Evaluation of ⁶⁴Cu(DO3A-xy-TPEP) as A Potential PET Radiotracer for Monitoring Tumor Multidrug Resistance

Abstract

In this study, we evaluated the potential of ⁶⁴Cu(DO3A-xy-TPEP) (DO3A-xy-TPEP = (2-(diphenylphosphoryl)ethyl)diphenyl(4-((4,7,10-tris(carboxymethyl)-1,4,7,10-tetraazacyclododecan-1-yl)methyl)benzyl)phosphonium) as a PET (positron emission tomography) radiotracer for noninvasive monitoring of multidrug resistance (MDR) transport function in several xenografted tumor models (MDR-negative: U87MG; MDR-positive: MDA-MB-435, MDA-MB-231, KB-3-1 and KB-v-1). It was found that ⁶⁴Cu(DO3A-xy-TPEP) has a high initial tumor uptake (5.27 ± 1.2 %ID/g at 5 min p.i.) and show a steady uptake increase between 30 and 120 min p.i. (2.09 ± 0.53 and 3.35 ± 1.27 %ID/g at 30 and 120 min p.i., respectively) in the MDR-negative U87MG glioma tumors. ⁶⁴Cu(DO3A-xy-TPEP) has a greater uptake difference between U87MG glioma and MDR-positive tumors (MDA-MB-231: 1.57 ± 0.04, 1.00 ± 0.17, and 0.93 ± 0.15; MDA-MB-435: 1.15 ± 0.19, 1.12 ± 0.20, and 0.81 ± 0.11; KB-3-1: 1.45 ± 0.31, 1.43 ± 0.16, and 1.08 ± 0.19; and KB-v-1: 1.63 ± 0.47, 1.81 ± 0.31, and 1.14 ± 0.22 %ID/g at 30, 60 and 120 min p.i., respectively) than ^99mTc-Sestamibi. Regardless of the source of MDR, the overall net effect is the rapid efflux of ⁶⁴Cu(DO3A-xy-TPEP) from tumor cells, which leads to a significant reduction of its tumor uptake. It was concluded that ⁶⁴Cu(DO3A-xy-TPEP) is more efficient than ^99mTc-Sestamibi as the substrate for MDR P-glycoproteins (MDR Pgps) and multidrug resistance-associated proteins (MRPs), and might be a more efficient radiotracer for noninvasive monitoring of the tumor MDR transport function. ⁶⁴Cu(DO3A-xy-TPEP) and ^99mTc-Sestamibi share almost identical subcellular distribution patterns in U87MG glioma tumors. Thus, it is reasonable to believe that ⁶⁴Cu(DO3A-xy-TPEP), like ^99mTc-Sestamibi, is able to localize in mitochondria due to the increased plasma and mitochondrial transmembrane potentials in tumor cells.

INTRODUCTION

Cancer is the second leading cause of death world wide. The most prevalent forms of the disease are solid tumors of the lung, breast, prostate, colon and rectum. Although the exact cause of cancer remains unknown, most cancer patients will survive after surgery, radiation therapy, and chemotherapy or a combination thereof if it can be detected at the early stage. Therefore, accurate early detection is highly desirable so that various therapeutic regiments can be given before the tumors become widely spread.

Multidrug resistance (MDR) remains a major obstacle in the treatment of cancer. The over-expression of multidrug resistance P-glycoproteins (MDR Pgps) and multidrug resistance-associated proteins (MRPs) in tumor cells has been shown to confer MDR onto drug-sensitive cells (1–3). Both MDR Pgps and MRPs belong to a large family of ATP-trafficking proteins that mediate the transport of a large number of drug molecules (3–6). Although the molecular mechanism is not entirely clear, the net effect of MDR Pgp and MRP overexpression is the reduced intracellular drug accumulation through an energy-dependent drug efflux in the MDR-positive cancer cells as compared with the drug-sensitive tumor cells (5, 6). While several different genes are associated with the MDR phenotype (6, 7), the MDR mediated by overexpression of MDR1 Pgp is one of the best characterized barriers to chemotherapeutics treatment. High levels of MDR1 Pgp expression has been detected in more than 50% of tumors (8–11). A correlation between the MDR1 Pgp expression and failure of chemotherapy or poor survival rates has been established (11). However, the lack of MDR1 Pgp in other MDR-positive tumors indicates that additional cellular changes also confer resistance to anticancer drugs (12–15). For example, MRP1 expression in retinoblastoma correlated well with the rare failure of chemotherapy (16). The lung resistance protein (LRP) and breast cancer resistance protein (BCRP) have also been observed in tumor cells (17, 18). More recently, the overexpression of a 40 kDa protein (P-40 or Annexin I) was reported in tumor cell lines w/o MDR1 Pgp (19, 20). Since MDR is a major obstacle in chemotherapy of cancer patients, there has been a continuing demand for radiotracers that are able to monitor the tumor MDR transport function in a noninvasive fashion (21–27).

Cationic radiotracers, such as ^99mTc-Sestamibi and ^99mTc-Tetrofosmin, tend to localize in tumor cells due to the increased negative mitochondrial potentials (28–35). ^99mTc-Sestamibi and ^99mTc-Tetrofosmin have been clinically used for both cancer early detection and non-invasive monitoring of the tumor MDR transport function by single photon emission computed tomography (SPECT) (21–27, 30). However, their diagnostic and prognostic values are often limited due to their insufficient tumor localization and high uptake in the heart, liver and muscle, which makes it very difficult to detect small lesions in the chest and abdominal regions. Thus, there is an unmet medical need for the radiotracers that are able to monitor noninvasively the MDR transport function in tumors.

