Abstract
Liposomes are important carriers for controlling the spatial and temporal distribution of drug molecules or other bioactive molecules. Radiolabeled liposomes have potential applications in diagnostic imaging and radionuclide therapy. The purpose of this study was to develop a practical method for labeling liposomes with therapeutic rhenium radionuclides, using 186Re as an example. Methods: An SNS pattern ligand, N,N-bis(2-mercaptoethyl)-N′,N′-diethylethylenediamine (BMEDA), and an S pattern ligand, benzene thiol (BT), were used to make 2 kinds of 186Re-SNS/S complexes, 186Re-BMEDA and 186Re-BMEDA + BT. These 186Re-SNS/S complexes were mixed with neutral liposomes encapsulating cysteine or (NH4)2SO4 to prepare 186Re-liposomes. The in vitro labeling stability of 186Re-liposomes was investigated by incubation in 50% fetal bovine serum/50% phosphate-buffered saline, pH 7.4, at 37°C. Rat distribution studies of 186Re-liposomes after intravenous injection were also performed. Results: The labeling efficiencies of 186Re-liposomes were 52.9%–81.3% depending on the 186Re-SNS/S complex chosen and whether cysteine- or (NH4)2SO4-encapsulated liposomes were used. 186Re-(NH4)2SO4 liposomes labeled with 186Re-BMEDA had the best in vitro labeling stability in serum with 89.8% ± 3.1% of the radioactivity associated with liposomes at 24 h and 76.2% ± 5.1% at 96 h. A specific activity of 1.85 GBq (50 mCi) of 186Re per 50 mg of phospholipid could be achieved with good labeling stability. Biodistributions were followed for 72 h and showed good in vivo stability for 186Re-liposomes that was characterized by a slow blood clearance and a gradually increasing spleen accumulation. 186Re-BMEDA alone had fast blood clearance and no accumulation in spleen. Conclusion: A practical method for labeling liposomes with 186Re using 186Re-SNS/S complexes is described. The labeled 186Re-liposomes were stable in serum and in vivo and could potentially be useful for radionuclide therapy.
Liposomes are double-membrane lipid vesicles. They have been widely studied as important carriers in controlling the spatial and temporal distribution of drug molecules or other bioactive molecules for targeted therapy. Liposomes have been investigated widely as universal carriers of tumor chemotherapeutic agents (1,2), as antigen carriers to stimulate immune response (1,3), as carriers of nucleic acid for gene therapy (1,4), and as carriers of antibiotics for infectious disease treatment (1).
Liposomes are promising carriers for radionuclide therapy for the following reasons:
Biocompatibility: Lipids and cholesterol (Chol) used for liposome manufacture are common constitutes of cell membranes and therefore are easily metabolized.
Varying uniform sizes: Liposomes with variable homogeneous particle size ranges can readily be produced by using the extrusion technique (5).
Modification of surface properties: The surface of liposomes can be modified with different kinds of functional groups, such as antibodies, folic acid, peptides, and so forth (6–9), enabling radiolabeled liposomes to be used for molecular imaging and targeted radionuclide therapy.
Controlled migration of liposomes and release of radioisotopes from liposomes: Use of different liposome components and different labeling methods to control the liposome migration and radioisotope release from liposomes may be helpful for delivering a uniform dose distribution in the tumor tissue (10).
Use of physical modalities for targeting: Targeted hyperthermia and radiation to the targeted tissues (11,12) can significantly increase the accumulation of radiolabeled liposomes to targeted tissues.
There are many studies using 99mTc-, 67Ga-, and 111In-labeled liposomes for nuclear imaging (13–16), which have shown that radiolabeled liposomes have good accumulation characteristics in tumor, infection, and inflammation in vivo (17,18). Small unilaminar liposomes (19) composed of saturated lipids and Chol have been shown to have reduced uptake by the reticuloendothelial system (RES) compared with liposomes with no Chol. Theoretic dose calculation studies by Emfietzoglou et al. (20) have suggested that if methods for labeling these small unilaminar liposomes with therapeutic radionuclides could be developed, these intravenously administered liposomes would deliver a high radiation dose to tumor tissues while sparing the red marrow and having acceptable doses to liver and spleen. These theoretic calculations were based on distribution studies of 67Ga-labeled liposomes. Surface modification of liposomes using polyethylene glycol (PEG) can change the in vivo distribution of liposomes significantly after intravenous injection (1,21,22). In addition to intravenous delivery, other methods of delivering radiolabeled liposomes are also possible. Harrington et al. (23) have studied the biodistribution of 111In-labeled pegylated liposomes via intratumoral or subcutaneous injection techniques. Their results showed that pegylated liposomes have potential as vehicles for intratumoral and subcutaneous drug delivery.
