HAP nanoparticles (nHAP) were purchased from Nanjing Emperor Nano Material Co. Ltd. (Nanjing, China). To overcome the disadvantages of nHAP as a gene delivery vector, we modified nHAP using branched PEI to prepare a new nanoparticle (which we named MnHAP), according to the protocol described by Ashokan et al (28) with some modifications. In brief, 200 mg nHAP powder was dissolved in double distilled water and stirred at 1200 rpm/min for 12 h. Then, PEI (MW 25 kDa, Sigma-Aldrich, St. Louis, MO, USA) was added to the nHAP microemulsion and stirring was continued for another 12 h, all at room temperature. The generated MnHAP microemulsion was centrifuged at 5000 rpm for 30 min after ultrasonification at 20 KHz for 5 min. The supernatant was collected for 48 h dialysis to remove the unreacted PEI. The resulting sample was lyophilized and stored at room temperature.

The particle size distribution and zeta potential of MnHAP preparations were measured by Malvern Zetasizer Nano ZS90 (Malvern Instruments, Ltd., Malvern, UK). The morphology of the nanoparticles was examined using a Hitachi H-7650 transmission electron microscope at 80 kV.

Different amounts of MnHAP (μg) were mixed with a fixed amount (0.25 μg) of negative control siRNA (sense: 5′-UUC UCC GAA CGU GUC ACG UTT-3′, anti-sense: 5′-ACG UGA CAC GUU CGG AGA ATT-3′, Shanghai GenePharm Co., Ltd) according to the MnHAP:siRNA weight ratios (0:1, 0.2:1, 0.4:1,1:1,2:1,4:1) by gentle pipetting for 30 min at room temperature. The above six mixtures were then electrophoresed on a 3% agarose gel at 100 V for 20 min. The bands were visualized under a UV transilluminator.

The human liver cancer cell line BEL-7402 was obtained from the Laboratory Animal Center of Sun Yat-sen University of China. The cells were grown in RPMI-1640 (Gibco, USA) supplemented with 10% fetal bovine serum (FBS, Gibco) under 5% CO₂ in air at 37°C.

For transfection efficiency analysis, BEL-7402 cells were plated in 6-well plates at a density of 5×10⁵/well and allowed to adhere for 24 h. Green fluorescent negative-control siRNA (FAM-siRNA) (2.5 μg) and 5.0 μg MnHAP, or 5 μl Lipofectamine 2000 were used per well (sense: 5′-UUC UCC GAA CGU GUC ACG UTT-3′; antisense: 5′-ACG UGA CAC GUU CGG AGA ATT-3′, Shanghai GenePharm Co., Ltd). At 4 h after transfection with FAM-siRNA, MnHAP/FAM-siRNA complexes and Lipofectamine 2000/ FAM-siRNA complexes, cells were harvested and analyzed by flow cytometry to determine the percentage of FAM-positive cells. A Nikon Eclipse TE300 fluorescent microscope (Nikon, Tokyo, Japan) was used for cell observations and recording images after washing cells 3 times with PBS to reduce background fluorescence.

A formazan-based assay (MTT assay, Sigma-Aldrich) was employed to determine the cytotoxicity of the MnHAP/siRNA mixture (weight ratio at 2:1). BEL-7402 cells were seeded in 96-well plates at a density of 10⁴/well with 100 μl culture medium. Cells were incubated for 24 h to allow complete adherence then the medium was replaced with fresh medium containing siRNA alone, MnHAP/siRNA complexes and Lipofectamine/siRNA complexes respectively; control wells received fresh medium. For each well, the amount of siRNA was 0.25 μg and the amount of MnHAP was 0.25 μg or 0.5 μl Lipofectamine. Four replicates of each condition were present and analyzed simultaneously.

To assess how treatments affected the cell number, medium in each well was aspirated after 4 h incubation and 100 μl fresh medium was added, followed by a further 24 h incubation after which time 10 μl MTT (5 mg/ml, MP Biomedicals, USA) was added to each well and the cells were again incubated in a CO₂ incubator for 4 h. The supernatant containing MTT substrate was aspirated and 100 μl DMSO added to dissolve the formazan deposits. The absorbance at 495 nm was measured on a spectrophotometer (Spectra Max M5, Molecular Devices), and the relative cell viability (%) was calculated as the percentage of absorbance of the treated cells sample relative to that of the untreated control cells.

BEL-7402 cells were harvested 48 h after siRNA transfection and lysed with lysis buffer for 30 min on ice. Total protein concentration was determined using a BCA protein assay. For western blotting, protein extracts were electrophoresed on SDS-PAGE gels and transferred to nitrocellulose membranes. After blocking with 5% non-fat milk powder, the membrane was first incubated overnight at 4°C with a mouse monoclonal antibody against GAPDH (Abcam). The blots were then washed and incubated with a goat anti-mouse IgG secondary antibody (Abcam). Bands were visualized using an ECL detection kit (Millipore) and intensities analyzed using Image J software. The relative expression of GAPDH was determined by dividing the densitometric value of the GAPDH band by that for its control (Tubulin).

