HepG2.2.15 cells, HepG2-N6 cells and pHBV1.3 plasmid were donated by Prof. Xiaoyong Zhang from the Department of Infectious Diseases, Nanfang Hospital. In particular, HepG2.2.15 cells, which can stably express HBsAg, HBV DNA and other substances in cell supernatant, are liver tumor cell lines of HepG2 cells with HBV genome integrated on their chromosomes (15), and HepG2-N6 cells are HepG2-derived cell lines that retain the characteristics of polarized hepatocytes, but display the morphology of a single columnar epithelium, enabling routine studies of virus transmission and release (16). In addition, HepG2-N6 cells highly express human NTCP (hNTCP) receptor, which is applicable for HBV targeting research. Mouse macrophage RAW 264.7 was donated by Prof. Longying Zha from the School of Public Health, Southern Medical University; HepG2 cells were kept in our laboratory; HBV/pMD18-T plasmid containing HBV DNA was constructed in our laboratory. HepG2 cells, HepG2-N6 cells, and RAW 264.7 cells were cultured in dulbecco’s modified eagle medium (DMEM) containing 10% serum, and HepG2.2.15 cells were cultured in 1640 medium containing 10% fetal bovine serum, non-essential amino acids, and G418 (200 μg/mL).

Six to eight weeks-old SPF grade C57BL/6J male mice were purchased from the Animal Experiment Centre of Southern Medical University and housed in a sterile SPF grade laminar flow chamber. The mice were given regular mouse chow and ad libitum under controlled conditions of temperature (20-25°C) and humidity (40-45%) using a 12:12 h light/dark cycle. All animal experiments were performed after receiving consent from Southern Medical University’s ethics review committee for animal experimentation.

The siRNA-X (1646-1664, GGUCUUACAUAAGAGGACU), siRNA-P (411-429, UCCUGCUGCUAUGCCUCAU), siRNA-C (2019-2039, AAGCCUUAGAGUCUCCUGAGC) against HBV X, P, and C genes respectively were selected by referring literature (17–20) and validated with cellular experiments, and synthesized by Guangzhou RiboBio, Co., Ltd.

The pHBV1.3 plasmid containing 1.3 times of the HBV genome was used for the construction of HBV infection models in vitro (21). To produce a cellular model of HBV infection, the pHBV1.3 plasmid was transfected into HepG2 cells using Lipofectamine 3000 (Gibco, Co., Ltd). siRNA was transfected simultaneously with the pHBV1.3 plasmid into HepG2 cells; HepG2.2.15 cells exclusively transfected with siRNA; the control group consisted of cells transfected with non-targeted control siRNA, and three replicate wells were set up for each cell line. After 72 h, the culture supernatants were collected and the levels of HBV DNA and HBsAg in the supernatants were measured. The HBV/pMD18-T plasmid containing HBV DNA was used as the standard of DNA levels. After fold dilution, quantitative real-time polymerase chain reaction (qPCR) was performed according to SYBR Premix Ex TaqTM II instructions (YEASEN, Co., Ltd). The lysis curve was the system default. The standard curve for HBV DNA was plotted using Ct values and the copy number was converted to international unit IU/mL. After centrifuging the supernatant, 50 μL was transferred to a new PCR tube, heated at 100°C for 2 min, then 1 μL was taken into the qPCR system (7), and the level of HBV DNA was calculated using the standard curve. After centrifugation to remove cell debris, the supernatant was obtained and HBsAg ELISA kits (Keygen Biotech, Co., Ltd) were used to detect HBV antigen levels according to the kit instructions.