Until now, no promising method has been described for labeling liposomes with the 186Re and 188Re therapeutic radioisotopes. We have previously described a method for labeling liposomes using 99mTc-SNS/S complexes (24), which have been studied previously as general radiopharmaceuticals intended for use mainly as brain imaging agents (25). Studies using the crystal structures of stable Re-SNS/S complexes and 186Re-labeled SNS/S complexes inferred that 99mTc-SNS/S complexes have a neutral core coordinate structure that is the same as the coordinate structure of Re-SNS/S complexes (26,27). The 186Re-SNS/S complexes can also be achieved by using routine practical radiolabeling methods (27). In this article, we report a liposome labeling method using 186Re-SNS/S pattern complexes (Fig. 1) and liposomes encapsulating cysteine or (NH4)2SO4.
Structure of 186Re/188Re-SNS/S complexes.
MATERIALS AND METHODS
Liposome Manufacture and Characterization
Two liposome formulations, cysteine-encapsulated liposomes and (NH4)2SO4-encapsulated liposomes (transmembrane ammonium gradient liposomes), were studied for 186Re labeling. All liposome formulations were comprised of the same lipid components, distearoyl phosphatidylcholine (DSPC) (Avanti Polar Lipids), Chol (Calbiochem), and α-tocopherol (Aldrich) at a molar percentage of 54:44:2. All lipids were used without further purification. Liposomes encapsulating cysteine were produced as previously outlined by Goins et al. (21). Liposomes encapsulating (NH4)2SO4 were prepared from a modification of the method by Maurer-Spurej et al. (28).
Cysteine-Encapsulated Liposomes.
DSPC, Chol, and α-tocopherol were mixed and dissolved in chloroform. Chloroform was then removed by rotary evaporation to form a lipid film. The lipid film was stored overnight in a vacuum desiccator to remove organic solvent. Samples were rehydrated with 300 mmol/L sucrose (Sigma) in sterile water for injection and warmed to 55°C for 15 min with periodic vortexing until the lipids were in suspension. The resultant multilamellar vesicles formed from rehydration were then frozen in liquid nitrogen and lyophilized. The resultant dry sugar-lipid preparations were then rehydrated with 200 mmol/L cysteine (Sigma) in Dulbecco’s phosphate-buffered saline (PBS), pH 6.3, at a total lipid concentration of 120 mmol/L. The solutions were then diluted at a v/v ratio of 1 part lipid suspension to 2 parts PBS, pH 6.3, containing 150 mmol/L sucrose, and 100 mmol/L cysteine. The diluted lipid suspensions were then extruded through a series of polycarbonate filters (extruder, Lipex Extruder; filter, Whatman Nucleopore) at 55°C: 400-nm liposomes: 2 passes, 2 μm; then 5 passes, 400 nm; 100-nm liposomes: 2 passes, 2 μm; then 2 passes, 400 nm; then 2 passes, 200 nm; then 5 passes, 100 nm. The extruded lipid solution was then washed in PBS, pH 6.3, containing 75 mmol/L sucrose and centrifuged at 200,000 × g for 45 min to remove unencapsulated sucrose and cysteine and to concentrate the liposome sample. The washing step was repeated 3 times. The final liposome pellet was resuspended in PBS, pH 6.3, containing 300 mmol/L sucrose at a lipid concentration of 120 mmol/L and stored in the refrigerator at 4°C for up to 3 mo until needed for radiolabeling studies.
Liposomal size was monitored using 488-nm laser light scattering (Dynamic Light Scattering; Brookhaven Instruments). The measured sizes were 312.6 ± 27.7 nm for 400-nm cysteine liposomes and 115.8 ± 11.7 nm for 100-nm cysteine liposomes. Phospholipid concentrations determined using Stewart’s method (29) were 44.4 mg/mL for 400- and 100-nm cysteine liposomes. Cysteine concentrations determined using a GSH-400 assay kit (OXIS International) were 5.5 mmol/L for 400-nm liposomes and 6.3 mmol/L for 100-nm liposomes. Bacterial and endotoxin tests were performed by Pathology Service Referral Laboratory, University Health System. No bacterial growth during 14 d of incubation at 37°C was observed for either 400- or 100-nm cysteine liposomes. A Gram-negative antigen screen test showed that the endotoxin level was >12.5 but <25 endotoxin units (EU)/mL for 100-nm cysteine liposomes, which were administered to rats.
(NH4)2SO4-Encapsulated Liposomes.
