphenyl(4-((3-(4-((1,4,7,10-tetrakis(carboxymethyl)-1,4,7,10-tetraazacyclododecan-2-yl)methyl)phenyl)thioureido)methyl)benzyl)phosphonium Acetate (DOTA-Bn-xy-TPEP)

TPEP-xy-NH₂ (6 mg, 0.01 mmol) and p-SCN-Bn-DOTA (6.7 mg, 0.01 mmol) were dissolved in the equal volume mixture (2 mL) of DMF and water. The pH value in the mixture was adjusted to 8.5 with 1 N NaOH. The resulting mixture was stirred at room temperature overnight and then purified by HPLC (Method 1). The fractions at ∼17.8 min were collected. Lyophilization of the collected fractions gave a white powder. The yield was 6.3 mg (56%). ¹H NMR (in D₂O): 1.90 (s, 3H, CH₃COO^-); 2.35-3.82 (m, 29 H), 4.22 (d, 2 H, PCH₂, J_PH = 16.2 Hz), 4.49 (s, 2 H, CH₂NH), 6.55 (dd, 2 H), 6.78 (m, 2 H), 7.16 (m, 2 H), 7.32-7.53 (m, 20 H), 7.69 (m, 3 H). ³¹P NMR: 30.62 (d, J = 52.4 Hz) and 42.84 (d, J = 52.4 Hz). ESI-MS: m/z = 1085.22 for [M]⁺ (calcd. 1085 for [C₅₈H₆₇N₆O₉P₂S]⁺). Anal. Calcd. for C₅₈H₆₇N₆O₉P₂S·CH₃CO₂: C, 62.92; H, 6.16; N, 7.34. Found: C, 62.80; H, 6.46; N, 7.39.

(2-(Diphenylphosphoryl)ethyl)diphenyl(4-((3-(4-((1,4,7-tris(carboxymethyl)-1,4,7-triazonan-2-yl)methyl)phenyl)thioureido)methyl)benzyl)phosphonium Acetate (NOTA-Bn-xy-TPEP)

NOTA-Bn-xy-TPEP was prepared according to the same procedure for DOTA-Bn-xy-TPEP using TPEP-xy-NH₂ (5 mg, 0.008 mmol) and p-SCN-Bn-NOTA (3.7 mg, 0.008 mmol). The fractions at ∼15.1 min (Method 1) were collected, combined and lyophilized to give a white powder. The yield was 4.6 mg (54%). ¹H NMR (in D₂O): 1.92 (s, 3H, CH₃COO^-); 2.35-3.70 (m, 23 H), 4.47 (s, 2 H, CH₂NH), 4.21 (d, 2 H, PCH₂, J_PH = 14.1 Hz), 6.53 (d, 2 H), 6.86 (d, 2 H), 7.02-7.22 (m, 4 H), 7.30-7.56 (m, 18 H), 7.68(m, 2 H). ³¹P NMR: 30.38 (d, J = 52.6 Hz), 42.66 (d, J = 52.6 Hz). ESI-MS: m/z = 984.06 for [M]⁺ (calcd 984 for [C₅₄H₆₀N₅O₇P₂S]⁺). Anal. Calcd. for C₅₄H₆₀N₅O₇P₂S·CH₃CO₂: C, 64.42; H, 6.08; N, 6.71. Found: C, 64.29; H, 6.00; N, 6.83.

⁶⁴Cu-Labeling

To a 5 mL vial were added 0.5 mL 0.1 M NaOAc buffer (pH = 6.9) containing 50 μg of the TPEP conjugate and 0.12 mL of ⁶⁴CuCl₂ solution (1.0 - 2.0 mCi) in 0.05 N HCl. The final pH was 5.0 - 5.5. The mixture was heated at 100 °C for 30 min. After cooling to room temperature, a sample of resulting solution was analyzed by radio-HPLC (Method 2). The radiochemical purity (RCP) was >90% for all ⁶⁴Cu radiotracers with specific activity being ∼50 Ci/mmol. RCP data and HPLC retention times of ⁶⁴Cu radiotracers are listed in Table 1.

Table 1.

HPLC retention time and log P values for ⁶⁴Cu-labeled TPEP cations

Compound	RCP (%)	Retention Time (min)	Log P value
64Cu(DO3A-xy-TPEP)	> 95	17.5	- 1.69 ± 0.11
64Cu(DO3A-PEG2-TPEP)	> 95	17.7	- 3.01 ± 0.09
64Cu(DO2A-xy-TPEP)+	> 95	17.5	- 1.90 ± 0.10
64Cu(DOTA-xy-TPEP)	> 90	19.9	- 2.45 ± 0.02
64Cu(DOTA-Bn-xy-TPEP)	> 98	20.2	- 1.92 ± 0.03
64Cu(NOTA-Bn-xy-TPEP)	> 98	20.2	- 1.75 ± 0.03

Dose Preparation

All newly synthesized ⁶⁴Cu radiotracers were purified by HPLC before being used for biodistribution studies. Volatiles in the HPLC mobile phases were completely removed under vacuum (< 10 mmHg) at 40 - 50 °C. The residue was dissolved in saline to ∼25 μCi/mL. The resulting solution was filtered with a 0.20 μ syringe-driven filter unit to eliminate any particles. Each animal was administered intravenously with ∼0.10 mL of the dose solution. For the imaging study, ⁶⁴Cu(DO3A-xy-TPEP) was prepared and the resulting mixture was used without purification. The reaction mixture was diluted to ∼5 mCi/mL with saline. The injected dose for each tumor-bearing mouse was about 250 μCi of ⁶⁴Cu(DO3A-xy-TPEP).

Solution Stability

For solution stability, ⁶⁴Cu(DO3A-xy-TPEP), ⁶⁴Cu(DO2A-xy-TPEP), ⁶⁴Cu(NOTA-Bn-xy-TPEP) and ⁶⁴Cu(DOTA-Bn-xy-TPEP)^- were prepared and purified by HPLC (Method 2). Volatiles in the HPLC mobile phase were removed under vacuum. The residue was dissolved in saline to ∼1 mCi/mL. Samples were analyzed by HPLC (Method 2) at 0, 1, 2, 4 and 12 h post purification. In the EDTA challenge experiment, the HPLC purified ⁶⁴Cu radiotracers were dissolved in 25 mM phosphate buffer (pH = 7.4) containing EDTA (1 mg/mL) to 1 mCi/mL. Samples of the resulting solution were analyzed by radio-HPLC (Method 2) at 0, 1, 2, 4 and 12 h post purification.