Previously, we reported ⁶⁴Cu-labeled organic phosphonium cations, such as ⁶⁴Cu(DO3A-xy-TPEP) (Figure 1: DO3A-xy-TPEP = (2-(diphenylphosphoryl)ethyl)diphenyl(4-((4,7,10-tris(carboxymethyl)-1,4,7,10-tetraazacyclododecan-1-yl)methyl)benzyl)phosphonium), as a potential PET (positron emission tomography) radiotracer in the athymic nude mice bearing U87MG human glioma xenografts (36–39). Both biodistribution and in vivo microPET imaging studies demonstrated that ⁶⁴Cu(DO3A-xy-TPEP) has higher tumor uptake with better tumor selectivity (as defined by their tumor/heart and tumor/muscle ratios) than ^99mTc-Sestamibi (Figure 1). These promising results lead us to explore the potential of ⁶⁴Cu(DO3A-xy-TPEP) as a PET radiotracer for noninvasive monitoring of the MDR transport function in athymic nude mice bearing MDR-positive tumor xenografts.

Figure 1.

Structures of ^99mTc-Sestamibi and ⁶⁴Cu(DO3A-xy-TPP).

The tumor cell lines used in this study include U87MG human glioma, MDA-MB-231 and MDA-MB-435 human breast cancer, KB-3-1 and KB-v-1. The U87MG glioma cell line was chosen because it has no over-expression of MDR1 Pgp (40–42). The U87MG cell line has been used for evaluation of the cellular uptake kinetics of ^99mTc-Sestamibi and ^99mTc-Tetrofosmin (43). MDA-MB-435 is the estrogen-independent human breast cancer cell line (metastatic ductal adenocarcinoma), and has a low level of MDR1 expression (44–47) and a high level expression of MRP2 and MRP4 (47, 48). MDA-MB-231 is also an estrogen-independent human breast cancer cell line, and is often considered “drug-sensitive” because it has little MDR1, MRP1 and BCRP overexpression (47, 48). However, recent studies have clearly demonstrated that the MDA-MB-231 breast tumor cell line has a high level expression of P-40, a 40 kDa protein (19, 20). The athymic nude mice bearing KB-3-1 and KB-v-1 tumor xenografts have been successfully used to evaluate cationic ^99mTc and ⁶⁸Ga radiotracers (49–56), as well as luminescent probes (57). High level expression of MDR Pgps or/and MRPs will result in a rapid efflux of the cationic radiotracer from tumor cells and the significant reduction in tumor localization of the radiotracer. Since it has been used for imaging the tumor MDR transport function by SPECT (21–27), ^99mTc-Sestamibi was as the “control”. We are interested in ⁶⁴Cu(DO3A-xy-TPEP) due its high tumor uptake and rapid liver clearance in nude mice bearing the MDR-negative U87MG human glioma xenografts (39). The main objective of this study is to evaluate the potential of ⁶⁴Cu(DO3A-xy-TPEP) as a new PET radiotracer for noninvasive monitoring of MDR transport function in several xenografted tumor models. It is important to note that MDR is often mediated by a combination of MDR Pgps (particularly MDR1 Pgp) and MRPs. In addition, MDR Pgps and MRPs are also overexpressed in several normal organs involved in excretory functions, including kidneys and liver (58–60).

EXPERIMENTAL SECTION

Materials

Chemicals were purchase from Sigma/Aldrich (St. Louis, MO). ⁶⁴Cu was produced using a CS-15 biomedical cyclotron at Washington University School of Medicine by the ⁶⁴Ni(p,n)⁶⁴Cu nuclear reaction. Na^99mTcO₄ was obtained from a commercial DuPont Pharma ⁹⁹Mo/^99mTc generator. Cardiolite® vials were obtained as a gift from Bristol Myers Squibb Medical Imaging (N. Billerica, MA), and were reconstituted according to the manufacturer’s package insert.

Methods

Radio-HPLC method used the LabAlliance HPLC system equipped with a UV/vis detector (λ = 254 nm), a β-ram IN-US detector and Zorbax C₁₈ analytical 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.

Doses Preparation

⁶⁴Cu(DO3A-xy-TPEP) was prepared according to the procedure described in our previous report (39), and was purified by HPLC. Volatiles in the HPLC mobile phases were evaporated under the reduced pressure at room temperature. Doses were prepared by dissolving the residue to 20 – 30 µCi/mL in saline. The resulting solution was filtered with a 0.20 micron filter unit before being injected into animals. The injection volume for each animal was 0.1 mL.

Animal Models

Biodistribution studies were performed 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 the Purdue University Animal Care and Use Committee (PACUC). Female athymic nu/nu mice (5 – 6 weeks) were purchased from Harlan (Charles River, MA). The mice were implanted with 5 × 10⁶ tumor cells into the mammary fat pad (MDA-MB-231 and MDA-MB-435 breast tumor cells) or shoulder flank (U87MG glioma, KB-3-1 and KB-v-1). U87MG glioma cells were cultured in the Minimum Essential Medium, Eagle with Earle’s Balanced Salt Solution (non-essential amino acids sodium pyruvate) (ATCC, Manassas, VA). The MDA-MB-435 and MDA-MB-231 breast cancer cells were grown in the RPMI Medium 1640 with L-Glutamine (GIBCO, Grand Island, NY). The DMEM medium with 4.5g/L D-Glucose, L-Glutamine and 110 mg/L sodium pyruvate (GIBCO, Grand Island, NY) was used for KB-v-1 and KB-3-1 tumor cells. In all cases, 10% FBS (Sigma, St. Louis, MO) and 5% Penicillin Streptomycin (GIBCO, Grand Island, NY) were added into the medium just before use. All cell lines were grown in humidified atmosphere of 5% carbon dioxide. All procedures will be performed in a laminar flow cabinet using the aseptic technique. Two to four weeks after inoculation, the tumor size was in the range of 0.1 – 0.5 g, and animals were used for biodistribution studies.