The experimental approach described in reference (22) was partially improved to prepare PSN and its controls, PreS/2-21-modified nanoparticles (PN), siRNA nanoparticles (SN), and lipid-like nanoparticles (LLN). TT3 is an organic compound produced the stepwise reaction with addition of propylene diamine, di-tert-butyl dicarbonate, sodium bicarbonate, 1,3,5-benzenetricarbonyl chloride, pyridine, trifluoroacetic acid, ethyl acetate, triethylamine, dodecaldehyde, and triacetyl, and was synthesized by WuXi AppTec. 1,2-distearoyl-sn-glycero-3-phosphocholine (DSPC) was purchased from Shanghai yuanye Bio-Technology, Co., Ltd. 1,2-distearoyl-sn-glycero-3-phosphoethanolamine-N[maleimide(polyethyleneglycol)-2000] (Mal-PEG2000-DSPE) was purchased from Shanghai Aladdin Biochemical Technology, Co., Ltd. The target peptide HBV PreS/2-21 and the lipid were synthesized by ChinaPeptides, Co., Ltd. The sequence of the target peptide is: NH₂-GTNLSVPNPLGFFPDHQLDP-COOH, synthesized with a stearylation modification at the N-terminal end and a cysteine (Cys) coupled at the C-terminal end. PreS/2-21 was dissolved in PBS solution (pH=7.0), Mal-PEG2000-DSPE was dissolved in N, N-Dimethylformamide (DMF), the two were combined at a molar ratio of Mal-PEG2000-DSPE: PreS/2-21 = 1:1.2 and the dehydration-condensation reaction was carried out slowly at 4°C for 8 h. PreS/2-21-Mal-PEG2000-DSPE (PMD) was obtained by lyophilizing the reaction products, followed by re-solubilizing it with anhydrous ethanol. The extrusion approach was used to synthesize PSN by TT3, cholesterol, DSPC, and PMD in a molar ratio of 50:10:38.5:1.5 (23). The aforementioned ingredients were also partially or completely used to make PN, SN, and LLN ( Table 1 ). The above liposomal nanoparticle extrusions were concentrated using an ultrafiltration tube and decontaminated using a 0.22 μm filter membrane.

The concentrations of siRNA were measured using an ultra-micro UV spectrophotometer (Denovix, Co., Ltd) and the number of moles of siRNA was calculated from the volume. The encapsulation rate was calculated as EE% = (moles of siRNA before packaging - moles of siRNA after packaging)/moles of siRNA before packaging × 100%. PSN was naturally dried on aluminum foil and plated with gold, and then the ultrastructural morphology was observed by scanning electron microscope (Hitachi, Co., Ltd). The particle size and zeta potential of liposomal nanoparticles were measured by Zetasizer Nano (Malvern, Co., Ltd): the liposomal nanoparticles diluted with ddH₂O from 1 μL to 1000 μL were placed in a potentiometric cuvette and then assayed and scanned 100 times in duplicate. The liposomal nanoparticles were stored at 4°C analyzed for the particle size every 5 days, and monitored continuously for 30 days to examine the temporal stability. 10 μL of liposomal nanoparticles were added to 90 μL of PBS containing 10% FBS and shaken at 37°C at 100 rpm/min. The particle size was measured at 0, 3, 6, 12, 24, and 48 h to observe the serum stability. In addition, we monitored the temporal stability of activity by storing the liposomal nanoparticles at 4°C and analyzed the inhibition of HBV DNA by PSN using HepG2.2.15 cells once a week for a month.

HepG2.2.15, HepG2-N6 and RAW263.7 cells (1×10⁴ cells/well) were grown for 24 h in 96-well plates, then incubated with varying dosages of PSN for 48 h. To assess cytotoxicity, MTT was applied to the samples, and the absorbance values at OD 490 nm were measured to calculate the 50% cytotoxic concentration (CC50) of PSN on different cells. Experiments were performed using the cell models of HepG2.2.15, HepG2 transfected with pHBV1.3 (pHBV1.3-HepG2), and HepG2-N6 cells. Treatment groups consisted of five PSN concentration gradient subgroups, as well as a negative control and a blank control group, each with three replicate wells. The cell supernatant was collected after 72 h of incubation to detect the level of HBV DNA, the inhibition rate of PSN against HBV was calculated, and the half effective concentration (EC50) was estimated by GraphPad Prism 7 software (GraphPad Software, Inc., San Diego, CA, USA). Therapeutic Index (TI) was calculated by the ratio of CC50 against EC50. RAW264 cells were seeded on 6-well plates at a density of 2×10⁵ per well and co-cultured with 25 μg/mL PSN. The cells and culture supernatants were collected respectively after co-cultivation for 12 h, and the expression of cytokines IFN-α, TNF-α, and IL-6 was detected by reverse transcription qPCR (RT-qPCR) and ELISA kits (MULTI SCIENCES(LIANKE) BIOTECH, CO., LTD.).