Partition Coefficient

All new ⁶⁴Cu radiotracers were purified before being used for partition coefficient determination. HPLC purification is needed to eliminate potential interference from other radio-impurities. After complete removal of volatiles in the HPLC mobile phases under vacuum (< 10 mmHg) at 40 - 50 °C, the residue was dissolved in a mixture of 3 mL saline and 3 mL n-octanol. The mixture was stirred vigorously for 20 min at room temperature, and was then transferred to an Eppendorf microcentrifuge tube. The tube was centrifuged at 12,500 rpm for 5 min. Samples in triplets from n-octanol and aqueous layers were obtained, and were counted a Perkin Elmer Wizard - 1480 automatic γ-counter (Shelton, CT). The log P value was reported as an average of the data obtained in three independent measurements.

Tumor-Bearing Animal Model

The ex-vivo biodistribution studies were performed using athymic nude mice bearing U87MG human glioma xenografts in compliance the NIH animal experiment guidelines (Principles of Laboratory Animal Care, NIH Publication No. 86-23, revised 1985). The animal protocol has been approved by Purdue University Animal Care and Use Committee (PACUC). Female athymic nu/nu mice (4 - 5 weeks of age) were purchased from Harlan (Charles River, MA). Each mouse was implanted with 5 × 10⁶ the U87MG glioma cells into the upper shoulder flank. Three to four weeks after inoculation, the tumor size was in the range of 0.3 - 0.6 g, and animals were used for both biodistribution and imaging studies.

Biodistribution Protocol

Twelve tumor-bearing mice (20 - 25 g) were randomly divided into four groups. The ⁶⁴Cu radiotracer (∼2.5 μCi dissolved in 0.1 mL saline) was administered each animal via tail vein. Three animals were sacrificed by sodium pentobarbital overdose (100 mg/kg) at 5, 30, 60, and 120 min p.i. The blood sample was withdrawn from the heart. Organs were excised, washed with saline, dried with absorbent tissue, weighed, and counted on a Perkin Elmer Wizard - 1480 γ-counter (Shelton, CT). Organs of interest included tumor, brain, heart, spleen, lungs, liver, kidneys, muscle and intestine. The organ uptake was calculated as a percentage of the injected dose per organ (%ID/organ) and a percentage of the injected dose per gram of organ tissue (%ID/g).

Calibration of microPET

Scanner activity calibration was performed to map between microPET image units and units of activity concentration. A pre-weighed 50 mL centrifuge tube was filled with solution containing ⁶⁴CuCl₂ (∼ 9.3 MBq as determined by the dose calibrator) was used to simulate whole body of the mouse. This tube was weighed, centered in the scanner aperture, and imaged for 30 min static image. From the sample weight and the density of 1 g/mL, the activity concentration was calculated as μCi/mL. Eight planes were acquired in the coronal section. A rectangular region of interest (ROI) (counts/pixel/s) was drawn on the middle of 8 coronal planes. The calibration factor was obtained by dividing the known radioactivity in the cylinder (μCi/mL) by the image ROI. This calibration factor was determined periodically and did not vary significantly with time.

MicroPET Imaging

MicroPET imaging of the tumor-bearing mice was performed using a microPET R4 rodent model scanner (Concorde Microsystems, Knoxville, TN). The tumor-bearing mice (n = 3) were imaged in the prone position in the microPET scanner. Each tumor-bearing mouse was injected with ∼250 μCi of ⁶⁴Cu(DO3A-xy-TPEP) via the tail vein, then anesthetized with 2% isoflurane and placed near the center of the FOV where the highest resolution and sensitivity are obtained. Multiple static scans were obtained at 0.5, 1.0, 2 and 20 h p.i. The images were reconstructed by a two-dimensional ordered subsets expectation maximum (OSEM) algorithm. No correction was necessary for attenuation or scatter. At each microPET scan, the ROIs were drawn over the tumor and major organs on decay-corrected whole-body coronal images. The average radioactivity concentration within the tumor or an organ was obtained from mean pixel values within the multiple ROI volume, which were converted to counts/mL/min by using the calibration constant C. Assuming that the tissue density is 1 g/mL, the ROIs were converted to counts/g/min, and were then divided by the total administered activity to obtain an imaging ROI-derived percentage administered activity per gram of tissue (%ID/g).

Metabolism

Metabolic stability was evaluated in normal nude mice for ⁶⁴Cu(DO3A-xy-TPEP), ⁶⁴Cu(DOTA-xy-TPEP), ⁶⁴Cu(DOTA-Bn-xy-TPEP)^- and ⁶⁴Cu(NOTA-Bn-xy-TPEP). Each mouse was administered with 100 μCi of ⁶⁴Cu radiotracer via tail vein. Urine samples were collected at 30 and 120 min p.i. by manual void, and were mixed with equal volume of acetonitrile. The mixture was centrifuged at 8,000 rpm. The supernatant was collected and filtered through a 0.20 μm Millex-LG syringe-driven filter unit to remove the precipitate and large proteins. The filtrate was analyzed by radio-HPLC (Method 2). The feces samples were collected at ∼120 min p.i., and were suspended in a mixture of 50% acetonitrile aqueous solution. The mixture was vortexed for 5 - 10 min. After centrifuging at 8,000 rpm for 5 min, the supernatant was collected and passed through a 0.20 μm Millex-LG syringe-driven filter unit to remove the precipitate or particles. The filtrate was then analyzed by radio-HPLC (Method 2).

Stability of ⁶⁴Cu(DO3A-xy-TPEP) in Liver

The liver tissue was harvested at 120 min p.i. from the mouse administered with ⁶⁴Cu(DO3A-xy-TPEP) (∼100 μCi), counted in a γ-counter (Perkin Elmer Wizard - 1480, Shelton, CT) for total liver radioactivity, and was then homogenized. The homogenate was mixed with 2 mL of saline. After centrifuging at 8,000 rpm for 5 min, the supernatant was collected and counted on a γ-counter (Perkin Elmer Wizard - 1480, Shelton, CT) to determine the percentage of radioactivity recovery. After filtration through a 0.20 μm Millex-LG filter unit to remove foreign particles, the filtrate was then analyzed by radio-HPLC (Method 2).

Data and Statistical Analysis

The biodistribution data and T/B ratios are reported as an average plus the standard variation based on results from three tumor-bearing mice at each time point. Comparison between two different radiotracers was made using the two-way ANOVA test (GraphPad Prim 5.0, San Diego, CA). The level of significance was set at p < 0.05.