Biodistribution Protocol

Twelve tumor-bearing mice (20 – 25 g) were divided into four groups. Each animal was administered with ⁶⁴Cu(DO3A-xy-TPEP) (2 – 3 µCi) via tail vein. Three animals were euthanized by sodium pentobarbital overdose (100 mg/kg) at 5, 30, 60, and 120 min postinjection (p.i.). Blood samples were withdrawn from the heart. The tumor and normal organs (brain, eyes, heart, intestine, kidneys, liver, lungs, muscle and spleen,) were excised, washed with saline, dried with absorbent tissue, weighed, and counted on a Perkin Elmer Wizard – 1480 γ-counter (Shelton, CT). The organ uptake was calculated as a percentage of the injected dose per organ (%ID/organ). The tumor uptake values are reported as an average ± standard deviation based on the results from three tumorbearing mice, unless specified, at each time point. The normal organ uptake values are reported as an average ± standard deviation based on the results from fifteen tumor-bearing mice (3 animals for each tumor type, and a total of five different tumor types) at each time point, unless specified. For comparison purposes, ^99mTc-Sestamibi was evaluated using the same protocol. Comparison between ⁶⁴Cu(DO3Axy-TPEP) and ^99mTc-Sestamibi was made using the one-way ANOVA test (GraphPad Prim 5.0, San Diego, CA). The level of significance was set at p < 0.05.

Blocking Experiment with Cyclosporin-A (Cys-A)

In this experiment, each tumor-bearing mice (n = 4) was treated with intraperitoneal injection of 0.1 mL DMSO solution containing 0.35 mg Cys-A (14 mg/kg) at 30 min before administration of radiotracer. In the “control” group (n = 4), each tumorbearing mouse was treated with the same volume of DMSO solution without Cys-A. Four animals were euthanized by sodium pentobarbital overdose (100 – 200 mg/kg) at 30 min p.i. The ex-vivo biodistribution was performed according to the procedure described above. The organ of uptake was calculated as a percentage of the injected dose per gram (%ID/g).

Subcelluar Distribution Characteristics

Subcellular distribution studies were performed using the athymic nude mice bearing U87MG glioma (n = 3) and MDA-MB-435 breast cancer (n = 3) xenografts according to literature method (61–66). Each tumor-bearing mouse was administered with ~30 µCi of ⁶⁴Cu(DO3A-xy-TPEP) via tail vein. Animals were sacrificed by intraperitoneal administration of sodium pentobarbital overdose (100 – 200 mg/kg) at 1 h p.i. The tumor was rapidly removed, and placed immediately in ice-cold (4 °C) Buffer I, containing 230 mM mannitol, 70 mM sucrose, 3 mM 3-(N-morpholino)propanesulfonic acid (MOPS), 2 mM ethylene glycol tetraacetic acid (EGTA), and 0.2 % bovine serum albumin (BSA) (w/v) with pH 7.2. The tumor tissue was then minced into 1 – 2 mm pieces, and homogenized using a tissue grinder (Wheaton, NJ). The resulting mixture was centrifuged at 500xg for 10 min at 4 °C. The supernatant (blood, extracellular components and debris) was discarded and pellet was re-suspended in 2 mL of ice-cold (4 °C) Buffer II (containing 110 mM KCl, 20 mM MOPS, 46 mM mannitol, 14 mM sucrose, 2 mM EGTA, pH 7.5). To the mixture was added 0.5 mg/mL of Trypsin protease (Gibco, Ground Island, NY). After 3 min homogenization, the protease reaction was terminated by adding 4 mL of Buffer II (4 °C) containing 0.2 % BSA. The homogenate was centrifuged at 5000xg for 5 min at 4 °C. Supernatant I was collected, and the pellet was re-suspended in 5 mL of ice-cold (4 °C) Buffer II with 0.2% BSA. After centrifugation at 500xg for 10 min (4 °C), pellet I (containing cellular fragments) was obtained. The supernatant was transferred and centrifuged at 3000xg for 10 min (4 °C) to obtain the membrane fragment (supernatant II) and mitochondrial component (pellet II), respectively. Supernatant I, supernatant II, pellet I and pellet II were counted by a Perkin-Elmer Wizard – 1480 automatic γ-counter (Shelton, CT). The percentage of radioactivity was calculated on the basis of the radioactivity counts in each subcellular component over the total radioactivity count before homogenization. The percentage of radioactivity recovery was calculated by the total radioactivity counts in supernatant I, supernatant II, pellet I and pellet II over the total radioactivity count before homogenization. All data were reported as an average of three independent measurements. ^99mTc-Sestamibi was evaluated using the same protocol for comparison purposes.