The lipid film samples prepared as described above were rehydrated with 300 mmol/L (NH4)2SO4 (Sigma) in sterile water for injection (120 mmol/L total lipid) and warmed to 55°C for 15 min with periodic vortexing until the lipids were in suspension. The resultant multilamellar vesicles formed from rehydration were then frozen in liquid nitrogen and thawed at 55°C for 5 cycles. The solutions were then diluted at a v/v ratio of 1 part lipid suspension to 1 part 300 mmol/L (NH4)2SO4 solution. The diluted lipid suspensions were then extruded through the series of polycarbonate filters as described above. The extruded lipid solution was then stored in the refrigerator at 4°C for up to 6 mo until needed for radiolabeling studies.
Liposomal sizes were 443.1 ± 29.9 nm for 400-nm (NH4)2SO4 liposomes and 147.7 ± 12.5 nm for 100-nm (NH4)2SO4 liposomes. Phospholipid concentrations were 21.9 mg/mL for 400-nm (NH4)2SO4 liposomes and 21.2 mg/mL for 100-nm (NH4)2SO4 liposomes. No bacterial growth was observed for both 400- and 100-nm (NH4)2SO4 liposomes. The endotoxin level was >5 but <12.5 EU/mL for the 100-nm (NH4)2SO4 liposomes.
Preparation of 186Re-SNS/S Complexes
An SNS pattern ligand, N,N-bis(2-mercaptoethyl)-N′,N′-diethylethylenediamine (BMEDA), and an S pattern ligand, benzene thiol (BT), were investigated. BMEDA was synthesized from a modification of the method by Corbin et al. (30), and the chemical structures were verified using 1H/13C NMR. BT was purchased from Aldrich. Two kinds of 186Re-SNS/S complexes (Fig. 2), 186Re-BMEDA and 186Re-BMEDA + BT, were studied as intermediates of liposome labeling.
Structures of 186Re-BMEDA (A) and 186Re-BMEDA + BT (B).
186Re-aluminum perrhenate (186Re-Al(ReO4)3) was purchased from Missouri University Research Reactor. Stannous chloride (Aldrich) was used as the reductant and glucoheptonate (GH) (Sigma) was used as an intermediate ligand to make 186Re-SNS/S complexes. Either 2.0 μL (2.2 mg) of BMEDA or 1.0 μL (1.1 mg) of BMEDA and 0.5 μL (0.5 mg) of BT was pipetted into a new vial. Then, 1.0 mL of 0.17 mol/L GH-0.10 mol/L acetate solution, pH 5.0, was added, followed by the addition of 80 μL of stannous chloride (15 mg/mL). The pH of the solution was adjusted to 5.0 with 1.0 mol/L NaOH. After flushing the solution with N2 gas, 44.4 MBq (1.2 mCi) or 504 MBq (13.6 mCi) of 186Re-Al(ReO4)3 (0.4 μg Re per mCi 186Re) was added. The vial was sealed and heated in an 80°C water bath for 1 h. The labeling efficiency of the 186Re-BMEDA complex was checked by paper chromatography with either methanol (Rf values: 186ReO4−, 0.4–0.7; 186Re-GH, 0.0–0.2; 186Re-BMEDA or 186Re-BMEDA + BT, 0.8–1.0) or saline (Rf values: 186ReO4−, 0.7–0.9; 186Re-GH, 0.7–1.0; 186Re-BMEDA or 186Re-BMEDA + BT, 0.0–0.2) as the eluent.
186Re-Liposome Labeling
For convenience, 400-nm liposomes were used for in vitro stability studies because a low-speed tabletop centrifuge could be used to quickly separate the liposome pellet from supernatant. Our previous studies using glutathione (GSH) liposomes labeled with 99mTc-SNS/S complexes suggested the good correlation between the in vitro labeling stability using 400-nm liposomes and the in vivo labeling stability using 100-nm liposomes. For rat distribution studies, 100-nm liposomes were used to permit better assessment of in vivo stability because it is known from the literature that liposomes with particle sizes of >100 nm are rapidly cleared from the blood (1).
400-nm Liposomes.
Immediately before radiolabeling, 0.2 mL of 400-nm (NH4)2SO4 liposomes containing 5 mg of DSPC (60 mmol/L total lipid) were prepared by dilution with 1.2 mL of PBS buffer, pH 7.4, and centrifugation at 11,000 × g for 10 min to remove the extraliposomal (NH4)2SO4. The supernatant was discarded and 0.60 mL of PBS buffer, pH 7.4, was added to resuspend liposomes. Cysteine liposomes (400 nm) (120 mmol/L, 0.10 mL) and 0.50 mL of PBS buffer, pH 7.4, were moved to a new vial for cysteine liposome labeling.