RESULTS

Synthesis of TPEP Conjugates

DO3A-xy-TPEP, DO2A-xy-TPEP and DO3A-PEG₂-TPEP were prepared according to Chart I. They were designed to examine the impact of linkers (xy vs. PEG₂) and molecular charge (⁶⁴Cu-DO3A vs. ⁶⁴Cu-DO2A) on organ uptake and excretion kinetics of the ⁶⁴Cu radiotracers. DOTA-xy-TPEP, DOTA-Bn-xy-TPEP, NOTA-Bn-xy-TPEP were designed to examine the impact of BFCs (DO3A vs. DOTA, DOTA-Bn and NOTA-Bn) on biodistribution patterns of ⁶⁴Culabeled TPEP cations. DOTA-xy-TPEP was prepared from the reaction of DOTA(OBu-t)₃-NHS with TPEP-xy-NH₂ (Chart II), followed by hydrolysis of DOTA(OBu-t)₃-xy-TPEP in 12 N HCl. DOTA-Bn-xy-TPEP and NOTA-Bn-xy-TPEP were prepared by reacting TPEP-xy-NH₂ with p-SCN-Bn-DOTA and p-SCN-Bn-NOTA (Chart II), respectively. All the TPEP conjugates, except DO3A-PEG₂-TPEP, were obtained as acetate salts after lyophilization since the HPLC mobile phases contain 0.1% acetic acid. They all have been characterized by NMR, ESI-MS and elemental analysis. The purity of the TPEP conjugates were >95% before being used for ⁶⁴Cu-labeling. Both TPEP-PEG₂ and DO3A-PEG₂-TPEP were isolated as tosylate salts, as evidenced by the proton signals of tosylate anion in their ¹H NMR spectra, and the molecular ion at m/z = 171 in their negative mode ESI-mass spectra. It is not clear why TPEP-PEG₂ and DO3A-PEG₂-TPEP were obtained as tosylate salts while other TPEP conjugates were isolated as acetate salts after HPLC purification and lyophilization.

Chart I.

Synthesis of TPEP and TPPP Conjugates.

Chart II.

Synthesis of TPEP Conjugates.

Radiochemistry

All ⁶⁴Cu radiotracers were prepared by reacting ⁶⁴CuCl₂ with the TPEP conjugate in 0.1 M NaOAc buffer (pH = 5.0 - 5.5) at 100 °C for 30 min, and were analyzed using the same reversed-phase HPLC method (Method 2). The RCP for all new ⁶⁴Cu radiotracers was >95% with the specific activity being >50 Ci/mmol. Their partition coefficients were determined in a 50%:50% (v:v) mixture of n-octanol and 25 mM phosphate buffer (pH = 7.4). Their log P values and HPLC retention times are listed in Table 1. All new ⁶⁴Cu radiotracers are very hydrophilic with their log P values < -1.50. BFCs and linkers have a significant impact on lipophilicity of the ⁶⁴Cu radiotracer. For example, ⁶⁴Cu(DOTA-xy-TPEP) (log P = -2.45 ± 0.02) is more hydrophilic than ⁶⁴Cu(DO3A-xy-TPEP) (log P = -1.69 ± 0.11) due to the replacement of DO3A with DOTA. Substitution of xylene with the PEG₂ linker made ⁶⁴Cu(DO3A-PEG₂-TPEP) (log P = -3.06 ± 0.08) more hydrophilic than ⁶⁴Cu(DO3A-xy-TPEP). Despite their charge difference, the lipophilicity of ⁶⁴Cu(DO3A-xy-TPEP) (log P = -1.69 ± 0.11) was close to that of ⁶⁴Cu(DO2A-xy-TPEP)⁺ (log P = -1.90 ± 0.10). ⁶⁴Cu(NOTA-Bn-xy-TPEP) (log P = -1.75 ± 0.03) also shared the similar lipophilicity with ⁶⁴Cu(DO3A-xy-TPEP) in spite of the extra aromatic phenyl group. Since NOTA-Bn has the same number of acetate chelating arms as DO3A, ⁶⁴Cu in ⁶⁴Cu-NOTA-Bn is most likely coordinated by six N₃O₃ donor atoms in the distorted-pseudo-prismatic coordination geometry as observed in the solid state structure of Na[Cu(NOTA)]·2NaBr·8H₂O (21). Therefore, it is reasonable to believe that ⁶⁴Cu(NOTA-Bn-xy-TPEP) exists as its Zwitterion form. In ⁶⁴Cu(DOTA-Bn-xy-TPEP)^-, however, ⁶⁴Cu is likely coordinated by only six (N₄O₂) of its eight donor atoms (N₄O₄) in the distorted octahedral coordination geometry, as observed in the solid state structure of Cu(DOTA)^2- with two acetate-O atoms remaining uncoordinated (22, 23). The extra acetate groups will results in the negative molecular charge under physiological conditions (pH = 7.4), ⁶⁴Cu(DOTA-Bn-xy-TPEP)^- (log P = -1.92 ± 0.03) is slightly more hydrophilic than ⁶⁴Cu(DO3A-xy-TPEP).

Solution Stability

The EDTA challenge experiment was used to study the solution stability of ⁶⁴Cu(DO3A-xy-TPEP), ⁶⁴Cu(DO2A-xy-TPEP)⁺, ⁶⁴Cu(DOTA-Bn-xy-TPEP)^- and ⁶⁴Cu(NOTA-Bn-xy-TPEP). These four ⁶⁴Cu radiotracers were selected because they represent four different types of BFCs (DO3A, DO2A, DOTA-Bn and NOTA-Bn) for ⁶⁴Cu chelation. The purpose of these studies was to demonstrate that the ⁶⁴Cu radiotracer remains intact before being injected into the tumor-bearing mice. Table 2 summarizes their solution stability data in the presence of excess EDTA (3 mM, pH = 7.4). On the basis of these data, it becomes quite clear that all four ⁶⁴Cu-labeled TPEP cations are stable for >12 h in the presence of 3 mM EDTA (25 mM phosphate buffer, pH = 7.4).

Table 2.

Solution stability data for the HPLC purified ⁶⁴Cu-labeled TPEP cations in the presence of excess EDTA (1 mg/mL in 25 mM phosphate buffer, pH = 7.5)

Time Post-Purification	0.5 h	1 h	2 h	4 h	12 h
64Cu(DO3A-xy-TPEP)	99	99	98	98	97
64Cu(DO2A-xy-TPEP)+	99	98	98	97	95
64Cu(DOTA-Bn-xy-TPEP)-	99	98	99	98	97
64Cu(NOTA-Bn-xy-TPEP)	99	98	99	98	97

Biodistribution Data

Biodistribution characteristics of six new ⁶⁴Cu radiotracers were evaluated in athymic nude mice bearing U87MG glioma xenografts. The Western Blot data (Figure SI) clearly demonstrated that there is no MDR1 Pgp expression on U87MG glioma xenografts. Selected biodistribution data and T/B ratios are listed in Tables 3 and 4, respectively. Detailed biodistribution data and T/B ratios are summarized in Tables SI-SVI. The main objective of these studies was to explore the impact of linkers and ⁶⁴Cu chelates on biodistribution characteristics and excretion kinetics of ⁶⁴Cu-labeled TPEP cations.