Western Blot Experiment

Western blot experiments were performed to demonstrate the presence or absence of MDR1 Pgp, MRP2 and MRP4 in different tumor tissues (U87MG glioma, MDA-MB-231, MDA-MB-435, KB-3-1 and KB-v-1). The tumor tissues (U87MG, MDA-MB-231, MDA-MB-435, KB-3-1 and KB-v-1) were harvested directly from the tumor-bearing nude mice. After homogenized, the tumor tissues (10 – 20 mg) were incubated with 500 µL of cell lysis buffer (Cell Signaling Technology, Danvers, MA) on ice for 5min. The homogenates were centrifuged at 14,000xg for 10 min. The supernatants were removed for further use. For tumor cells, MDA-MB-231 and MDA-MB-435, medium was removed at first. After washed with PBS, cells were incubated with cell lysis buffer (Cell Signaling Technology, Danvers, MA) on ice for 5 min. The tumor cells were scraped and the mixture was sonicated briefly. The homogenates were centrifuged at 14,000xg for 10 min. The supernatant was removed for further use. The protein content was determined by using the BCA™ Protein Assay Kit (Pierce, Rockford, IL). The protein (15 µg per lane) was subjected to electrophoresis on SDS– polyacrylamide gel at 160V for 60 min and transferred electrophoretically on polyvinylidene difluoride membranes at 70 mA for 120 min. The blots were blocked with TBS-T containing 5% non-fat dry milk powder for 1 h. Then, the blotting membranes were incubated with 1 mL of the primary antibody (1:200) against MDR1 (JSB-1, mouse IgG₁ supernatant, Santa Cruz Biotechnology, CA), MRP2 (M2II-12, mouse IgG2a supernatant, Santa Cruz Biotechnology, CA) or MRP4 (M4I-10, rat IgG2a supernatant, Santa Cruz Biotechnology, CA) separately over night. Primary antibody solution was removed and the membranes were incubated for 1 h with secondary antibody (goat anti-mouse IgG-HRP, Santa Cruz Biotechnology, CA) diluted with 1:5000. Finally, membranes were developed using the ECL Western Blotting Substrate (Pierce, Rockford, IL) according to the manufacturer's instructions. The positive control for MDR1 was Human PGP Membrane from BD Gentest (San Jose, CA). Human MRP2 Membrane (BD Gentest, San Jose, CA) was used as the positive control of MRP2.

RESULTS

Biodistribution Characteristics

The athymic nude mice bearing U87MG, MDA-MB-231, MDA-MB-435, KB-3-1 and KB-v-1 tumor xenografts were used to evaluate tumor uptake and biodistribution characteristics of ⁶⁴Cu(DO3A-xy-TPEP). For comparison purpose, ^99mTc-Sestamibi was also evaluated in the same animal model since it has been clinically used for early cancer detection and for imaging MDR transport function in cancer patients (21–27, 30). Biodistribution data for ⁶⁴Cu(DO3A-xy-TPEP) and ^99mTc-Sestamibi are summarized in Table 1 and Table 2, respectively. Figure 2 compares the tumor uptake (%ID/g) of ⁶⁴Cu(DO3A-xy-TPEP) and ^99mTc-Sestamibi. Figure 3 illustrates the tumor uptake differences between U87MG glioma and the MDR-positive tumors (MDA-MB-231, MDA-MB-435, KB-3-1 and KB-v-1) for ⁶⁴Cu(DO3A-xy-TPEP) and ^99mTc-Sestamibi.

Table 1.

Biodistribution data for ⁶⁴Cu(DO3A-xy-TPEP) in xenografted tumor-bearing athymic nude mice. The tumor uptake is reported as an average ± standard deviation based on the results from three tumor-bearing mice, unless specified, at each time point. The normal organ uptake values are reported as an average ± standard deviation based on the results from fifteen tumor-bearing mice (3 animals for each tumor type, and a total of five different tumor types) at each time point.

Organ	5 min	30 min	60 min	120 min
Blood	9.52 ± 1.19 (n = 9)	2.09 ± 0.90 (n = 15)	1.30 ± 1.10 (n = 15)	0.84 ± 0.62 (n = 15)
Brain	0.22 ± 0.06 (n = 9)	0.10 ± 0.09 (n = 15)	0.06 ± 0.05 (n = 15)	0.02 ± 0.02 (n = 15)
Heart	4.04 ± 0.70 (n = 9)	1.25 ± 0.48 (n = 15)	0.92 ± 0.43 (n = 15)	0.55 ± 0.09 (n = 15)
Intestine	4.06 ± 0.78 (n = 9)	3.23 ± 1.22 (n = 15)	2.87 ± 1.28 (n = 15)	2.61 ± 1.26 (n = 15)
Kidney	31.79 ± 8.02 (n = 9)	6.77 ± 1.63 (n = 15)	4.11 ± 1.11 (n = 15)	2.98 ± 0.53 (n = 15)
Liver	8.03 ± 1.33 (n = 9)	8.74 ± 3.30 (n = 15)	8.66 ± 4.85 (n = 15)	5.59 ± 1.39 (n = 15)
Lungs	7.72 ± 0.74 (n = 9)	2.50 ± 0.61 (n = 15)	2.19 ± 0.91 (n = 15)	1.51 ± 0.18 (n = 15)
Muscle	3.07 ± 0.45 (n = 9)	0.82 ± 0.39 (n = 15)	0.81 ± 0.98 (n = 15)	0.10 ± 0.06 (n = 15)
Spleen	2.73 ± 0.31 (n = 9)	1.14 ± 0.37 (n = 15)	0.84 ± 0.40 (n = 15)	0.56 ± 0.17 (n = 15)
U87MG	5.27 ± 0.99 (n = 3)	2.09 ± 0.53 (n = 3)	2.58 ± 0.31 (n = 3)	3.35 ± 1.27 (n = 3)
MDA-MB-435	2.73 ± 0.55 (n = 3)	1.05 ± 0.19 (n = 3)	1.12 ± 0.20 (n = 3)	0.81 ± 0.11 (n = 3)
MDA-MB-231	3.54 ± 0.15 (n = 3)	1.57 ± 0.04 (n = 3)	1.10 ± 0.17 (n = 3)	0.93 ± 0.15 (n = 3)
KB-v-1		2.00 ± 0.48 (n = 6)	1.58 ± 0.47 (n = 6)	0.93 ± 0.22 (n = 3)
KB-3-1		1.97 ± 0.61 (n = 6)	1.46 ± 0.26 (n = 6)	1.16 ± 0.19 (n = 3)

Table 2.

