The folate receptor is an attractive target for selective tumor delivery of liposomal doxorubicin (DXR) because it is abundantly expressed in a large percentage of tumors. This study examined the effect of polyethylene glycol (PEG) spacer length and folate ligand density on the targeting ability of folate-modified liposomes. Liposomes were modified with folate-derivatized PEG-distearoylphosphatidylethanolamine with PEG molecular weights of 2000, 3400, or 5000. The association of DXR-loaded liposomes with KB cells, which overexpress the folate receptor, was evaluated by flow cytometry at various ratios of folate modification. A low ratio of folate modification with a sufficiently long PEG chain showed the highest folate receptor-mediated association with the cells, but did not show the highest in vitro cytotoxicity. DXR release from folate-modified liposomes in endosomes might be different. These findings will be useful for designing folate receptor-targeting carriers.
Antitumor drug delivery systems with nanoscopic dimensions have received much attention due to their unique accumulation behavior at the tumor site. Various nanoparticulate carriers such as liposomes, polymer conjugates, polymeric micelles, and nanoparticles are utilized for selective delivery of various anticancer drugs to tumors in a passive targeting manner [
Receptor-mediated endocytosis pathways have been exploited for tumor-specific targeting of nanocarriers and intracellular delivery of their contents. Modification of carriers with a ligand directed to an overexpressed receptor in cancer cells can improve selectivity and facilitate the movement of carriers into the intracellular compartment. One such candidate ligand is folic acid because the folate receptor-
Ligand density per drug carrier and spacer length are important in designing suitable carriers for targeting. However, the optimal ligand density on liposomes is controversial. Different densities of folate in liposomes (ligand/total lipid molar ratio) ranging between 0.01% and 1.0% have been reported in the literature as sufficient to promote liposome binding to the folate receptor on cells [
In this study, folate-mediated association of DXR-loaded liposomes with human oral carcinoma KB cells, which overexpress the folate receptor, was evaluated in terms of PEG spacer length and the ratio of modification with the folate ligand. Enhanced association of DXR in KB cells was shown with an extremely low ratio of folate modification and a sufficiently long PEG spacer length, but high cytotoxicity of DXR was observed with a high ratio of folate modification.
Hydrogenated soybean phosphatidylcholine (HSPC), aminopoly(ethyleneglycol)-distearoylphosphatidylethanolamine (amino-PEG-DSPE, PEG mean molecular weight of 2000, 3400, and 5000), and methoxy-PEG5000-DSPE (mPEG5000-DSPE) were obtained from NOF Corporation (Tokyo, Japan). Cholesterol (Ch), doxorubicin (DXR) hydrochloride, folic acid, and HPLC-grade acetonitrile were purchased from Wako Pure Chemical Industries, Ltd. (Osaka, Japan). Folate-derivatized PEG-DSPE (F-PEG2000-, F-PEG3400-, and F-PEG5000-DSPE), which are conjugates of folic acid and amino-PEG-DSPE, were synthesized as reported previously [
Liposomes were prepared from HSPC/Ch (55/45 mol/mol). All lipids were dissolved in chloroform, which was removed by evaporation. Lipophilic fluorescent marker DiI-labeled liposomes were prepared by the same procedure, but with the addition of DiI (0.4 mol% of total lipid) to the lipid mixture and without DXR loading. The film was hydrated with MgSO4 aqueous solution (300 mM, adjusted to pH 3.5 with HCl) and sonication. The resulting mean diameter of liposomes was about 130 nm, as determined by the dynamic light scattering method (ELS-800; Otsuka Electronics Co., Ltd., Osaka, Japan) at 25°C after diluting the liposome suspension with water.
