The pHBV1.3 plasmid (genotype D, GenBank accession number V01460.1) was constructed as previously reported (He et al., 2016). The plasmid expressing N-terminally hemagglutinin (HA)-tagged-IFI27 fragment was PCR amplified and inserted into EcoRI and Xhol sites of the pCAGGS vector. The construction of the reporter plasmids pGL3-SP1-Luc, pGL3-SP2-Luc, pGL3-EnhII/Cp-Luc, and pGL3-EnhI/Xp-Luc was followed as reported previously (Hao et al., 2015). Two IFI27 short hairpin RNAs (shRNAs) and control or non-targeting shRNA (shControl) containing enhanced green fluorescence protein (eGFP) were inserted into pLKO-1-puro. The targets for the shRNAs are as follows: shControl: 5′-GCAGAAGAACGGC ATCAAG-3′, ShIFI27-4: 5′-CTCCGGATTGACCAAGTTCAT-3′, and ShIFI27-5: 5′-CCCTGCAGAGAAGAGAACCAT-3′.

All animal experiments were performed following the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. Mice were housed under specific pathogen-free (SPF) conditions in individually ventilated cages. The protocol was approved by the Institutional Animal Care and Use Committee of Wuhan University (project license WDSKY0201302).

HepG2, Huh7 cells, HepG2.2.15, and human embryonic kidney 293T (HEK-293T) cells were cultivated in Dulbecco’s modified Eagle’s medium (complete DMEM) comprising 10% fetal bovine serum (FBS), 100 U/ml penicillin, and 100 mg/ml streptomycin. Transfection of cells was carried out with Lipofectamine 3000 (Invitrogen) according to the manufacturer’s instructions. The cells were retained at 37°C in a humidified 5% CO₂ atmosphere.

For the generation of HepG2 cells in which IFI27 was knocked out (IFI27-KO), the gene-editing technology CRISPR/Cas9 was used. The gRNA targeting Cas9 to IFI27 gene was cloned. HEK-293T cells were co-transfected with Lenti-IFI27-gRNA-CRISPR-Cas9 plasmid together with the packaging vector pMD2.G and psPAX using Neofect reagent (Neofect Biotech, Beijing, China), and the lentiviruses were harvested after 48 and 72 h. These lentiviral particles were then transduced into HepG2 cells in the presence of polybrene (8 μg/ml), with the subsequent selection of puromycin (2 μg/ml). Cells were plated in a 96-well plate at ∼1 cell per well to get a single clone. Western blotting for individual clones was used to detect the expression of IFI27. KO cells were also confirmed by sequencing the targeted loci. The sequence of gRNA is as follows: 5′-CTCTGCCGTAGTTTTGCCCC-3′.

The extraction technique of HBV DNA from intracellular core particles was adopted from a previously described method with a few modifications (He et al., 2016). In brief, HepG2 cells were co-transfected with pHBV1.3 and pSV-β-gal along with pCAGGS-HA-IFI27 or its control vector (pCAGGS) in the quantities as shown in the related figures. The HepG2 cells were lysed in NP-40 lysis buffer [50 mM Tris–HCl (pH 7.0) and 0.5% NP-40] at 4°C after 96 h post-transfection and then centrifuged at 13,000 rpm. The supernatants were collected and digested with RNase A and DNase I (Thermo Fisher Scientific) in the presence of DNase I buffer at 37°C for 2 h, DNase I was inactivated subsequently at 65°C for 20 min in the presence of 10 mM EDTA, and then proteinase K was used to digest the protein along with 1% SDS overnight at 55°C. Finally, the digested samples were extracted with phenol:chloroform. After that, DNA samples were precipitated with ethanol and then resolved in 30 μl Tris-EDTA (TE) buffer. The viral DNA was then subjected to qPCR. The extracellular encapsidated DNA of HBV from supernatant or sera was extracted following a previously described protocol (Tian et al., 2011; Hao et al., 2015). The HBV replicative intermediate and extracellular HBV DNA were then analyzed by qPCR using primers as mentioned in Table 1.