First, the expression of NTCP receptors on the surface of HepG2 cells, HepG2-N6 cells were detected by flow cytometry, fluorescence microscopy, and western blot. The binding ability of PSN to NTCP receptors were detected by flow cytometry and fluorescence microscopy. HepG2 cells and HepG2-N6 cells were plated in six-well plates and incubated for 12 hours followed by a 30 min incubation with PBS or PSN (40 μL), then were stained with FITC-labeled NTCP antibody (ImmunoClone, IC03828F), and an equal amount of cells stained with FITC-labeled isotype control antibody was used as a negative control and analyzed by flow cytometry and fluorescence microscopy. HepG2-N6 cells and HepG2 cells were seeded in six-well plates. After 12-hours incubation, the cells were processed for total protein extraction, concentration determination, and western blot analysis. The protein concentration was measured with Pierce™ BCA Protein Assay Kit (ThermoFisher #23225) following instructions. Western blot was done according to standard methods. The expression level of β-actin was used as an internal control. The primary antibodies used in western blot were NTCP polyclonal antibody (Signalway Antibody Co., United States) and β-actin antibody (Fude Biological Technology Co., Hangzhou, China), and the secondary antibody was horseradish peroxidase-conjugated goat anti-rabbit IgG (Bioworld, United States). Furthermore, liposomal nanoparticles PSN-Cy3 and SN-Cy3 were prepared by Cy3-labeled siRNA-X, and the targeting of PSN-Cy3 to NTCP receptors was detected by HepG2-N6 cells, HepG2 cells and Hela cells. The Cy3-labelled siRNA-X was synthesized by Guangzhou RiboBio, Co., Ltd., and the liposomal nanoparticles were prepared as before, light protection throughout the process. Cells were inoculated into six-well plates and incubated for 12 h, then 40 μL PSN-Cy3 or SN-Cy3 was added and incubated for 30 min protected from light, while PBS-treated cells were used as a negative control. The fluorescence was measured by flow cytometry, observed under a fluorescence microscope and photographed.

Experiments were performed using the cell models of HepG2.2.15, pHBV1.3-HepG2, and HepG2-N6 cells. Among them, HepG2-N6 cell model was infected with HBV produced in HepG2.2.15 cell culture supernatant and concentrated using PEG8000, and the infection efficiency was verified by western blot. The experiments were divided into five groups: negative control, PSN, PN, SN, and LLN group. Approximately 12 h after the cells were inoculated onto the six-well plates, cells had attained 80% confluency, and 5 nM siRNA-containing nanoparticles or an equivalent amount of nanoparticles were introduced to the cells. Cell supernatants and cells were collected separately after 72 h of incubation, and the expression of HBxAg, pgRNA, HBV DNA and cccDNA were detected using RT-qPCR or qPCR. The primer sequences are shown in Table S1 . The levels of HBsAg and HBeAg in cell supernatants were detected by ELISA kits (Keygen Biotech, Co., Ltd). The cells were processed for protein extraction, concentration determination, and western blot analysis. Antibody to hepatitis B core antigen (anti-HBc, a gift from Prof. Xiaoyong Zhang, the Department of Infectious Medicine, Nanfang Hospital, China) and horseradish peroxidase-conjugated goat anti-rabbit IgG (Zhongshan Golden Bridge Biotechnology Co., Beijing, China) were incubated as described previously.