Table 3.

Selected biodistribution data for ⁶⁴Cu radiotracers in athymic nude mice (n = 3 unless specified) bearing U87MG glioma xenografts. The organ uptake is expressed as %ID/gram

Time	Tumor	Blood	Lungs	Liver	Kidneys	Heart	Muscle
64Cu(DO3A-xy-TPEP)
5 min	5.27±1.21	9.52±1.27	7.72±0.78	8.03±1.41	31.79±8.51	4.04±0.75	3.07±0.48
30 min	2.09±0.53	1.68±0.39	2.27±0.48	6.48±1.39	6.79±0.97	0.94±0.19	0.55±0.17
60 min	2.58±0.31	0.65±0.12	1.84±0.38	6.40±1.94	4.24±1.08	0.67±0.17	0.22±0.11
120 min*	2.35±0.57	0.36±0.07	1.51±0.18	5.59±1.39	2.98±0.53	0.55±0.09	0.10±0.06
64Cu(DO3A-PEG2-TPEP)
5 min	5.06±0.84	7.19±1.40	6.08±0.57	6.44±1.29	29.11±2.57	3.64±0.25	2.45±0.36
30 min	1.24±0.31	1.03±0.09	1.63±0.27	3.76±0.06	4.69±0.19	0.71±0.06	0.39±0.13
60 min	1.41±0.35	0.49±0.11	1.08±0.07	3.20±0.56	2.70±0.19	0.42±0.03	0.17±0.10
120 min	1.60±0.45	0.33±0.05	1.06±0.08	2.91±0.12	2.53±0.44	0.44±0.07	0.07±0.02
64Cu(DO2A-xy-TPEP)+
5 min	5.06±0.43	9.25±0.84	7.51±0.78	3.95±0.16	25.60±1.21	3.26±0.24	2.81±0.22
30 min	1.66±0.21	2.13±0.73	2.18±0.43	2.18±0.29	5.99±1.85	0.78±0.23	0.51±0.25
60 min	0.78±0.32	0.44±0.17	0.89±0.12	1.91±0.02	2.44±0.43	0.41±0.26	0.25±0.11
120 min	0.51±0.08	0.24±0.13	0.47±0.06	1.49±0.12	1.28±0.10	0.10±0.06	0.07±0.03
64Cu(DOTA-xy-TPEP)
5 min	4.84±0.96	11.65±1.62	7.27±0.63	5.61±0.37	21.61±5.63	3.96±0.91	2.65±0.61
30 min	1.99±0.41	2.99±0.03	2.78±0.32	4.47±0.45	7.72±0.76	1.14±0.19	1.31±0.65
60 min	1.78±0.65	1.74±0.51	2.00±0.36	3.93±0.84	5.53±0.56	0.90±0.23	0.31±0.12
120 min	1.39±0.59	0.33±0.01	1.00±0.28	2.99±0.21	2.84±0.50	0.30±0.07	0.01±0.01
64Cu(DOTA-Bn-xy-TPEP)
5 min	3.03±0.52	14.35±2.38	7.25±1.28	10.32±2.02	18.07±5.96	4.07±0.66	2.65±0.43
30 min	2.47±0.09	4.45±0.27	3.95±0.14	9.24±0.98	9.02±0.13	1.86±0.23	1.06±0.04
60 min	2.68±0.07	2.29±0.58	3.48±0.54	9.28±0.91	7.72±0.25	1.73±0.43	0.56±0.09
120 min	1.91±0.30	1.14±0.18	2.31±0.16	8.66±0.12	5.44±0.40	1.03±0.17	0.23±0.03
64Cu(NOTA-Bn-xy-TPEP)
5 min	2.74±0.05	11.55±1.93	8.39±1.05	11.17±1.63	13.52±2.14	4.09±0.96	1.83±0.83
30 min	1.06±0.23	2.58±0.84	2.30±0.61	5.86±1.37	3.28±0.68	0.80±0.35	0.92±0.42
60 min	0.47±0.04	0.92±0.04	0.81±0.09	3.99±0.93	1.37±0.09	0.19±0.00	0.10±0.06
120 min	0.28±0.04	0.30±0.07	0.35±0.08	2.20±0.18	1.08±0.06	0.04±0.02	0.02±0.03

^*n = 6 due to difficulty to determine the radiotracer uptake in muscle.

Table 4.

T/B ratios for ⁶⁴Cu radiotracers in athymic nude mice (n = 3 unless specified) bearing U87MG human glioma xenografts

Time	Tumor/Blood	Tumor/Heart	Tumor/Lung	Tumor/Muscle	Tumor/Liver
64Cu(DO3A-xy-TPEP)
5 min	0.57±0.13	1.26±0.34	0.63±0.15	1.57±0.49	0.68±0.27
30 min	1.48±0.52	3.06±0.82	1.11±0.53	5.08±4.37	0.43±0.18
60 min	4.66±0.65	4.73±0.45	1.45±0.56	17.96±9.11	0.54±0.10
120 min*	5.87±1.91	5.42±0.54	1.39±0.39	20.42±9.72	0.64±0.15
64Cu(DO3A-PEG2-TPEP)
5 min	0.72±0.15	1.38±0.16	0.84±0.17	2.09±0.39	0.80±0.13
30 min	1.21±0.31	1.78±0.54	0.77±0.15	3.38±0.96	0.33±0.18
60 min	2.99±1.03	3.40±1.00	1.31±0.37	9.99±0.4.74	0.45±0.15
120 min	4.83±1.05	3.66±0.77	1.51±0.36	23.95±9.63	0.55±0.16
64Cu(DO2A-xy-TPEP)+
5 min	0.55±0.06	1.55±0.14	0.68±0.08	1.81±0.23	1.28±0.12
30 min	0.87±0.37	2.30±0.83	0.79±0.20	4.04±2.24	0.78±0.16
60 min	1.94±0.84	2.26±1.00	0.90±0.44	3.64±2.35	0.41±0.17
120 min	2.63±1.47	6.32±2.92	1.09±0.11	7.83±3.05	0.34±0.06
64Cu(DOTA-xy-TPEP)
5 min	0.41±0.04	1.24±0.18	0.66±0.07	1.86±0.36	0.86±0.12
30 min	0.67±0.14	1.78±0.51	0.71±0.07	1.65±0.39	0.45±0.07
60 min	1.08±0.46	2.05±0.79	0.89±0.21	5.18±1.17	0.47±0.22
120 min	4.21±1.66	4.61±1.90	1.36±0.29	7.65±1.04	0.47±0.23
64Cu(DOTA-Bn-xy-TPEP)
5 min	0.21±0.04	0.75±0.08	0.42±0.06	1.15±0.07	0.30±0.07
30 min	0.56±0.04	1.34±0.17	0.63±0.04	2.32±0.07	0.27±0.02
60 min	1.21±0.26	1.61±0.36	0.78±0.11	4.92±0.83	0.29±0.02
120 min	1.69±0.26	1.88±0.43	0.83±0.14	8.00±1.54	0.22±0.04
64Cu(NOTA-Bn-xy-TPEP)
5 min	0.24±0.03	0.69±0.14	0.33±0.04	1.68±0.63	0.25±0.03
30 min	0.42±0.07	1.42±0.47	0.46±0.03	1.25±0.36	0.18±0.01
60 min	0.51±0.03	2.42±0.25	0.58±0.07	5.59±2.86	0.12±0.02
120 min	0.94±0.19	3.19±1.12	0.80±0.08	5.22±0.40	0.13±0.01

^*n = 6 due to difficulty to determine the tumor/muscle muscle ratios.