Freshly washed 400-nm (NH4)2SO4 liposomes were mixed with 22.2 MBq (0.6 mCi) or 252 MBq (6.8 mCi) of 186Re-BMEDA or 186Re-BMEDA + BT solution adjusted to pH 7.0 and incubated at 37°C for 1 h. The 400-nm cysteine liposomes prepared as described above were mixed with 22.2 MBq (0.6 mCi) of 186Re-BMEDA or 186Re-BMEDA + BT solution (pH 7.0) and incubated at 37°C for 1 h. The labeling efficiency was determined as the ratio of the activity associated with liposome pellet after centrifugation.
100-nm Liposomes.
Immediately before radiolabeling, 2.0 mL of 100-nm (NH4)2SO4 liposomes containing 50 mg of DSPC were diluted with 2.0 mL of PBS buffer, pH 7.4, and centrifuged at 47,000 × g for 45 min to remove the extraliposomal (NH4)2SO4. The supernatant was discarded and 1.0 mL of PBS buffer, pH 7.4, was added to resuspend liposomes. Cysteine liposomes (100 nm) (120 mmol/L, 1.0 mL) were moved to a new vial for cysteine liposome labeling.
The preparation of 186Re-BMEDA for the labeling process of 100-nm liposomes for biodistribution studies was similar to that described above. 186Re-Perrhenate (3.7 GBq [100 mCi]) and 4.5 μL (5.0 mg) of BMEDA were used to make 186Re-BMEDA. The liposomes encapsulating cysteine or (NH4)2SO4 prepared as described above were mixed with 2.22 GBq (60 mCi) of the 186Re-BMEDA solution and incubated at 37°C for 1 h. Sephadex G-25 column chromatography with PBS buffer, pH 7.4, was used to separate 100-nm radiolabeled liposomes from free 186Re-BMEDA. The liposomes are eluted from the column first and can be conveniently collected and visualized due to their opacity. The labeling efficiency was determined from the 186Re activity before and after separation using a Radix dose calibrator.
In Vitro Labeling Stability Study of 186Re-Liposomes
The in vitro labeling stabilities of 400-nm 186Re-cysteine liposomes and 400-nm 186Re-(NH4)2SO4 liposomes labeled using 186Re-BMEDA or 186Re-BMEDA + BT were studied comparably. After separation of 400-nm 186Re-liposomes from free 186Re-SNS/S complexes by centrifugation, evaluation of the in vitro labeling stabilities of 186Re-liposomes was performed by incubating 186Re-liposomes (18.5 MBq [0.5 mCi] or 185 MBq [5 mCi] on average) in 1.6 mL of 50% fetal bovine serum (FBS) (GIBCO)-PBS buffer, pH 7.4, at 37°C. At certain times after serum incubation, 40 μL of 186Re-liposome solution was removed with a micropipette to a test tube and counted for total radioactivity using a Minaxiγ A5550 γ-counter (Packard). Then, the incubation solution was centrifuged at 11,000 × g for 10 min and 40 μL of supernatant was removed to a fresh test tube and counted for radioactivity that did not associate with the liposome pellet. The liposome pellet and the remaining supernatant were resuspended and allowed to continue incubation until the next time point.
In Vivo Distribution of 186Re-Liposomes in Normal Rats
The 100-nm 186Re-liposomes were used for normal rat biodistribution studies. To ensure there was no free 186Re-BMEDA in the liposome solutions, 100-nm 186Re-liposome solutions were separated twice with Sephadex G-25 column chromatography before intravenous injection.
The animal experiments were performed according to the National Institutes of Health Animal Use and Guidelines and were approved by our Institutional Animal Care Committee. Rat distributions after intravenous injection of 100-nm 186Re-cysteine liposomes or 100-nm 186Re-(NH4)2SO4 liposomes labeled with 186Re-BMEDA were investigated in normal Sprague–Dawley male rats. For comparison, normal rat distributions of free 186Re-BMEDA were also studied. After the rats were anesthetized by inhalation with isoflurane (3% in 100% oxygen), each rat (380 g on average) was injected with radiolabeled liposomes containing 114.7–133.2 MBq (3.1–3.6 mCi) of 186Re and 4.2 mg of DSPC or 133.2 MBq (3.6 mCi) of free 186Re-BMEDA. Planar images of the anesthetized rats in the prone position were collected at various times with a Picker Dyna 4 γ-camera interfaced to a Pinnacle computer workstation (MedaSys) (acquisition time: 1 min per image at baseline and at 1 and 4 h; 2 min per image at 24 and 72 h.). A low-energy, high-resolution collimator was used. The energy window was set at 137 keV ± 20%. The image size was set at 64 × 64.