DXR was encapsulated into liposomes using the ionophore-mediated loading method [
For comparison of loading procedures, DXR was encapsulated in liposomes by the pH gradient method [
The folate ligand was inserted into preformed liposomes by the postinsertion technique [
The release of drug from the liposomes in phosphate-buffered saline (PBS, pH 7.4 or 5.0) was monitored by a dialysis method. The dialysis was done at 37°C using seamless cellulose tube membranes (Spectrum, Houston, TX, USA) with a molecular weight cutoff of 300,000 Da and PBS as the sink solution. The initial concentration of DXR-loaded liposomes was 0.2 mg/mL. The sample volume in the dialysis bag was 1 mL, and the sink volume was 100 mL. The concentration of drug was analyzed at various times points during dialysis.
KB cells were obtained from the Cell Resource Center for Biomedical Research, Tohoku University (Miyagi, Japan). The cells were cultured in folate-deficient RPMI 1640 medium with 10% heat-inactivated FBS and kanamycin sulfate (50
The cells were prepared by plating 3 × 105 cells/well in a 12-well culture plate 1 day before the assay. The cells were incubated with DXR-loaded liposomes or DiI-labeled liposomes containing 20
KB cells were incubated with DXR-loaded liposomes (20
For efficient drug delivery to the target site, drugs should be stably entrapped in liposomes. In this study, an ionophore-mediated pH gradient method utilizing MgSO4 was applied to load DXR into liposomes because this method can effectively encapsulate drugs [
DXR release profile of liposomes loaded by the MgSO4/ionophore method (■) and pH gradient method using citrate buffer (○) in PBS (pH 7.4) at 37°C. Each value represents the mean
The average particle size of liposomes used in this study was approximately 130 nm, and the folate modification did not change the sizes of liposomes. More than 80% of folate ligand was inserted in liposomes at each ratio, which was confirmed after the separation of folate-modified liposomes by ultracentrifugation (100,000
In this study, the cellular association of folate-modified liposomes was evaluated in KB cells with respect to PEG spacer length and modification ratio by flow cytometry based on DXR fluorescence (Figure
Association of folate-modified liposomes with KB cells with 2-h incubation was determined by fluorescence of DXR-loaded liposomes (a) and DiI-labeled liposomes (b) using flow cytometry. (a) Folate modification from 0.03 to 1.0 mol% of F2-, F3-, and F5-PEG-DSPE. Inset: F5-PEG-DSPE modification from 0.01 to 0.3 mol%. (b) Folate modification from 0.01 to 0.3 mol% of F2-, F3-, and F5-PEG-DSPE.
Next, we confirmed whether folate might mediate cellular association with KB cells by mPEG-DSPE modification and free ligand competition (Figure
Association of DXR-loaded liposomes with KB cells with 1-h incubation was determined by flow cytometry. Cells were incubated with each liposome in folate-free medium or medium containing 1 mM folic acid.
The effect of incubation time on the cellular association of liposomes was then examined (Figure
Cellular association of DXR-loaded liposomes with time was determined by flow cytometry. KB cells were incubated with F5-L modified at 0.03 or 0.3 mol% or without folate modification (NF-L).
The effect of the folate modification ratio on cytotoxicity in KB cells was evaluated using the WST-8 assay (Figure
Cytotoxicity of DXR-loaded liposomes against KB cells. Cells were incubated with each liposome at a DXR concentration of 20
The release of free drug from liposomes is involved in cytotoxicity or antitumor activity [
DXR release profile of folate-modified liposomes (F5-L) in PBS (pH 5.0) at 37°C. Each point represents the mean of 2 experiments.
In this study, the effects of PEG spacer length and ligand density on folate receptor-targeted liposomes were evaluated. A low ratio of folate modification with a sufficiently long PEG chain (F-PEG5000-DSPE) increased folate receptor-mediated association, but a high ratio of folate modification enhanced in vitro cytotoxicity. This information will be useful for designing folate receptor-targeting carriers.
This work was supported in part by a Grant-in-Aid for Young Scientists (B) from the Ministry of Education, Culture, Sports, Science and Technology of Japan and by the Open Research Center Project. The authors would like to thank Mr. Ken Kajihara for his assistance in the experimental work.