Comparison with _99mTc-Sestamibi and ₆₄Cu(DO3A-xy-TPP)

Figure 2 compares the selected organ uptake and T/B ratios between ⁶⁴Cu(DO3A-xy-TPEP), ⁶⁴Cu(DO3A-xy-TPP) and ^99mTc-Sestamibi. This comparison allows us to demonstrate advantages of the ⁶⁴Cu-labeled TPEP cations over ^99mTc-Sestamibi with respect to tumor uptake and T/B ratios, and to assess the impact of “targeting moiety” (TPEP vs. TPP) on biodistribution patterns of ⁶⁴Cu radiotracer. ^99mTc-Sestamibi was used as the “control” since it has been clinically used for imaging tumors and monitoring the tumor MDR transport functions (25-33). Biodistribution data and T/B ratios for ⁶⁴Cu(DO3A-xy-TPP) and ^99mTc-Sestamibi were obtained from our previous report (14).

Figure 2.

Direct comparison of organ uptake and T/B ratios between ⁶⁴Cu(DO3A-xy-TPEP), ⁶⁴Cu(DO3A-xy-TPP) and ^99mTc-Sestamibi in the athymic nude mice bearing U87MG human glioma xenografts.

The most striking differences between ⁶⁴Cu(DO3A-xy-TPEP) and ^99mTc-Sestamibi are their uptake in the heart and muscle and their T/B (tumor/heart, tumor/liver, tumor/lung and tumor/muscle) ratios. For example, the heart uptake of ⁶⁴Cu(DO3A-xy-TPEP) was <1% ID/g at >30 min p.i., while the heart uptake of ^99mTc-Sestamibi was 19.22 ± 7.62 at 5 min p.i. and 19.19 ± 5.32 %ID/g at 120 min p.i. (14). The muscle uptake of ⁶⁴Cu(DO3A-xy-TPEP) was almost undetectable at ≥60 min p.i. while ^99mTc-Sestamibi had a high muscle uptake (4.84 ± 1.22 and 5.45 ± 1.24 %ID/g at 5 and 120 min p.i., respectively) over the 2 h period (Table 3). ⁶⁴Cu(DO3A-xy-TPEP) had the tumor uptake that is significantly (p <0.05) higher than that of ^99mTc-Sestamibi at all four time points. The tumor/heart ratio of ⁶⁴Cu(DO3A-xy-TPEP) at 120 min p.i. was more than 50x better than that of ^99mTc-Sestamibi. Its tumor/lung and tumor/liver ratios were much better than those of ^99mTc-Sestamibi (Figure 3). ⁶⁴Cu(DO3A-xy-TPEP) had a tumor uptake that is well comparable to that of ⁶⁴Cu(DO3A-xy-TPP) at >30 min p.i.; but its liver uptake was significantly lower (p < 0.01) than that of ⁶⁴Cu(DO3A-xy-TPP) at all four time points (Table 3). As a result, ⁶⁴Cu(DO3A-xy-TPEP) has tumor/liver ratios >3x better than those of ⁶⁴Cu(DO3A-xy-TPP). Thus, TPEP is better than TPP as the mitochondrion-targeting molecules.

Figure 3.

Impact of BFCs on organ uptake and T/B ratios of the ⁶⁴Cu-labeled TPEP cations.

Impact of BFCs

Figure 3 compares the organ uptake (heart, liver and muscle) and T/B ratios of ⁶⁴Cu(DO3A-xy-TPEP), ⁶⁴Cu(DO2A-xy-TPEP)⁺, ⁶⁴Cu(DOTA-Bn-xy-TPEP)^- and ⁶⁴Cu(NOTA-Bn-xy-TPEP). This comparison allows us to assess the impact of BFCs on biological properties of ⁶⁴Cu radiotracers. ⁶⁴Cu(DO3A-xy-TPEP) has the lipophilicity (log P = -1.69 ± 0.11) close to that of ⁶⁴Cu(DO2A-xy-TPEP)⁺ (log p = -1.90 ± 0.10) despite their different charge. While the tumor uptake of ⁶⁴Cu(DO3A-xy-TPEP) remained unchanged between 30 and 120 min p.i., there was a significant tumor washout for ⁶⁴Cu(DO2A-xy-TPEP)⁺ over the 2 h period. The liver uptake of ⁶⁴Cu(DO2A-xy-TPEP)⁺ was significantly (p < 0.01) lower than that of ⁶⁴Cu(DO3A-xy-TPEP) at all four time points. Its tumor/liver ratio at 120 min p.i. was also significantly (p < 0.05) lower. The lipophilicity of ⁶⁴Cu(DOTA-Bn-xy-TPEP)^- (log p = -1.92 ± 0.03) is close to that of ⁶⁴Cu(DO3A-xy-TPEP). Even though they share a similar tumor uptake, the uptake of ⁶⁴Cu(DOTA-Bn-xy-TPEP)^- in the heart, liver and muscle was significantly higher at >30 min p.i. As a result, the T/B ratios of ⁶⁴Cu(DOTA-Bn-xy-TPEP)^- was lower than that those of ⁶⁴Cu(DO3A-xy-TPEP). ⁶⁴Cu(NOTA-Bn-xy-TPEP) has the same overall charge as ⁶⁴Cu(DO3A-xy-TPEP); but its tumor uptake and tumor/heart ratios were much lower than those of ⁶⁴Cu(DO3A-xy-TPEP). Apparently, the BFCs and their ⁶⁴Cu chelates have a significant impact on both tumor uptake and T/B ratios of the radiotracer.