After 72 h, anesthetized rats were euthanized by cervical dislocation and the biodistribution of 186Re-liposomes or free 186Re-BMEDA in various rat tissues was measured with the Minaxiγ A5550 γ-counter. Femur with bone marrow was taken as representative of bone and bone marrow. Bowel activity was determined by counting an aliquot of bowel plus contents after digestion in saturated NaOH. The 72-h total urine activity was determined by counting an aliquot of urine sample from each rat. Total blood, bone, muscle, and skin mass of rats were calculated as 5.4%, 10%, 40%, and 13% of total body weight, respectively (31,32).
Statistical Analysis
The MiniTab program (MiniTab Inc.) was used to perform statistical analysis. All average values are given as mean ± SD. The comparison of labeling efficiency and percentage injected dose (%ID) per organ between groups was determined using 1-way ANOVA. The acceptable probability for a significant difference was P < 0.05.
RESULTS
186Re-Liposome Labeling Efficiency
The labeling efficiencies of 400-nm cysteine or 400-nm (NH4)2SO4 liposomes labeled with 22.2 MBq (0.6 mCi) of 186Re-BMEDA or 186Re-BMEDA + BT are shown in Table 1. The labeling efficiency of 186Re-cysteine liposomes labeled with 186Re-BMEDA was significantly lower than that of 186Re-(NH4)2SO4 liposomes labeled with 186Re-BMEDA or 186Re-BMEDA + BT (P < 0.05). The labeling efficiency of 400-nm 186Re-(NH4)2SO4 liposomes labeled with 252 MBq (6.8 mCi) of 186Re-BMEDA was 74.3% ± 6.4% (n = 3).
Labeling Efficiencies of 400-nm 186Re-Liposomes Labeled with 186Re-BMEDA and 186Re-BMEDA + BT (n = 3)
In Vitro Labeling Stability of 186Re-Liposomes
The in vitro labeling stabilities of 186Re-cysteine liposomes and 186Re-(NH4)2SO4 liposomes labeled with 22.2 MBq (0.6 mCi) of 186Re-BMEDA or 186Re-BMEDA + BT are shown in Figure 3. 186Re-(NH4)2SO4 liposomes labeled with 186Re-BMEDA had the best in vitro labeling stability after incubation in 50% FBS-PBS, pH 7.4, at 37°C for 96 h (Fig. 3A). There was 89.8% ± 3.1% (n = 3) radioactivity associated with (NH4)2SO4 liposomes at 24 h and 76.2% ± 5.1% (n = 3) radioactivity associated with liposomes at 96 h. In contrast, 186Re-(NH4)2SO4 liposomes labeled with 186Re-BMEDA + BT had lower in vitro stability compared with 186Re-(NH4)2SO4 liposomes labeled with 186Re-BMEDA (P < 0.05 at 48, 72, and 96 h) (Fig. 3A). 186Re-(NH4)2SO4 liposomes labeled with 186Re-BMEDA and with 186Re-BMEDA + BT behaved in a similar fashion with a linear 186Re release from the liposomes (R2 = 0.995 and 0.999, respectively; P < 0.0001), although 186Re-(NH4)2SO4 liposomes labeled with 186Re-BMEDA + BT showed faster 186Re release.
(A) In vitro labeling stability (n = 3) of 400-nm 186Re-(NH4)2SO4 liposomes labeled with 186Re-BMEDA and with 186Re-BMEDA + BT in 50% FBS-PBS buffer, pH 7.4, at 37°C. (B) In vitro labeling stability (n = 3) of 400-nm 186Re-cysteine liposomes labeled with 186Re-BMEDA and with 186Re-BMEDA + BT in 50% FBS-PBS buffer at 37°C. 186Re-(NH4)2SO4 liposomes labeled with 186Re-BMEDA had the best in vitro labeling stability of the liposomes tested. This stability could be retained for 96 h in 50% FBS-PBS buffer at 37°C (*P < 0.05).
186Re-Cysteine liposomes showed less in vitro labeling stability compared with 186Re-(NH4)2SO4 liposomes. A similar in vitro labeling stability for 186Re-cysteine liposomes labeled with 186Re-BMEDA and with 186Re-BMEDA + BT was observed (Fig. 3B).
The effect of high specific activity (10 times more 186Re activity per mg of lipid) on the in vitro stability of 186Re-(NH4)2SO4 liposomes labeled with 186Re-BMEDA was investigated. A labeling efficiency of 74.3% ± 6.4% was achieved using 2.52 GBq (68 mCi) of 186Re per 50 mg of DSPC. A specific activity of 1.87 ± 0.16 GBq (50.5 ± 4.4 mCi) of 186Re per 50 mg of DSPC labeled to (NH4)2SO4 liposomes also showed good in vitro labeling stability up to 96 h of incubation in 50% FBS-PBS buffer at 37°C (Fig. 4). There were 89.1% ± 0.6% (n = 3) radioactivity associated with liposomes at 24 h and 66.7% ± 1.5% (n = 3) radioactivity associated with liposomes at 96 h (P < 0.05 at 48, 72, and 96 h).