Biodistribution data for ^99mTc-Sestamibi in xenografted tumor-bearing athymic nude mice. The tumor uptake is reported as an average ± standard deviation based on the results from three tumorbearing mice at each time point, unless specified. The normal organ uptake values are reported as an average ± standard deviation based on the results from nine tumor-bearing mice (3 animals for each tumor type, and a total of three different tumor types) at each time point.

Organ	5 min	30 min	60 min	120 min
Blood	0.87 ± 0.44 (n = 6)	0.26 ± 0.28 (n = 9)	0.28 ± 0.34 (n = 9)	0.08 ± 0.03 (n = 9)
Brain	0.22 ± 0.05 (n = 6)	0.15 ± 0.05 (n = 9)	0.11 ± 0.02 (n = 9)	0.08 ± 0.02 (n = 9)
Heart	19.22 ± 6.96 (n = 6)	20.52 ± 1.11 (n = 9)	20.07 ± 3.99 (n = 9)	20.88 ± 4.55 (n = 9)
Intestine	9.80 ± 2.74 (n = 6)	11.13 ± 1.62 (n = 9)	11.11 ± 4.54 (n = 9)	5.29 ± 2.52 (n = 9)
Kidney	37.45 ± 12.0 (n = 6)	25.50 ± 5.56 (n = 9)	20.24 ± 4.59 (n = 9)	12.47 ± 3.90 (n = 9)
Liver	14.37 ± 0.92 (n = 6)	9.82 ± 2.12 (n = 9)	8.37 ± 1.90 (n = 9)	5.40 ± 0.42 (n = 9)
Lungs	6.20 ± 2.36 (n = 6)	3.03 ± 1.04 (n = 9)	2.17 ± 0.80 (n = 9)	1.72 ± 0.47 (n = 9)
Muscle	4.84 ± 1.12 (n = 6)	5.02 ± 1.01 (n = 9)	5.25 ± 1.67 (n = 9)	5.22 ± 1.22 (n = 9)
Spleen	3.21 ± 1.27 (n = 6)	1.77 ± 0.554 (n = 9)	1.32 ± 0.33 (n = 9)	1.32 ± 0.90 (n = 9)
U87MG	3.06 ± 0.94 (n = 3)	2.11 ± 0.38 (n = 3)	1.47 ± 0.54 (n = 3)	1.55 ± 0.36 (n = 3)
MDA-MB-435	1.18 ± 0.17 (n = 3)	1.09 ± 0.20 (n = 3)	0.96 ± 0.42 (n = 3)	0.94 ± 0.21 (n = 3)
KB-v-1		0.92 ± 0.61 (n = 3)	1.05 ± 0.86 (n = 3)	0.95 ± 0.47 (n = 3)
KB-3-1		0.53 ± 0.09 (n = 3)	0.31 ± 0.09 (n = 3)	0.22 ± 0.03 (n = 3)

Figure 2.

Comparison of tumor uptake for ⁶⁴Cu(DO3A-xy-TPEP) (top) and ^99mTc-Sestamibi (bottom) in athymic nude mice bearing different tumor xenografts (U87MG glioma, MDA-MB-231 and MDA-MB-435 breast cancer, KB-3-1 and KB-v-1).

Figure 3.

Tumor uptake difference between the MDR-negative U87MG glioma and the MDR-positive tumors (MDA-MB-231, MDA-MB-435, KB-3-1 and KB-v-1) for ⁶⁴Cu(DO3A-xy-TPEP) (top) and ^99mTc-Sestamibi (bottom).