The recombinant adeno-associated virus AAV-HBV-002 applied to construct the hepatitis B model was purchased from PackGene Biotech, Co., Ltd. AAV-HBV-002 containing 1.3× HBV genome, is characterized by the production of HBV DNA, HBeAg and HBsAg, and the genotype is C2 and serotype is adr. Sixteen mice were each injected with AAV-HBV-002 into the tail vein at a dose of 1×10¹¹ vg. Another group of eight mice was injected with an equal volume of PBS as normal controls. Every four days, the mice’s mental condition, nutrition, and water consumption were assessed, and their body weight was measured and recorded. During collection every four days, a total of nearly 500 μL of orbital blood were collected and stored at -80°C. On fifteenth day, the expression of HBV DNA, HBeAg and HBsAg in serum was measured to determine whether the model had been successfully constructed. Figure 2 depicts the process of constructing a hepatitis B model in mice by injecting virus into the tail vein.

Fresh mouse blood was taken to prepare 2% erythrocyte suspension, to which different concentrations of PSN were added, and those with distilled water were used as the positive control group. After incubating at 37°C for 3 h, the colour of the supernatant was observed to monitor whether haemolysis had occurred. The 16 of mice who successfully constructed were randomly divided into PSN group and hepatitis B model group, and 8 untreated normal mice were selected as blank control group. The mice in the PSN group were given PSN and the other two groups were given saline in equal doses. In the PSN group, the effective dose of 5 mg/kg siRNA was administered every other day for 15 days, for a total of 8 doses. After completion of administration, the mice were anesthetized with 5% anhydrous ether, dissected, and perfused with PBS, then 1.5-2.0 mL of blood was collected. qPCR was used to detect the relative expression of HBV DNA (24). The primers and probe sequences were shown in Table S1 . The levels of antigens HBeAg and HBsAg, transaminases of ALT and AST were detected by ELISA kits (NanJing JianCheng Bioengineering Institute). The expression of IFN-α, TNF-α, and IL-6 was detected by RT-qPCR and ELISA kits. The important organs of the mouse, such as heart, liver, kidney, spleen, lung and brain, were removed and fixed in 4% paraformaldehyde solution, before being embedded, sectioned, and HE staining. The procedure for administration of PSN intervention to mice is shown in Figure 2 .

Six of 6-8 weeks C57BL/6J male mice were selected to construct a mouse hepatitis B virus model. After successful modelling, the mice were randomly divided into two groups with 3 mice in each group, and the experimental and control groups were treated with PSN-Cy7 and SN-Cy7, respectively, with a tail vein injection dose of 5 mg/kg. The fluorescence intensity in liver, kidney, lung, spleen and brain tissues was observed by a multi-modal small animal live imaging system at 1 and 3 h after injection.

All data reported are representative of at least three independent experiments. The experimental results were statistically evaluated using SPSS v19.0 software and were reported as mean ± standard deviation (Mean ± SD). Comparisons between two groups of data were made using the t-test for two independent samples, and the Shapior-Wilk test for normality and chi-square test was used between multiple groups of data, with the test level α = 0.05 (two-sided), and P<0.05 was considered a statistically significant difference.

The pHBV1.3-HepG2 and HepG2.2.15 cells were inoculated into six-well plates and the experiment was divided into five groups: PBS negative control group, siRNA-P, siRNA-C, siRNA-X, and siRNA-NC. As demonstrated in Figure 3 , three of the siRNAs reduced HBV expression with statistically significant differences. For the inhibition of HBV DNA in the supernatant of HepG2.2.15 cells, siRNA-X was more effective than siRNA-C; for the inhibition of HBsAg, siRNA-X was the most effective ( Figures 3A–D ).