Effect of high specific activity on in vitro labeling stability of 400-nm 186Re-(NH4)2SO4 liposomes labeled with 186Re-BMEDA. With (NH4)2SO4 liposomes, specific activity of 1.85 GBq (50 mCi) of 186Re per 50 mg of DSPC (1 mL 120 mmol/L liposomes) labeled to (NH4)2SO4 liposomes also showed good in vitro labeling stability up to 96 h of incubation in 50% FBS-PBS buffer at 37°C (*P < 0.05).
Normal Rat Distribution of 186Re-Liposomes
The labeling efficiency for 100-nm 186Re-cysteine liposomes was 70.0% and for 100-nm 186Re-(NH4)2SO4 liposomes was 82.6%. The γ-camera images of rats acquired at different times after intravenous injection of 100-nm 186Re-(NH4)2SO4 liposomes labeled with 186Re-BMEDA, 100-nm 186Re-cysteine liposomes labeled with 186Re-BMEDA, or 186Re-BMEDA alone are shown in Figure 5. Both 100-nm 186Re-liposome formulations tested showed slow blood clearances and spleen accumulations. The spleen accumulation is stable even at 72 h after intravenous injection. This spleen uptake is a common feature of liposome distribution with the liposome composition used in this study after intravenous injection in rats (1,17). Other liposome formulations that have surface modification with PEG have greatly reduced spleen accumulation (21,22). The 186Re-liposomes also have delayed excretion from the hepatobiliary system compared with 186Re-BMEDA alone. 186Re-BMEDA alone had a faster blood clearance and did not show any spleen accumulation (Fig. 5). 186Re-BMEDA alone also showed significant excretion from the hepatobiliary system and high activity in kidneys.
γ-Camera images of normal rats via intravenous injection method. (Top row) Images of rat at baseline and at 1, 4, 24, and 72 h after intravenous injection of 100-nm 186Re-(NH4)2SO4 liposomes labeled with 186Re-BMEDA. (Middle row) Images of rat at corresponding times after intravenous injection of 100-nm 186Re-cysteine liposomes labeled with 186Re-BMEDA. (Bottom row) Images of rat at corresponding times after intravenous injection of 186Re-BMEDA alone. 186Re-(NH4)2SO4 liposomes and 186Re-cysteine liposomes labeled with 186Re-BMEDA showed slow blood clearances and spleen accumulations, which are common features of liposome distribution after intravenous injection in rats. 186Re-BMEDA alone showed fast blood clearance, fast excretion from bowel and urine, and no spleen accumulation.
Normal rat distributions of 186Re-cysteine liposomes and 186Re-(NH4)2SO4 liposomes labeled with 186Re-BMEDA at 72 h are listed in Table 2. For comparison, biodistributions of the free 186Re-BMEDA used for liposome labeling were also obtained. The 100-nm 186Re-(NH4)2SO4 liposomes and 100-nm 186Re-cysteine liposomes labeled with 186Re-BMEDA showed significant spleen accumulation at 72 h after intravenous injection (P < 0.001). This shows the common feature of liposome distribution after intravenous injection in rats (1,17). 186Re-BMEDA alone did not have any spleen accumulation.
Normal Rat Distributions of 2 Kinds of 100-nm 186Re-Liposomes Labeled with 186Re-BMEDA and 186Re-BMEDA Alone at 72 Hours (n = 4)
Comparisons between 186Re-cysteine liposomes and 186Re-(NH4)2SO4 liposomes indicate that 186Re-(NH4)2SO4 liposomes have significantly higher radioactivity in spleen, blood, liver, kidney, muscle, and bone with bone marrow at 72 h (P < 0.001 for spleen, liver, kidney, and bone with bone marrow; P < 0.05 for blood and muscle); 186Re-(NH4)2SO4 liposomes have significantly higher radioactivity in the total activity of bowel and feces (P < 0.001) and significantly lower excretion from urine at 72 h (P < 0.01).
DISCUSSION
There are 3 types of liposome labeling methods (33). An ideal labeling method consists of trapping the radioisotopes within the inner space of premanufactured liposomes with high labeling efficiency (34). To perform this labeling method, a radiolabeled chemical with a certain lipophilicity is required so that it can go across the lipophilic double membrane of the liposome. After crossing the lipophilic double membrane, this radiolabeled chemical can transfer to a hydrophilic chemical within the inner space of the liposome where it could potentially be trapped in a stable fashion.
