Phospholipids (soybean lecithin for injection use, with phosphatidylcholine content > 70%) were purchased from Shanghai Tai-Wei Pharmaceutical Co., Ltd. (Shanghai, China). Cholesterol was obtained from Amresco (Solon, OH, USA). Curcumin was obtained from Sigma (St. Louis, MO, USA). Transferrin was obtained from Sigma (St. Louis, MO, USA). Bovine serum albumin-V was obtained from Amresco (Solon, OH, USA). Improved PRMI-1640 medium, fetal bovine serum and pancreatic enzyme were purchased from HyClone (Logan, UT, USA). Poloxamer 188 (F68) was obtained from BASF (China) Co., Ltd. (Shanghai, China). All other chemical reagents used in this study were of analytical grade or better. Oral squamous cell carcinoma cell line HN4 was provided by the state key laboratory of west China stomatology, Sichuan university.
Preparation of liposomal curcumin (Cur-Lips)
In this experiment, the modified ethanol injection method according to our previous works was used to prepare Cur-Lips . The specific methods are as follows: Combining curcumin, phospholipids, and cholesterol in ethanol was as an organic phase. The phosphate buffer solution (PBS, 0.01Mol/L, pH 7.4) containing F68 was used as the aqueous phase. F68 was served as a surfactant to narrow the size distributions. Then the organic phase was injected into the aqueous phase under magnetic agitation slowly, and the organic solvent was removed by roto evaporator in a water bath at 37°C for 30 minutes without light. The resulting suspension was first centrifuged at a low speed (3 000r·min− 1, 5min) to remove the free curcumin, then centrifuged at a high speed (16 000r·min− 1, 10min) to discard the supernatant. Finally, the precipitate was suspended in PBS (pH 7.4) to obtain Cur-Lips. For Cur-Lips, the mass ratio of phospholipid to cholesterol was 5:1, the mass fraction of F68 was 1%, the ratio of organic phase to aqueous phase was 10:1, the final concentration of phospholipid was 10mg/mL, and the final concentration of curcumin was 0.4mg/mL.
Preparation of transferrin-modified liposomal curcumin (Tf-Cur-Lips)
Tf-Cur-Lips were prepared by post-insertion method. Specific methods: Transferrin micelles were obtained by mercaptoylation of transferrin in a water bath at 37°C for 2h. Then the transferrin micelles were incubated with the prepared liposomal curcumin in a water bath at 37°C for 4h and was added to the dialysis bag (with an immobilized molecular weight of 10KDa) for dialysis for 12h . Transferrin was modified to liposomal curcumin by covalent binding of the amine group of transferrin to the carboxyl group of the liposomes. Free curcumin and transferrin were removed to obtain Tf-Cur-Lips.
Characterization of Cur-Lips and Tf-Cur-Lips
The particle size, size distribution and ζ-potential of Cur-Lips and Tf-Cur-Lips were examined by dynamic light scattering (DLS) and electrophoretic light scattering (ELS) technologies, respectively, using the instrument of Zetasizer Nano ZS90 (Malvern Instruments Ltd., Malvern, UK). The morphology of Cur-Lips and Tf-Cur-Lips was observed by scanning electron microscopy (SEM, INSPECT F, FEI, Eindhoven, The Netherlands). Before SEM, one drop of the properly diluted Cur-Lips or Tf-Cur-Lips suspension was placed on a clean glass sheet, followed drying in the air. The samples were coated with gold and observed under the scanning electron microscopy.
The entrapment efficiency of curcumin in Cur-Lips and Tf-Cur-Lips was determined by low temperature and high-speed centrifugation method. An appropriate amount of Cur-Lips or Tf-Cur-Lips was taken in a precise amount and centrifuged at 16,000 r·min− 1 for 5min. After centrifugation, the supernatant (1mL to 10mL) was diluted to the scale with anhydrous ethanol and filtered through 0.22µm organic filter membrane. The content of free curcumin was calculated by fluorescence spectrophotometry (Ex:458nm, Em:548nm) in the supernatant and presented as W1, i.e. Then, 1mL Cur-Lips or Tf-Cur-Lips was added to 10mL measuring bottle, anhydrous ethanol was added to the scale, and then filtered through 0.22µm organic filter membrane. The content of free curcumin was determined by fluorescence spectrophotometry in the filtrate and presented as W, i.e. The encapsulation efficiency (EE) of the curcumin was calculated using the following equation: EE% = (W - W1) / W × 100%.