Linker Effects

Figure 4 illustrates a comparison of the tumor uptake and tumor/heart ratios between ⁶⁴Cu(DO3A-xy-TPEP), ⁶⁴Cu(DO3A-PEG₂-TPEP) and ⁶⁴Cu(DOTA-xy-TPEP). The addition of an extra acetamide group between DO3A and TPEP resulted in a significant reduction of liver uptake. However, the tumor/liver ratios (Table 4) of ⁶⁴Cu(DOTA-xy-TPEP) and ⁶⁴Cu(DO3A-xy-TPEP) were comparable within the experimental error, probably due to the faster tumor washout of ⁶⁴Cu(DOTA-xy-TPEP). Substitution of xylene with PEG₂ also leads to a significant reduction of the uptake in tumor and liver. As a result, ⁶⁴Cu(DO3A-PEG₂-TPEP) and ⁶⁴Cu(DO3A-xy-TPEP) shared very similar tumor/liver ratios.

Figure 4.

Impact of the linker on organ uptake and T/B ratios of the ⁶⁴Cu-labeled TPEP cations.

Pet Imaging

We performed a microPET imaging study on ⁶⁴Cu(DO3A-xy-TPEP) using the athymic nude mice bearing U87MG glioma xenografts. Figure 5 shows microPET images of the tumor-bearing mouse administered with ∼250 μCi of ⁶⁴Cu(DO3A-xy-TPEP) at 1, 4 and 24 h p.i. The tumor was clearly visualized as early as 1 h p.i. with very high T/B contrast. No significant radioactivity accumulation was detected in the brain, heart and muscle. After normalization, the tumor uptake of ⁶⁴Cu(DO3A-xy-TPEP) was 2.57 ± 0.41 %ID/g, 3.51 ± 1.20 %ID/g, and 3.11 ± 1.44 %ID/g at 1, 4 and 24 h p.i., respectively. These data are completely consistent with that observed in the ex-vivo biodistribution study (Figure 3), and have clearly demonstrated that ⁶⁴Cu(DO3A-xy-TPEP) is useful for imaging the MDR-negative tumors.

Figure 5.

The decay-corrected whole-body coronal microPET images (top) and the radioactivity accumulation quantification (n = 3, mean ± SD) in selected organs (bottom) of athymic nude mice bearing U87MG tumor administered with ∼250 μCi of ⁶⁴Cu(DO3A-xy-TPEP). Arrows indicate the presence of glioma tumors.

Metabolic Properties

Metabolism studies were performed on ⁶⁴Cu(DO3A-xy-TPEP), ⁶⁴Cu(DOTA-xy-TPEP), ⁶⁴Cu(DOTA-Bn-xy-TPEP)^- and ⁶⁴Cu(NOTA-Bn-xy-TPEP) using normal mice. These four ⁶⁴Cu radiotracers were selected because of their difference in ⁶⁴Cu chelate. Since they were excreted from the renal and hepatobiliary routes, we tried to collect both urine and feces samples from the normal mice administered with the ⁶⁴Cu radiotracer. A reversed-phase HPLC method (Method 2) was used to analyze the collected urine and feces samples. Figure 6 shows representative radio-HPLC chromatograms of ⁶⁴Cu(DO3A-xy-TPEP) (left) and ⁶⁴Cu(DOTA-xy-TPEP) (right) in saline before injection, in urine at 30 and 120 min p.i., and in feces at 120 min p.i. There was very little metabolite (<2%) detectable for ⁶⁴Cu(DO3A-xy-TPEP) in both urine and feces samples. Similar metabolic stability was also observed for ⁶⁴Cu(DOTA-Bn-xy-TPEP)^- and ⁶⁴Cu(NOTA-Bn-xy-TPEP) (Figure SII). ⁶⁴Cu(DOTA-xy-TPEP) is excreted without significant metabolism via renal route (Figure 6). However, only 15% of it remained intact in feces, and 85% of radioactivity appeared at 4 min. Thus, the extra acetamido group has a significant impact on metabolic stability of ⁶⁴Cu(DOTA-xy-TPEP) during its hepatobiliary excretion while it had no effect on its metabolic stability during renal excretion.

Figure 6.

Typical reversed-phase radio-HPLC chromatograms (Method 3) of ⁶⁴Cu(DO3A-xy-TPEP) (left) and ⁶⁴Cu(DOTA-xy-TPEP) (right) in saline before injection (A), in the urine sample at 30 min p.i. (B), in the urine sample at 120 min p.i. (C), and in the feces sample (D) at 120 min p.i.

Transchelation of ₆₄Cu(DO3A-xy-TPEP) in Liver

We examined the in vivo stability of ⁶⁴Cu(DO3A-xy-TPEP) in the liver. About 50% of total liver activity was recovered in the liver homogenate. The liver homogenate samples were analyzed using the reversed-phase and size-exclusion radio-HPLC methods. Figure SIII shows radio-HPLC chromatograms of the supernatant from liver homogenate. Using the size-exclusion HPLC method, we were able to identify four major radiometric peaks in its HPLC chromatogram (Figure SIII: bottom). While the identity of these radiometric peaks are not known, one thing is sure that ⁶⁴Cu(DO3A-xy-TPEP) underwent extensive tranchelation in the liver during the 2 h study period.

DISCUSSION

For the new radiotracer to be successful as a tumor imaging agent by PET or SPECT (single photon emission computed tomography), it must show a high tumor uptake. Since most of high-incidence tumors occur in torso (namely lung, colorectal and breast cancers metastatic to the lymphatic system), it is highly desirable for the radiotracer to clear from the liver and lungs so that the high T/B ratios can be achieved and clinically useful tumor images can be obtained in a short period of time. Cationic radiotracers, such as ^99mTc-Sestamibi and ^99mTc-Tetrofosmin, have been clinically used for imaging tumors of different origin and the transport function of MDR Pgp by SPECT (25-33). However, their cancer diagnostic values are often limited due to their insufficient tumor localization and high uptake in the non-cancerous organs, such as heart, liver and muscle, which makes it difficult to detect small lesions in the chest and abdominal regions. Radiolabeled lipophilic organic cations, such as 4-(¹⁸F-benzyl)triphenylphosphonium (¹⁸F-BzTPP), have also been proposed as radiotracers for tumor imaging by PET (34-38). Their high uptake in the heart and liver may also impose a significant challenge for early detection of small tumors in the chest and abdominal regions. Thus, there is an unmet need for the radiotracers that have high tumor-selectivity (high tumor uptake with little accumulation in non-cancerous organs) and are sensitive to the mitochondrial potential changes at the early stage of tumor growth.

In our previous studies, we have demonstrated that the ⁶⁴Cu-labeled TPP cations, such as ⁶⁴Cu(DO3A-xy-TPP), are able to localize in the mitochondria of U87MG glioma cells. MicroPET imaging data also show that the tumor could be visualized clearly as early as 30 min p.i. (14). Considering both the tumor uptake selectivity, ⁶⁴Cu(DO3A-xy-TPP) has a significant advantage over ^99mTc-Sestamibi, the radiopharmaceutical currently available for both myocardial perfusion and tumor imaging (24-33, 39-45). However, the high liver uptake of ⁶⁴Cu(DO3A-xy-TPP) remains a significant challenge for its clinical applications as a new PET radiotracer for imaging tumors.