Phillips et al. (13) developed an effective 99mTc-liposome labeling method with 99mTc-hexamethylpropyleneamine oxime (HMPAO) and GSH-encapsulated liposomes. By using this labeling method, the labeling can be conveniently performed after liposomes are manufactured, resulting in very stable 99mTc-liposomes. 99mTc-Liposomes have been widely investigated as diagnostic imaging agents. These studies have demonstrated that 99mTc-liposomes can be effective agents for infection and inflammation imaging, tumor imaging, lymphoscintigraphy, and blood-pool measurement (9,17,18,21).
It is hypothesized that the lipophilic 99mTc-HMPAO crosses the double membrane of liposomes, after which it reacts with GSH and transforms into a hydrophilic composite in the inner space of liposomes. The 99mTc is thus trapped in the inner space of liposomes.
99mTc-HMPAO was introduced as a brain imaging agent. Since 99mTc-HMPAO is a 99mTc-N4 pattern complex, it has low stability for remaining in the original neutral coordinate structure (35). Although rhenium and technetium share similar chemical characteristics, there have been no reports of radiolabeled Re-HMPAO. The most likely reason, which can be derived from studies of technetium and rhenium chemistry, is that rhenium probably does not form a stable neutral coordinate structure with HMPAO.
Hafeli et al. (33) reported a method for labeling liposomes with 186Re/188Re. These liposomes were labeled by incorporating a 186Re/188Re complex into liposomes during liposome manufacture, making this method impractical in a clinical setting. The labeling efficiency with this method also was low (method I, 0%–20.9%; method II, 45.5% ± 5.3%), and a biodistribution study of 186/188Re-labeled liposomes was not reported.
Another report of labeling liposomes with a therapeutic radionuclide has been described. Utkhede et al. reported a 90Y-liposome labeling method using a cation ionophore and diethylenetriaminepentaacetic acid–encapsulated liposomes (36). The 90Y uptake by liposomes was >95%. The limitation of this method is that the labeling must be performed at a temperature of >41°C, which makes the labeling impractical for use with temperature-sensitive liposomes. In addition, 90Y may not be an ideal radionuclide for liposome labeling because of the following inherent physical and biochemical characteristics: (a) 90Y3+ is a bone-seeking chemical (37), so that the dissociated 90Y3+ accumulates in bone, potentially resulting in a high radiation dose to bone marrow; and (b) 90Y is a pure β-emitter (38), which makes it difficult to trace the in vivo distribution of 90Y-labeled agents directly.
99mTc-SNS/S pattern complexes (25) were initially studied for potential use as brain imaging agents. Because no stable nuclides exist for technetium, rhenium was used as a technetium analog to analyze the structure of these SNS/S pattern complexes. It was proven (26,27) that Re-SNS/S complexes have a chemistry similar to that of 99mTc-SNS/S complexes and have a core coordinate structure of the complex that is neutral so that lipophilic complexes can be made with this kind of coordinate system. The lipophilic characteristic of 99mTc-SNS/S, 186Re-SNS/S, or 188Re-SNS/S complexes suggests that these complexes can cross the lipophilic double membrane of a liposome.
Nock et al. (39) suggested that a ligand exchange behavior occurred between 99mTc-SNS/S complexes and other molecules containing a thiol group, such as GSH and cysteine. This suggests that 99mTc-SNS/S and 186Re/188Re-SNS/S complexes may change into hydrophilic composites after they enter into the inner space of liposomes and react with preencapsulated GSH, cysteine, or other hydrophilic chemicals containing a thiol group and be trapped.
Liposomes encapsulating citrate or other weak acids with lower inner space pH (pH gradient liposomes) or encapsulating (NH4)2SO4 (ammonium gradient liposomes) have been used in the entrapment of chemotherapeutic agents for drug delivery (1,2,28). The mechanism of the drug entrapment in these kinds of liposomes was studied, and it has been shown that the chemotherapeutic agents containing an amine group can be entrapped in the lower pH environment of liposome inner space. The 186Re complexes we have studied also contain amine groups that make the entrapment of 186Re-SNS/S complexes by the ammonium gradient mechanism possible. In the ammonium gradient situation, after radiolabeled complexes enter into the liposome’s inner space, the amine group of the complex is protonized and the complex becomes hydrophilic and trapped. It would also be possible to use both the ligand exchange mechanism with GSH/cysteine and the ammonium or pH gradient mechanism described above to potentially further improve the labeling of liposomes with 186Re/188Re.