In general, the biodistribution of ⁶⁴Cu(DO3A-xy-TPEP) in non-cancerous organs of the MDRpositive tumor-bearing mice was almost identical to that obtained from the mice bearing U87MG human glioma xenografts. There was no size-dependence for the tumor uptake of ⁶⁴Cu(DO3A-xy-TPEP) and ^99mTc-Sestamibi. However, ⁶⁴Cu(DO3A-xy-TPEP) and ^99mTc-Sestamibi showed significant differences in their uptake in both tumors and normal organs. For example, ⁶⁴Cu(DO3A-xy-TPEP) had a rapid tumor washout in the MDR-negative glioma model between 5 min (5.27 ± 1.21 %ID/g) and 30 min (2.09 ± 0.53 %ID/g) p.i. due to its fast blood clearance, and a steady tumor uptake increase between 30 min (2.09 ± 0.53 %ID/g) and 120 min (3.35 ± 1.27 %ID/g) p.i. probably due to its slow kinetics to penetrate the cellular and mitochondrial transmembranes (39). In the MDR-positive tumor-bearing mice, ⁶⁴Cu(DO3A-xy-TPEP) showed a rapid tumor washout between 5 and 30 min p.i. (Figure 3: top), and its tumor uptake was relatively unchanged (ranging from 1.0 %ID/g to 1.5 %ID/g depending on the tumor type) in between 30 and 120 min p.i. The tumor uptake of ^99mTc-Sestamibi decreased steadily in all four tumor-bearing animal models over the 2 h period (Figure 3: bottom). The most striking difference between ⁶⁴Cu(DO3A-xy-TPEP) (Table 1) and ^99mTc-Sestamibi (Table 2) was their uptake in heart and muscle. For example, the heart uptake was <1% ID/g for ⁶⁴Cu(DO3A-xy-TPEP) at >30 min p.i., whereas ^99mTc-Sestamibi had the heart uptake of 19.22 ± 7.62 % ID/g at 5 min p.i. and 19.19 ± 5.32 %ID/g at 120 min p.i. The muscle uptake of ⁶⁴Cu(DO3A-xy-TPEP) was undetectable at >30 min p.i. In contrast, ^99mTc-Sestamibi had a high muscle uptake over the 2 h study period (4.84 ± 1.22 and 5.45 ± 1.24 %ID/g at 5 and 120 min p.i., respectively). Another significant difference was their tumor uptake in the MDR-negative U87MG glioma and the MDR-positive tumor xenografts. For example, ⁶⁴Cu(DO3A-xy-TPEP) had the tumor uptake of 5.27 ± 0.99, 2.09 ± 0.53, 2.58 ± 0.31, and 3.35 ± 1.27 %ID/g at 5, 30, 60 and 120 min p.i., respectively, in U87MG glioma. Its tumor uptake values in the MDR-positive tumors were significantly lower than that in glioma (Figure 2). The glioma uptake values of ^99mTc-Sestamibi were 3.06 ± 0.94, 2.11 ± 0.38, 1.47 ± 0.54, and 1.55 ± 0.36 %ID/g at 5, 30, 60 and 120 min p.i., respectively, and its uptake values in the MDA-MB-435 breast tumor were 1.18 ± 0.17, 1.09 ± 0.20, 0.96 ± 0.42, and 0.94 ± 0.21%ID/g at 5, 30, 60 and 120 min p.i., respectively.

Western Blot Experiments

Western blot experiments were performed to demonstrate the presence or absence of MDR1 Pgp, MRP2 and MRP4 in the U87MG, MDA-MB-231, MDA-MB-435, KB-3-1 and KB-v-1 tumor xenografts. Tumor cells were directly isolated from the tumor tissues. The total protein concentration was determined to be 4.5 ± 0.4, 4.0 ± 0.2, 3.3 ± 0.2, 4.3 ± 0.3 and 4.1 ± 0.2 mg/mL for the xenografted MDA-MB-435, MDA-MB-231, U87MG, KB-v-1 and KB-3-1 tumors, respectively. Figure 4 illustrates the Western blotting data for the U87MG, MDA-MB-231, MDA-MB-435, KB-3-1 and KB-v-1 tumor xenografts. The positive control was the Human PGP Membrane for MDR1 Pgp. The positive control for MPR2 was Human MRP2 Membrane. No positive control is commercially available for MRP4. The blotting data clearly demonstrated that there was very little or no expression of MDR1 Pgp, MRP2 and MRP4 in the U87MG glioma xenografts. However, there is a high expression of MDR1 Pgp, MRP2 and MRP4 in the MDA-MB-231, MDA-MB-435, KB-3-1 and KB-v-1 tumors.

Figure 4.

Western blot data to demonstrate the presence/absence of MDR1 (A and B), MRP2 (C) and MRP4 (D) in U87MG (MDR1, MRP2 and MRP4-negative), MDA-MB-435 (MDR1-negative in tumor cells but MDR1-positive in tumor tissue, MRP2 and MRP4-positive), MDA-MB-231 (MDR1-negative in tumor cells but MDR1-positive in tumor tissue, MRP2 and MRP4-positive), KB-v-1 and KB-3-1 (MDR1, MRP2 and MRP4-positive).

It is important to note that MDA-MB-435 and MDA-MB-231 are often known “drug sensitive” or MDR1-negative breast cancer cell lines (47, 48). The results from Western blotting studies also showed that the MDA-MB-435 and MDA-MB-231 human breast cancer cell lines have little expression of MDR1 Pgp (Figure 4B). However, the MDA-MB-435 and MDA-MB-231 breast cancer cells isolated from the xenografted MDA-MB-435 and MDA-MB-231 breast tumors are MDR1-positive (Figure 4B). They also show a significant expression of MRP2 and MRP4.

Impact of Cys-A on Tumor Uptake and Radiotracer Liver Clearance Kinetics

To further demonstrate that the tumor washout is caused by the tumor MDR transport function, we performed blocking experiments using ⁶⁴Cu(DO3A-xy-TPEP) and ^99mTc-Sestamibi as radiotracers in the athymic nude mice bearing KB-3-1 and KB-v-1 tumor xenografts. Cys-A, a well-known wide-spectrum MDR modulator (67), was injected as the blocking agent 1 h prior administration of ⁶⁴Cu(DO3A-xy-TPEP) or ^99mTc-Sestamibi. The ex-vivo biodistribution was performed at 30 min p.i. Figure 5 compares uptake of ⁶⁴Cu(DO3A-xy-TPEP) and ^99mTc-Sestamibi in the tumor and selected normal organs in the absence/presence of Cys-A. As expected, pre-treatment of the tumor-bearing mice with excess Cys-A resulted in a significant increase in tumor uptake of both ⁶⁴Cu(DO3A-xy-TPEP) (Figure 5: bottom) and ^99mTc-Sestamibi (Figure 5: top). The tumor uptake difference was 0.83 %ID/g in KB-3-1 and 1.70 KBv-1 %ID/g w/o Cys-A for ⁶⁴Cu(DO3A-xy-TPEP) while there was no significant difference for ^99mTc-Sestamibi within the experimental error, suggesting that ⁶⁴Cu(DO3A-xy-TPEP) might be a more efficient radiotracer for noninvasive monitoring of the MDR transport function.