The particle sizes of nanoparticles ranged from (83.85 ± 11.99) nm to (116.30 ± 8.05) nm ( Figures 4A, D ). The zeta potential of PSN was (-103.01 ± 2.61) mV at 37°C ( Figure 4B ). Under the scanning electron microscope, the PSNs were spherical, with smooth surface and uniform size distribution, and the diameters were mostly about 100 nm ( Figure 4C ). The siRNA concentration in the filtrate before and after encapsulation was measured to calculate the encapsulation rate. The results showed that PSN achieved an encapsulation rate of (84.75 ± 0.30) %, which was higher than the others. The encapsulation rates and particle sizes of the nanoparticles are shown in Figure 4D .

Liposomal nanoparticles were stored at 4°C and particle size was tested every 5 days. There were some fluctuations in particle size, but no fragmentation or fusion into large particles ( Figure 5A ). It was also mixed with 10% serum and incubated at 37°C for 48 h to test the stability in serum, showing little fluctuation in particle size and no fragmentation or fusion into large particles ( Figure 5B ). HepG2.2.15 cells were treated with PSN stored at 4°C for different periods of time, showing that more than 60% of HBV DNA inhibition was retained for at least 28 days, and the activity was relatively stable ( Figure 5C ).

We examined the CC50 and EC50 of PSN against three cell models and calculated the TI (CC50/EC50). The results ( Table 2 ) showed that the TIs of PSN to three kinds of cells were close to or exceeded 100, indicating that PSN can specifically inhibit HBV with less damage to host cells and maintain good cell safety. RAW264 cells treated with 25 μg/mL PSN for 12 h did not cause significant changes in the expression of cytokines IFN-α, TNF-α, and IL-6, either in the cells or culture supernatants ( Figure S1 ).

After labeling with NTCP-FITC antibody, the expression of NTCP on the surface of each cell was detected by flow cytometry. The results ( Figure 6A ) showed that the fluorescence intensity of HepG2-N6 cells was high, while HepG2-N6 cells was significantly lower after PSN treatment with statistically difference (P<0.0001) ( Figure 6G ). The fluorescence intensity of HepG2 cells incubated with NTCP-FITC antibody was lower (P<0.0001) ( Figure 6G ), and the fluorescence intensity of HepG2 cells did not change significantly after PSN treatment. The same results were obtained by fluorescence microscopy, as detailed in Figures S2A–C . The results indicate that HepG2-N6 cells have a high level of NTCP receptors on surface, and PSN specifically binds to NTCP and induces a reduction in its endocytosis. And the results ( Figure 6I ) by western blot analysis further indicated that HepG2-N6 cells have a high level of NTCP receptors.

The Cy3-labeled siRNA was used to quantify the targeting of PSN to various cells. Flow cytometry results showed that the fluorescence intensity of PSN-Cy3 in HepG2-N6 cells was significantly higher than that in HepG2 cells as well as Hela cells (P<0.0001, P<0.001) ( Figures 6B, C, H ); the fluorescence intensity in PSN-Cy3-treated HepG2-N6 cells was significantly higher than that in SN-Cy3 group (P<0.001) ( Figures 6D, H ); while the fluorescence intensity in PSN-Cy3-treated HepG2 cells and Hela cells had almost the same fluorescence intensity as the SN-Cy3 group ( Figures 6E, F ). Under the fluorescence microscope, PSN-Cy3 showed the highest brightness in HepG2-N6 cells, but weak fluorescent signal in HepG2 cells and Hela cells ( Figures S2D–I ). The results indicate that PSN has strong NTCP binding ability and can be endocytosed by HepG2-N6 cells with high NTCP expression, but less by HepG2 cells and Hela cells with low expression of NTCP receptor.

The inhibitory effect of liposomal nanoparticles on HBV was examined using HepG2.2.15 cells. Except for LLN, all liposomal nanoparticles had a certain inhibitory effect on HBV with statistically significant difference. The inhibition rate of PSN on pgRNA and HBV DNA was stronger, reaching 95.81% and 83.53%, respectively, and on HBxAg, cccDNA, HBeAg, and HBsAg was around 40% ( Figures S3A–F ). The inhibitory impact of PSN on HBV was further investigated using HepG2 cells transfected with pHBV1.3 plasmid as a model of HBV-infected cells, which showed similarly result ( Figures S3G–L ).