In this study, there was similar in vitro labeling stability of 186Re-cysteine liposomes labeled with 186Re-BMEDA compared with that of 186Re-cysteine liposomes labeled with 186Re-BMEDA + BT. But 186Re-(NH4)2SO4 liposomes labeled with 186Re-BMEDA + BT showed lower in vitro labeling stability compared with that of 186Re-(NH4)2SO4 liposomes labeled with 186Re-BMEDA (P < 0.05 at 48, 72, and 96 h). The reason for this difference is likely related to the fact that 186Re-BMEDA has more amine groups that can be protonized. 186Re-BMEDA + BT has less amine groups available and is more lipophilic compared with 186Re-BMEDA. In comparison, 186Re-cysteine liposomes have similar hydrophilic structures after the ligand exchange reaction between cysteine and 186Re-BMEDA or 186Re-BMEDA + BT.
Typical distributions of standard liposome formulations in the 100- to 200-nm size range that are administered via intravenous injection have the slow blood-pool clearance and spleen accumulation (1,17). Our experiments showed the stability of 186Re-liposomes labeled with 186Re-BMEDA. A significantly higher spleen accumulation was observed even at 72 h after intravenous injection of 186Re-liposomes compared with 186Re-BMEDA alone (P < 0.001). 186Re-BMEDA alone showed fast blood clearance, fast excretion from bowel and urine, and no spleen accumulation (Fig. 5).
The labeling method reported in this article may be applied to different kinds of liposomes with different lipid compositions. To show the feasibility of the labeling method, we used nonspecifically targeted liposomes made from DSPC and Chol. These liposomes used in the normal distribution studies ranged in size from 115 to 150 nm. The liposome dose was only 4.2 mg of DSPC per rat. These liposomes were used to assess in vivo stability by monitoring the normal distribution in the spleen. Although high spleen uptake provides assurance of in vivo liposome stability, this high spleen uptake may not be desirable for intravenous liposome radiotherapy. Changes in liposome size and surface characteristics can greatly affect liposome biodistribution. Small liposomes of <100 nm with a high Chol content were found to have relatively low spleen uptake and excellent calculated dose distribution to tumors (19). In addition, liposome surface modifications using PEG (1,21,22) or functional groups (6–9)—such as antibodies and the biotin–avidin system—and physical modalities (11,12)—such as hyperthermia and radiation—can change the behavior of radiolabeled liposomes significantly, potentially resulting in much lower levels in liver and spleen and much higher accumulation in tumors.
The amount of liposomes injected can also change the in vivo behavior of liposomes. Lower spleen and liver radioactivity may also be achieved by injecting unlabeled liposomes beforehand, thus saturating the RES (1). In addition, some reports have suggested that radiolabeled liposomes can be used to treat tumors using intratumoral injection techniques (23,40). This flexibility in the preparation of different liposome structures and the use of various administration techniques indicates the great potential of using radiolabeled liposomes in radionuclide therapy.
By using the above methods, liposomes can be radiolabeled with high specific activity to meet the requirements of clinical treatment. Based on our experiments, a specific activity of 1.85 GBq (50 mCi) of 186Re per 50 mg of DSPC can be readily achieved. Considering the presence of carrier in 186Re-perrhenate and the half-life difference between 186Re and 188Re, 1.85 GBq (50 mCi) of 186Re corresponds to >22.2 GBq (>600 mCi) of carrier-free 188Re on the same 50-mg lipid dose.
CONCLUSION
We have introduced a method of labeling liposomes with 186Re by using 186Re-SNS/S complexes and liposomes encapsulating hydrophilic chemicals containing thiol groups or an ammonium gradient. This labeling method is convenient and permits achievement of high specific activities. 186Re/188Re-Liposomes have potential for clinical radionuclide therapy applications.
Acknowledgments
The authors thank the San Antonio Area Foundation for financial support.
Footnotes
Received Jun. 17, 2003; revision accepted Sep. 8, 2003.
For correspondence or reprints contact: William T. Phillips, MD, Department of Radiology, MSC 7800, UTHSCSA, 7703 Floyd Curl Dr., San Antonio, TX 78229-3900.
E-mail: phillips{at}uthscsa.edu
REFERENCES
In this issue
Jump to section
Related Articles
Cited By...
- Evaluation of Injectable Chitosan-based Co-cross-linking Hydrogel for Local Delivery of 188Re-LIPO-DOX to Breast-tumor-bearing Mouse Model
- MicroSPECT/CT Imaging and Pharmacokinetics of 188Re-(DXR)-liposome in Human Colorectal Adenocarcinoma-bearing Mice
- Cancer Nanotargeted Radiopharmaceuticals for Tumor Imaging and Therapy
- Nanoliposomes for Cancer Imaging and Therapy: Labeling Methods and Potential Applications
- Intraoperative 186Re-Liposome Radionuclide Therapy in a Head and Neck Squamous Cell Carcinoma Xenograft Positive Surgical Margin Model