In this study, we used TPEP as the mitochondrion-targeting molecule. We are interested TPEP cation because of the phosphoryl group that may improve radiotracer excretion kinetics from the liver. As expected, the liver uptake of ⁶⁴Cu(DO3A-xy-TPEP) was significantly lower (p < 0.01) than that of ⁶⁴Cu(DO3A-xy-TPP) (Figure 3), even though ⁶⁴Cu(DO3A-xy-TPEP) (log P = -1.69 ± 0.11) is 10x more lipophilic than ⁶⁴Cu(DO3A-xy-TPP) (log P = -2.67 ± 0.21). The tumor uptake of ⁶⁴Cu(DO3A-xy-TPEP) is comparable to that of ⁶⁴Cu(DO3A-xy-TPP) at >30 min p.i. (Figure 2). As a result, tumor/liver ratios of ⁶⁴Cu(DO3A-xy-TPEP) are >3x better than those of ⁶⁴Cu(DO3A-xy-TPP). On the basis of these results, we believe that ⁶⁴Cu(DO3A-xy-TPEP) is a better PET radiotracer than ⁶⁴Cu(DO3A-xy-TPP) for imaging the MDR-negative tumors. ⁶⁴Cu(DO3A-xy-TPEP) also has significant advantages over ^99mTc-Sestamibi with respect to the tumor uptake and T/B (tumor/heart, tumor/liver, tumor/lung and tumor/muscle) ratios.

The exact localization mechanism remains unknown even though the results from in vitro assays show that ⁶⁴Cu(DO3A-xy-TPP) is able to localize in mitochondria of glioma cells (14). Many factors can influence biological properties of the ⁶⁴Cu-labeled TPEP cations. Previously, we found that replacing DO3A in ⁶⁴Cu(DO3A-xy-TPP) with DO2A leads to formation of ⁶⁴Cu(DO2A-xy-TPP)⁺, and results in a dramatic reduction in liver uptake without altering its tumor uptake. In this study, a similar reduction in liver uptake was seen for ⁶⁴Cu(DO2A-xy-TPEP)⁺ (Figure 3); but ⁶⁴Cu(DO2A-xy-TPEP)⁺ had a rapid tumor washout over the 2 h study period. As a result, its tumor/liver ratio is significantly lower (p < 0.01) than that of ⁶⁴Cu(DO3A-xy-TPEP).

⁶⁴Cu(NOTA-Bn-xy-TPEP) has the same molecular charge as ⁶⁴Cu(DO3A-xy-TPEP). They also share the almost identical lipophilicity with log P values being -1.69 ± 0.11 and -1.75 ± 0.03, respectively. However, the tumor uptake and T/B ratios of ⁶⁴Cu(NOTA-Bn-xy-TPEP) are significantly lower (p < 0.01) than those of ⁶⁴Cu(DO3A-xy-TPEP) (Figure 3). Similar adverse effect was also seen for the ⁶⁴Cu-labeled TPP analogs (16). Among six ⁶⁴Cu radiotracers evaluated in this study, ⁶⁴Cu(NOTA-Bn-xy-TPEP) has the lowest tumor uptake and poorest T/B ratios in non-cancerous organs, such as the heart, liver, lungs and muscle. Thus, it is reasonable to believe that the coordination geometry of the ⁶⁴Cu-DO3A chelates might play a significant role in tumor uptake and tumor retention of the ⁶⁴Cu-labeled TPP and TPEP cations.

The lipophilicity (log p = -1.92 ± 0.03) of ⁶⁴Cu(DOTA-Bn-xy-TPEP)^- is also close to that of ⁶⁴Cu(DO3A-xy-TPEP) (log p = -1.69 ± 0.11) despite of their difference in molecular charge. Their tumor uptake (Figure 3) is comparable within the experimental error. It seems that the negative charge in⁶⁴Cu(DOTA-Bn-xy-TPEP)^- does not affect its capability to localize in tumor cells. However, the uptake of ⁶⁴Cu(DOTA-Bn-xy-TPEP)^- in the heart, liver and muscle was significantly higher than that of ⁶⁴Cu(DO3A-xy-TPEP) at >30 min p.i. As a result, its T/B ratios was significantly lower than that ⁶⁴Cu(DO3A-xy-TPEP).

Linkers have a significant impact on tumor uptake and tumor/heart ratios of ⁶⁴Cu-labeled TPEP cations. For example, the extra acetamido group in ⁶⁴Cu(DOTA-xy-TPEP) results in a significantly lower liver uptake than that of ⁶⁴Cu(DO3A-xy-TPEP). However, the tumor/liver ratios of ⁶⁴Cu(DOTA-xy-TPEP) and ⁶⁴Cu(DO3A-xy-TPEP) compare well within the experimental error (Figure 4) due to a faster tumor washout of ⁶⁴Cu(DOTA-xy-TPEP). Substitution of xylene with PEG₂ leads to a significant reduction of the tumor and liver uptake. As a result, the tumor/liver ratios of ⁶⁴Cu(DO3A-PEG₂-TPEP) and ⁶⁴Cu(DO3A-xy-TPEP) are also comparable.

In addition, linkers also affect the metabolic stability of ⁶⁴Cu radiotracers. For example, ⁶⁴Cu(DO3A-xy-TPEP), ⁶⁴Cu(DOTA-Bn-xy-TPEP)^- and ⁶⁴Cu(NOTA-Bn-xy-TPEP) have no metabolism during their excretion via renal and hepatobiliary routes as evidenced by the lack of metabolite(s) in urine and feces samples. ⁶⁴Cu(DOTA-xy-TPEP) remains intact during excretion via the renal excretion; but only ∼15% of it remains intact in the feces sample (Figure 6). Since the only difference between ⁶⁴Cu(DO3A-xy-TPEP) and ⁶⁴Cu(DOTA-xy-TPEP) is the extra acetamido group, the metabolic instability of ⁶⁴Cu(DOTA-xy-TPEP) is most likely caused by the cleavage of CO-NH bond during hepatobiliary excretion.

The radioactivity detected in urine and feces samples represents only the portion excreted from the renal and hepatobiliary routes. The remaining activity is still “trapped” in organ tissues. We examined the stability of ⁶⁴Cu(DO3A-xy-TPEP) in the liver, and found that there was no intact ⁶⁴Cu(DO3A-xy-TPEP) in the liver homogenate (Figure SIII: top). While the identity of these four major radiometric peaks (Figure SIII: bottom) remains unknown, the size-exclusion HPLC chromatographic pattern of ⁶⁴Cu(DO3A-xy-TPEP) is very similar to that of the ⁶⁴Cu-labeled TETA-Octreotide in the rat liver homogenate (46). Based on the results from stability studies of ⁶⁴Cu complexes with tetraazamacrocycles (46-50), we believe that these radiometric peaks are likely caused by transchelation of ⁶⁴Cu to the proteins, such as superoxide dismutase (SOD) abundant in the liver and kidneys (51).