In vitro release study of Cur-Lips and Tf-Cur-Lips
The in vitro release property of Cur-Lips and Tf-Cur-Lips was observed using the dynamic dialysis method . As a control, we measured the release rate of curcumin dissolved in PBS (pH7.4, 0.01Mol/L). 1mL curcumin solution, Cur-Lips or Tf-Cur-Lips (the final concentration of curcumin was 2mg/mL in these three groups) were added to dialysis bags with a cutoff molecular weight of 10KDa (only curcumin was allowed to pass, but liposomes were not). After tying the two ends of the dialysis bag tightly, they were immersed in 10mL of the release medium (PBS containing 0.5% Tween® 80, pH 7.4) and placed in a horizontal shaker (37°C, 100 r·min − 1). All the fluid outside the dialysis bag was removed at a fixed time point (0.5,1, 2, 4, 6, 8, 24, 48, 72,96,168h) and 10mL of fresh release medium was added to the release system. The collected samples were stored in the refrigerator at -80°C. After all samples were collected, the samples were 16000 r·min− 1 and centrifugated for 5min. A supernatant of 200 µL was taken to analyze the curcumin content (Em:458nm; Ex: 548nm).
Pharmacokinetics of Cur-Lips and Tf-Cur-Lips in rats
Eighteen healthy adult SD male rats of 200-250g were divided into three groups of six. Curcumin solution (dissolved in pH7.4, 0.01Mol/L PBS as a control group), Cur-Lips or Tf-Cur-Lips were intravenously injected into male SD rats with a final curcumin concentration of 200µg/kg (n = 6). At a predetermined time-point (15min, 30min, 1h, 2h, 4h, 6h, 8h, 24h, 48h, 72h), venous blood was taken 200µL and added to the heparin-treated EP tube. An aliquot of 50µL plasma was added into the tubes containing 10ng of emodin which was used as the inner standard. 100µL methanol and 100µL chloroform were added to extracts curcumin and emodin by vortex and sonication for 5 min in sequence. Plasma was separated by 16000 r·min− 1 centrifugation twice for 3min and 200µL fresh methanol was added to the residue followed by vortex and centrifugation again. The liquid was combined and dried by air flow. The residue was dissolved in 200µL of methanol by vortex for 3 min. The contents of curcumin in plasma at each time point were determined by HPLC internal standard method using a fluorescent detector (Em:458nm; Ex: 548nm), and the plasma drug concentration-time curve of curcumin was drawn to compare the differences among the groups.
Inhibition of Cur-Lips and Tf-Cur-Lips on HN4 cells proliferation
HN4 cells were cultured in vitro, and the logarithmic growth phase cells were taken. After digestion with trypsin, about 5000 cells per well were inoculated into 96-well plates. After the cells were further cultured in an incubator for 24h, curcumin solution (dissolved in pH7.4, 0.01Mol/L PBS as a control group), Cur-Lips or Tf-Cur-Lips were added, respectively. The final concentration of curcumin in each group was diluted to 0,10,20,40,80µmol/L with a modified serum free PRMI-1640 medium, and 5 secondary pores were set in each group. After treatment for 1d and 3d, the culture medium was discarded for each group, and the modified PRMI-1640 medium containing 10% cck-8 was added. After incubation at 37°C for 30min, the absorbance value at 450nm was detected, and the proliferation inhibition rate of HN4 cells was calculated.
Extraction of total RNA, RT-PCR and real-time PCR
HN4 cells were cultured in vitro and logarithmic growth phase cells were added with curcumin concentration of 40µmol/L curcumin solution (dissolved in pH7.4, 0.01Mol/L PBS as a control group), Cur-Lips or Tf-Cur-Lips for 24h, and serum-free modified PRMI-1640 medium was added as the control group. Total RNA, RT-PCR and real-time PCR were extracted to evaluate the expression of apoptosis-related genes P53 and Fas at transcriptional levels in HN4 cells. The total RNA of each group was extracted using the Simply. P total RNA extraction kit (BioFlux, hangzhou, China), and then the PrimeScript RT reagent kit with cDNA Eraser (Takara Bio, Tokyo, Japan) was used to transect 1µg of total RNA into cDNA. Method outlined in 'Molecular Cloning: A Laboratory Manual' (2001, 3rd edn). Total RNA and cDNA were detected using agar-gel electrophoresis. CDNA samples were amplified with RT-PCR kit (Tian-gen, Beijing, China) to establish the standard curve of specific genes. The primers of each gene used in this experiment are shown in Table 1. These primers were all from the established GenBank sequence. For real-time PCR, expression of target genes in HN4 cells were quantified using SYBR Premix Ex TaqTM (Perfect Real Time) kit (Takara) carried out on an ABI 7300 system (ABI, Foster City, CA, USA). For each reaction, a melting curve was generated to test primer dimer formation. The relative quantification of mRNA level was carried out by the double-standard curve method. To evaluate the PCR efficiency, the amplification of GAPDH (D-glyceraldehyde-3-phosphate dehydrogenase) was used as the control.
Primer sequences of target genes and GAPDH for real-time PCR assay
Primer sequence (5′–3′)
All assays in this study were repeated at least thrice. The t test was used for comparison between the two groups and a one-way ANOVA was used to analyze differences among groups. In all tables and figures, representative data were presented as the mean ± standard deviation. P values < 0.05 were considered statistically significant.