In addition, we detected HBV core antigen in HepG2-N6 cells infected with HBV by western blot, and the results ( Figure 7F ) showed that the core antigen in HBV-infected HepG2-N6 cells was significantly reduced after PSN treatment. We also tested the anti-HBV effect of PSN using HepG2-N6 cells before, during and after HBV infection respectively, and found that adding PSN before or at the time of HBV infection had a better pgRNA, HBeAg inhibition effect than after HBV infection, and the results were statistically different ( Figures 7 , S4 ). It showed that PreS/2-21 on the surface of nanoparticles could compete with HBV to bind NTCP receptors and function as a certain HBV entry inhibitor.

The results of the haemolytic assay showed no haemolysis in the PSN group, with all red blood cells sinking and the upper layer being a slightly cloudy yellow solution. In contrast, the distilled water group showed haemolysis and the solution was red ( Figure S5 ). After 15 days of PSN treatment (5 mg/kg, qod, 8 time), all mice were anaesthetized and dissected, and pathological sections were taken from liver, lung, heart, kidney, spleen and brain tissues for HE staining. During modeling and administration, the mental status and body weight of the mice did not fluctuate apparently ( Figure S6 ). The hepatocytes in the blank control group were normal in structure, with the nucleus in the centre of the cell and normal in size (left side of Figure 8A ). The liver tissue of the mice in the hepatitis B model group was diffusely edema, the liver cells were obviously enlarged with indistinct boundaries, and there were a large number of vacuoles (middle of Figure 8A ). The hepatocytes in the PSN group had clear borders and no evident pathological changes were observed (right side of Figure 8A ). The rest of the tissues in the hepatitis B model group and the PSN groups showed no abnormalities ( Figures 8B–F ).

After successful modelling, the experimental group was given PSN intervention for 15 days, while a model control group and a blank control group were set up and given an equal volume of saline. After 15 days of treatment, a high copy of HBV DNA could be detected in the hepatitis B model group at a concentration of (49.13 ± 2.68)×10⁵ IU/mL. In the PSN group, the level of HBV DNA was lower than that in the hepatitis B model group, at (37.13 ± 3.35)×10⁵ IU/mL ( Figure 9A ), with an inhibition rate of 24.43%. The relative expression of cccDNA was also lower in the PSN group compared to the hepatitis B model group ( Figure 9B ), and the inhibition rate of cccDNA by PSN was 28.93%. After PSN intervention, the expression of HBeAg decreased from (117.97 ± 38.14) PEI U/mL to (52.56 ± 17.55) PEI U/mL and the expression of HBsAg decreased from (112.38 ± 37.61) IU/mL to (68.79 ± 28.90) IU/mL with statistically significant difference ( Figures 9C, D ), and the inhibition rates of PSN on HBeAg and HBsAg were 55.45% and 38.79%, respectively.

After 15 days of PSN treatment, the expression of ALT and AST were lower than those in the hepatitis B model group and close to those in the blank control group ( Figure 10A ). The mRNA expression of cytokines IFN-α, TNF-α and IL-6 was significantly reduced after PSN treatment, which appeared to be closer to the levels of the control group ( Figure S7 ). Compared with the model group, the protein expression of IL-6 and IFN-α in the serum of mice in the PSN group decreased, among which IL-6 decreased with statistical difference (P < 0.01), while the expression level of TNF-α appeared to be relatively low among the three groups ( Figure 10B ).

After 1 h of injection of PSN-Cy7 and SN-Cy7, in vivo imaging of mice revealed that the fluorescence in the PSN group was concentrated in the liver and gastrointestinal tract, while in the SN group was concentrated in the gastrointestinal tract with only weak fluorescence in the liver ( Figures 11A, B ). After 3 h of injection, the fluorescence in the PSN group was concentrated in the gastrointestinal tract, while some mice in the control group only had fluorescence in the gastrointestinal tract or even no fluorescence in the body ( Figures 11C, D ).