Figure 5.

Impact of Cys-A on the uptake of ⁶⁴Cu(DO3A-xy-TPEP) and ^99mTc-Sestamibi in selected organs of the athymic nude mice bearing KB-v-1 and KB-3-1 tumor xenografts.

In addition, the uptake of ⁶⁴Cu(DO3A-xy-TPEP) and ^99mTc-Sestamibi in kidneys, liver and lungs, was also significantly increased, and the radioactivity excretion was delayed in the presence of Cys-A. For example, the liver uptake was 8.03 ± 1.33 %ID/g for ⁶⁴Cu(DO3A-xy-TPEP) and 9.82 ± 2.12 %ID/g for ^99mTc-Sestamibi in the absence of Cys-A at 30 min p.i. In presence of Cys-A, however, the liver uptake values increased to 17.18 ± 2.47 %ID/g for ⁶⁴Cu(DO3A-xy-TPEP) and 23.68 ± 5.08 %ID/g for ^99mTc-Sestamibi at the same time point. Similar increase (Figure 5) was also observed in the heart, kidneys, lungs and spleen. The increase of muscle uptake of ⁶⁴Cu(DO3A-xy-TPEP) is probably caused by its high blood radioactivity accumulation in the presence of excess Cys-A. The impact of Cys-A on the excretion kinetics of ⁶⁴Cu(DO3A-xy-TPEP) was more significant than that of ^99mTc-Sestamibi.

Subcellular Distribution Characteristics

We performed the subcellular distribution studies on ⁶⁴Cu(DO3A-xy-TPEP) and ^99mTc-Sestamibi using the standard differential centrifugation method (Figure 6) according to the literature (61–65). The main objective of these studies is to elucidate the tumor localization mechanism of ⁶⁴Cu(DO3A-xy-TPEP). Figure 7 illustrates the subcellular distribution characteristics of ⁶⁴Cu(DO3A-xy-TPEP) and ^99mTc-Sestamibi in the xenografted U87MG glioma and MDA-MB-435 tumors. The radioactivity recovery was >95% for both ⁶⁴Cu(DO3A-xy-TPEP) and ^99mTc-Sestamibi. After cellular fractionation, majority of the radioactivity was found in supernatants I (66.5 ± 12.5% for ⁶⁴Cu(DO3A-xy-TPEP) and 63.5 ± 4.7% for ^99mTc-Sestamibi) and II (8.5 ± 0.5% for ⁶⁴Cu(DO3A-xy-TPEP) and 14.5 ± 2.6% for ^99mTc-Sestamibi). Only a small portion of the radioactivity was found in the mitochondrial fraction (Figure 7: top) for ⁶⁴Cu(DO3A-xy-TPEP) (2.25 ± 0.21%) and ^99mTc-Sestamibi (3.33 ± 1.50%). It is clear that ⁶⁴Cu(DO3A-xy-TPEP) and ^99mTc-Sestamibi share almost identical subcellular distribution patterns in the U87MG glioma tumors. In addition, the subcellular distribution characteristics of ⁶⁴Cu(DO3A-xy-TPEP) are very similar in the xenografted U87MG glioma and MDA-MB-435 breast tumors (Figure 6: bottom).

Figure 6.

Schematic presentation of the standard differential centrifugation process.

Figure 7.

Top: subcellular distribution properties of ⁶⁴Cu(DO3A-xy-TPEP) and ^99mTc-Sestamibi in U87MG human glioma xenografts; Bottom: direct comparison of subcellular distribution patterns of ⁶⁴Cu(DO3A-xy-TPEP) in the U87MG glioma and MDA-MB-435 breast tumor xenografts.

DISCUSSION

During last two decades, cationic radiotracers, such as ^99mTc-Sestamibi, have been widely used for early cancer detection and for noninvasive monitoring of the tumor MDR transport function by SPECT (21–35). The overwhelming emphasis of the prior literature has been focused on the tumor MDR transport function mediated by MDR1 Pgp most likely due to the fact that MDR1 Pgp is one of the best biologically characterized barriers to chemotherapy. It is very important to emphasize that the tumor MDR transport function can be mediated by MDR Pgps and MRPs (1–11), as well as other drug transporting proteins, such as LRP (17) and BCRP (18). Regardless of the source of MDR, the overall net effect is the rapid efflux of cationic radiotracers from tumor cells, which often leads to a significant reduction of the radiotracer tumor uptake. There is an inverse relationship between the radiotracer tumor uptake and the MDR transport function. If there is no MDR Pgps and/or MRPs in tumor cells, cationic radiotracers will have relatively high tumor uptake with no significant tumor washout. If there is a high level expression of MDR Pgps and/or MRPs, cationic radiotracers are expected to show a rapid tumor radioactivity washout with low tumor uptake. If the radiotracer is efficient as a substrate for MDR Pgps and MRPs, there would be less radiotracer uptake in the MDR-positive tumors. The ideal radiotracers useful for noninvasive monitoring of the tumor MDR transport function would be those which provide the largest uptake difference between the MDR-negative and MDR-positive tumors, and have low uptake in non-cancerous organs, such as the heart, liver, lungs and muscle.