The metabolism of the ⁶⁴Cu-labeled biomolecules (antibodies and peptides) and ⁶⁴Cu complexes of tetraazamacrocycles has been investigated extensively (46-59). These studies clearly show that kinetic inertness of the ⁶⁴Cu chelate is particularly important for the in vivo stability of ⁶⁴Cu radiotracers, and their instability is often caused by transchelation of ⁶⁴Cu from the ⁶⁴Cu-BFC chelate to proteins (46, 51). For target-specific ⁶⁴Cu radiotracers, the tumor uptake is predominantly determined by receptor binding of targeting biomolecules (antibodies or peptides). The BFCs for ⁶⁴Cu chelation should be those which form the ⁶⁴Cu-BFC chelates with high kinetic inertness in order to minimize the liver radioactivity. For the ⁶⁴Cu-labeled TPEP cations, however, the BFC contributes greatly to the radiotracer tumor uptake and excretion kinetics from non-cancerous organs, such as liver and lungs. The optimal BFC should be those which result in the ⁶⁴Cu radiotracer with high tumor uptake and the best T/B, particularly tumor/heart and tumor/liver, ratios. From this point view, DO3A and DOTA are suitable BFCs for ⁶⁴Cu-labeling of TPEP cations. It must be noted that the radioactivity “trapped” inside the liver due to transchelation of ⁶⁴Cu is only a small portion of the injected radioactivity. Whenever possible, one has to consider the radioactivity excreted from renal and hepatobiliary routes (urine and feces samples), as well as the radioactivity “trapped” in various organ tissues.

CONCLUSIONS

In summary, we evaluated six new ⁶⁴Cu-labeled TPEP cations for their biodistribution properties and excretion kinetics using the athymic nude mice bearing U87MG human glioma xenografts. It was found that most the ⁶⁴Cu radiotracers described in this study have significant advantages over ^99mTc-Sestamibi with respect to their tumor/heart and tumor/muscle ratios. TPEP seems to be a better mitochondrion-targeting molecule than TPP because of the lower liver uptake and better tumor/liver ratios of ⁶⁴Cu(DO3A-xy-TPEP) than that of ⁶⁴Cu(DO3A-xy-TPP). BFCs and molecular charge also have significant impact on biological properties of ⁶⁴Cu-labeled TPEP cations. For example, replacing DO3A with DO2A results in ⁶⁴Cu(DO2A-xy-TPEP)⁺, which has a lower tumor uptake and T/B ratios than ⁶⁴Cu(DO3A-xy-TPEP). Substitution of DO3A with NOTA-Bn has a significant adverse effect on tumor uptake of the ⁶⁴Cu radiotracer. However, the use of DOTA-Bn to replace DO3A has minimal impact on the radiotracer tumor uptake; but the tumor/liver ratio of ⁶⁴Cu(DOTA-Bn-xy-TPEP)^- is not as good as that of ⁶⁴Cu(DO3A-xy-TPEP), probably due to the aromatic benzene ring in DOTA-Bn. Addition of an extra acetamido group in ⁶⁴Cu(DOTA-xy-TPEP) results in a lower liver uptake; but the tumor/liver ratios of ⁶⁴Cu(DOTA-xy-TPEP) and ⁶⁴Cu(DO3A-xy-TPEP) are well comparable, probably due to a faster tumor washout of ⁶⁴Cu(DOTA-xy-TPEP). Substitution of xylene with PEG₂ also leads to a significant reduction in both tumor and liver uptake. As a result, ⁶⁴Cu(DO3A-PEG₂-TPEP) and ⁶⁴Cu(DO3A-xy-TPEP) share very similar tumor/liver ratios. On the basis of both tumor uptake and T/B ratios of the ⁶⁴Cu radiotracers, we believe that both DO3A and DOTA are suitable BFCs for the ⁶⁴Cu-labeling of TPP cations. Among the six ⁶⁴Cu radiotracers evaluated in this tumor-bearing animal model, ⁶⁴Cu(DO3A-xy-TPEP) is of particularly interest due to its higher tumor uptake and better T/B ratios (particularly tumor/heart, tumor/liver and tumor/muscle). ⁶⁴Cu(DO3A-xy-TPEP) will be further evaluated for its potential as a new PET radiotracer to monitor the MDR transport function in tumors of different origin.

ABBREVIATIONS

DO2A

1,4,7,10-tetraazacyclododecane-1,7-diacetic acid

DO3A

1,4,7,10-tetraazacyclododecane-4,7,10-triacetic acid

DOTA

1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid

NOTA

1,4,7-triaazacyclononane-1,4,7-triacetic acid

DO2A-xy-TPP

(4-((4,10-bis(carboxymethyl)-1,4,7,10-tetraazacyclododecan-1-yl)methyl)benzyl)triphenylphosphonium

DO3A-xy-TPP

triphenyl(4-((4,7,10-tris(carboxymethyl)-1,4,7,10-tetraazacyclododecan-1-yl)methyl)benzyl)phosphonium

DO2A-xy-TPEP

(4-((4,10-bis(carboxymethyl)-1,4,7,10-tetraazacyclododecan-1-yl)methyl)benzyl)(2-(diphenylphosphoryl)ethyl)diphenylphosphonium

DO3A-xy-TPEP

(2-(diphenylphosphoryl)ethyl)diphenyl(4-((4,7,10-tris(carboxymethyl)-1,4,7,10-tetraazacyclododecan-1-yl)methyl)benzyl)phosphonium

DO3A-PEG₂-TPEP

(2-(diphenylphosphoryl)ethyl)diphenyl(4-((2-(4,7,10-tris(carboxymethyl)-1,4,7,10-tetraazacyclododecan-1-yl)acetamido)methyl)benzyl)phosphonium

DOTA-Bn-xy-TPEP

(2-(Diphenylphosphoryl)ethyl)diphenyl(4-((3-(4-((1,4,7,10-tetrakis(carboxymethyl)-1,4,7,10-tetraazacyclododecan-2-yl)methyl)phenyl)thioureido)methyl)benzyl)phosphonium

NOTA-Bn-xy-TPEP

(2-(Diphenylphosphoryl)ethyl)diphenyl(4-((3-(4-((1,4,7-tris(carboxymethyl)-1,4,7-triazonan-2-yl)methyl)phenyl)thioureido)methyl)benzyl)phosphonium