All in vitro assays were performed at least twice independently, with 3 or 4 replicates of each data point. Student’s t-test was employed for statistical analysis using the SPSS 13.0 software package (SPSS Inc., Chicago, IL, USA).

To prevent nHAP aggregation and enhance transfection efficiency, nHAP was coated with branched PEI using published reaction conditions to generate MnHAP (28). The appearance of the newly formed MnHAP microemulsion was clear/milky and appearance was stable for more than 7 days without formation of precipitates, whereas nHAP alone sank quickly from the solution and formed aggregates. As seen using light microscopy, the particle dispersion of MnHAP was better than that of nHAP (Fig. 1A) and their average size was 184.5±20.5 nm (Fig. 1B). Surface charge is an important property of nanoparticles. When it exceeds ±30 mV, particles in suspension could be kept well-dispersed and stably suspended (21). The surface charge of nHAP without modification is −10.47 mV (data not shown), which implies that it is difficult for the dispersed particles to remain stable. In the present study, the measurement of surface zeta potential of MnHAP revealed a positive surface charge of +35.4 mV, which not only enables the particles to avoid agglomeration but also facilitates binding to the negatively charged siRNA to form MnHAP/siRNA complexes (Fig. 1C). Moreover, when imaged at ×1,5000 magnification using transmission electron microscopy (TEM) MnHAP particles were of uniform size and spherical with smooth surfaces (Fig. 1D).

To assess the siRNA binding efficiency of MnHAP, a gel retardation assay was carried out at six weight ratios of MnHAP to siRNA while maintaining the amount of siRNA. Fig. 2 illustrates that with the increase of MnHAP/ siRNA weight ratio, the migration of siRNA bands was significantly retarded. When the ratio of the MnHAP/siRNA was ≥2:1, no free siRNA could be detected, which indicated that the siRNA was completely confined to the sample well. Therefore, the 2:1 ratio could be the minimal ratio for siRNA being totally bound by MnHAP and this ratio was used in the subsequent experiments.

The in vitro transfection studies on BEL-7402 cells revealed that the proportion of FAM-positive cells increased remarkably following transfection with MnHAP/siRNA and Lipofectamine 2000/siRNA complexes compared to that with siRNA alone. According to the quantitative analysis shown in Fig. 3, the transfection efficiency of the group treated with FAM-labeled siRNA was <0.50% (0.48±0.02%), whereas the efficiency was increased significantly for the MnHAP/siRNA treatment group (96.49±0.34%), which is almost the same as that of Lipofectamine 2000/siRNA (94.42±0.50%). The FAM-positive cells in each group were observed and counted using fluorescence microscopy (Fig. 4A–C). The morphological features of the positive cells were confirmed and the positive cell percentages were in agreement with those from flow cytometry analyses.

MTT assays were performed on BEL-7402 cells to evaluate the cytotoxicity of MnHAP. The results showed that the cell viability following exposure to MnHAP/siRNA complex was 81.60±2.00%, whereas for the commercially available transfection agent Lipofectamine 2000/siRNA complex, the cell viability dropped to 64.40±1.69% (Fig. 5). As siRNA itself had no obvious cytotoxic effect on BEL-7402 cells, the cytotoxicity of MnHAP/ siRNA and Lipofectamine 2000/siRNA complexes were quite possibly due to MnHAP and Lipofectamine 2000 themselves. Even so, MnHAP was less toxic than Lipofectamine 2000 as demonstrated in Fig. 5.

The cytoxicity of MnHAP is similar to that of nHAP according to a previous study of BEL-7402 cells in which the cytotoxic effect was attributed to inhibition of proliferation and induction of apoptosis by nHAP (29). Therefore, our modification by PEI did not change the cytotoxicity of nHAP. However, in some studies, HAP without nano-processing was viewed as a non-toxic biocompatible biomaterial (5,30). It has been reported that HAP cytotoxicity seemed to be related to its various physical characteristics including size, shape, charge and surface area as well as to the cell lines used, and that the degree of cell toxicity is positively associated with the amount of HAP taken up into cells (31). Compared to micro-size HAP, nano-sized nHAP particles are much more easily taken up by cells. The intracellular nHAP results in the rapid release of calcium, and this is the main reason for nHAP cytotoxicity (32).

In this study, we found that MnHAP could bind siRNA to form nano-complexes capable of knocking down GAPDH expression in BEL-7402 cells in vitro. Fig. 6, shows western blot analysis using tubulin as an internal control; GAPDH protein levels were lower when MnHAP delivered siRNA-GAPDH compared to the control siRNA-NC. Lipofectamine 2000 was similarly effective. The strong gene silencing activity of MnHAP loaded with siRNA-GAPDH is probably due to the high stability of complexes in the culture medium.

In conclusion, we have successfully developed MnHAP, a novel transfection vector that can carry siRNA into BEL-7402 cells efficiently. This study shows that an MnHAP emulsion is stable for a longer time compared with that of nHAP, and this ensures its successful binding to siRNA for transfection. The high transfection efficiency and decreased cytotoxicity relative to Lipofectamine 2000 support that MnHAP may be a good candidate vector for biological and medical research. To further explore its biological properties and the potential medical applications, we will conduct further studies on the modification of MnHAP with ligands and on the delivery of other functional genes to target specific organs.