Since the uptake of ⁶⁴Cu(DO3A-xy-TPEP) in normal organs is almost identical in different xenografted tumor-bearing models, their tumor uptake difference should, in theory, reflect the MDR transport function in the MDR-negative vs. MDR-positive tumors. For example, the U87MG glioma tumors have no significant expression of MDR1 Pgp and MRPs, as demonstrated by Western blot experiment (Figure 4). As a result, ⁶⁴Cu(DO3A-xy-TPEP) has a high initial tumor uptake, and show a steady uptake increase between 30 and 120 min p.i. (Figure 2: top). Similar results are obtained for ^99mTc-Sestamibi (Figure 2: bottom). The xenografted MDA-MB-435, MDA-MB-231, KB-3-1 and KBv-1 tumors all have high levels of MDR1 Pgp, MRP2 and MRP4 expression (Figure 4A). As a result, both ⁶⁴Cu(DO3A-xy-TPEP) and ^99mTc-Sestamibi show a significant washout from tumor over the 2 h period. Since ⁶⁴Cu(DO3A-xy-TPEP) has a greater tumor uptake difference between U87MG glioma and MDA-MB-231, MDA-MB-435, KB-3-1 and KB-v-1 than ^99mTc-Sestamibi, it is reasonable to believe that ⁶⁴Cu(DO3A-xy-TPEP) is more efficient than ^99mTc-Sestamibi as the substrate for MDR Pgps and MRPs, and might be more efficient for noninvasive monitoring of the tumor MDR transport function. This conclusion is supported by the blocking experiments, which show a significant tumor uptake increase for ⁶⁴Cu(DO3A-xy-TPEP) and ^99mTc-Sestamibi in the presence of excess Cys-A (Figure 5).

Both ^99mTc-Sestamibi and ^99mTc-Tetrafosmin have been studied for their potential to monitor the tumor MDR in pre-clinical animal models and in clinical studies (28–35). The MDR1 Pgp and MRPs are also overexpressed in normal organs, such as kidneys and liver (58–60). As demonstrated in the blocking study (Figure 5), the pre-administration of tumor bearing mice with excess Cys-A resulted in a significant increase of radioactivity accumulation in non-cancerous organs, such as kidneys, liver and lungs, and the delayed radioactivity excretion. Therefore, the MDR Pgp transport function is at least partially responsible of the fast efflux of ⁶⁴Cu(DO3A-xy-TPEP) from the kidneys, liver, and lungs.

The subcellular distribution characteristics of ⁶⁴Cu(DO3A-xy-TPEP) was compared with those of ^99mTc-Sestamibi, a well-know radiotracer clinically useful for cancer detection by SPECT. The standard differential centrifugation technique has been successfully used to study the myocardial localization mechanism of cationic ^99mTc radiotracers (61–66), such as ^99mTc-Sestamibi. As expected, only a small fraction (3 – 4%) of ⁶⁴Cu radioactivity is found in the mitochondrial fraction, and majority of the ⁶⁴Cu radioactivity is found in supernatants I and II. Almost identical distribution patterns were observed for ^99mTc-Sestamibi. It has been suggested the tissue homogenization and differential centrifugation techniques may damage the mitochondrial structure (61–66), which allows the leakage of cationic radiotracers, such as ⁶⁴Cu(DO3A-xy-TPEP) and ^99mTc-Sestamibi, from tumor mitochondria. Since ⁶⁴Cu(DO3A-xy-TPEP) and ^99mTc-Sestamibi share almost identical subcellular distribution patterns in the U87MG glioma, we believe that ⁶⁴Cu(DO3A-xy-TPEP), like ^99mTc-Sestamibi, is able to localize in the tumor mitochondria due to the increased plasma and mitochondrial transmembrane potentials in tumor cells as compared to normal epithelial cells (68–72). This conclusion is supported by the results from in vitro cellular assays, which showed that ⁶⁴Cu(DO3A-xy-TPP) (DO3A-xy-TPP: triphenyl(4-((4,7,10-tris(carboxymethyl)-1,4,7,10-tetraazacyclododecan-1-yl)methyl)benzyl)phosphonium) was able to localize in U87MG human glioma cells (36).

CONCLUSION

In this study, we evaluated the potential of ⁶⁴Cu(DO3A-xy-TPEP) as a new PET radiotracer for noninvasive monitoring of MDR transport function in several xenografted tumor models. It is important to note that MDR is often mediated by a combination of MDR Pgps (particularly MDR1 Pgp) and MRPs. ⁶⁴Cu(DO3A-xy-TPEP) has a high initial tumor uptake and show a steady uptake increase between 30 and 120 min p.i. in the MDR-negative U87MG glioma, which has little expression of MDR1 Pgp and MRPs. In the xenografted MDA-MB-435, MDA-MB-231, KB-3-1 and KB-v-1 tumors, which have high levels of MDR1 Pgp, MRP2 and MRP4, ⁶⁴Cu(DO3A-xy-TPEP) and ^99mTc-Sestamibi show a significant tumor washout over the 2 h period. Since ⁶⁴Cu(DO3A-xy-TPEP) has a greater tumor uptake difference between the MDR-negative U87MG glioma and MDR-positive tumors (MDA-MB-231, MDA-MB-435, KB-3-1 and KB-v-1) than ^99mTc-Sestamibi, we believe that ⁶⁴Cu(DO3A-xy-TPEP) is more efficient than ^99mTc-Sestamibi as the substrate for MDR Pgps and MRPs, and might be a more efficient radiotracer for noninvasive monitoring of the MDR transport function in tumors of different origin. Regardless of the source of MDR, the overall net effect is the rapid efflux of cationic radiotracers from tumor cells, which often leads to a significant reduction of the radiotracer tumor uptake.