HepG2 cells were lysed in cold lysis buffer [50 mM Tris–HCl, 100 mM NaCl (pH 8.0), 5 mM EDTA, 1% SDS) and a cocktail (protease inhibitor) was added. The samples were separated on SDS polyacrylamide gel electrophoresis (PAGE) and then transferred to nitrocellulose membrane. The membrane was incubated for 1 h in 5% non-fat dried skim milk to block non-specific binding. Following the membrane incubation at 4°C with target-specific primary antibodies such as rabbit anti-IFI27 (Biorbyt, Catalog number: orb337785) or anti-Ha-IFI27 (Abcam Trading Company, Shanghai, China) overnight. The membrane was incubated with a secondary antibody for 1 h and the result was normalized to reference internal control β-actin (ABclonal, Woburn, MA, United States). The signal of the western blot was observed with an enhanced chemiluminescence (ECL) substrate (Millipore, Billerica, MA, United States).

The total RNA from HepG2 cells and tissue from mice liver was isolated using TRIzol reagent (Invitrogen, Carlsbad, CA, United States). The cDNA was synthesized using Reverse Transcription Kit (Takara Biomedical Technology). PCR reactions were prepared using SYBR Green Fast qPCR Master Mix Kit (Yeasen Biotechnology, Shanghai). The mRNA of the housekeeping glycolysis gene glyceraldehyde-3-phosphate dehydrogenase (GAPDH) functioned as an endogenous standard. The ΔΔCt method was used and all the primers used in qRT-PCR are mentioned in Table 1.

To perform the northern blot, 4 μg of RNA samples were taken and then resolved on MOPS-buffered 1.5% agarose gel comprising 2.2 M formaldehyde and transferred to a nylon membrane (GE Healthcare). The hybridization was done adopting the manufacturer’s instructions containing the DIG-labeled probe. The probe was generated with a DIG probe synthesis kit (Roche, Germany). The rRNAs (28S and 18S) act as an internal control to detect the quantity of total RNA.

The levels of HBsAg and HBeAg antigen in culture supernatants of transfected HepG2 cells or mice serum were assessed using an enzyme-linked immunosorbent assay (ELISA) according to the manufacturer’s protocol (Kehua District, Shanghai). The activity of β-galactosidase was used to normalize the values in cell lysates and measured by a Beta-Glo kit (Promega).

HepG2 cells plated in a 24-well plate were transfected with HBV reporter plasmids (200 ng) along with an indicated amount of pCAGGS-HA-IFI27 plasmid (250 ng) or pCAGGS (control vector), and the pRL-TK plasmid (50 ng) was used for the control of transfection efficiency. At 48 h post-transfection, the cells were lysed and subjected to luciferase activity assay using the Dual-Glo system (Promega, Madison, WI, United States).

For HBV promoter analysis, chromatin immunoprecipitation (ChIP) assay was performed using the standard protocol of the ChIP Assay kit (Beyotime, Biotech, Catalog number: P2078). Briefly, HepG2 cells were grown to confluency in 10-cm dishes and transfected with HBV1.3 with IFI27 or control vector (pCAGGS). At 72 h post-transfection, the cells were cross-linked with formaldehyde for 10 min at 37°C, followed by neutralization using 125 mM glycine. The cells were then lysed with cold lysis buffer (50 mM Tris–HCl, 150 mM NaCl, pH 8.0, 10 mM EDTA, 1% SDS) in combination with a cocktail (protein inhibitor). Then, the lysates were sonicated by three pulses for 15 s on ice to break the genome into 200–1,000 bp in size. The antibodies used for immunoprecipitation were the anti-HA antibody (Abcam, Catalog number: ab9110) at 4°C overnight. Normal mouse IgG and anti-RNA polymerase II were used as negative and positive controls, respectively. The precipitate was then washed, and immunoprecipitated DNA fragments were extracted with phenol–chloroform extraction. The immunoprecipitated DNA fragments were amplified with PCR using EnhII/Cp (nt 1715–1815) primers shown in Table 1.

To test HBV replication in vivo, 6–8-week-old male mice (C57BL/6) were used. We randomly divided a total of 10 mice into two groups (five mice each). The replication-competent vector pHBV1.3 (10 μg) and pSV-β-gal (5 μg) were hydrodynamically co-delivered together with pCAGGS-HA-IFI27 expression constructs or empty vector pCAGGS (20 μg) into the tail vein of mice within 8 s in a volume of saline equivalent to 10% of the mouse body weight. These mice were sacrificed after 4 days post-injection and mice livers and blood were processed for analysis. The mice sera were collected and examined for analysis of HBsAg, HBeAg, and HBV DNA. Western blot was performed with a small slice of mice liver to detect IFI27 and HA-tag protein. For HBV RNA analysis, a small piece of liver tissue was homogenized in TRIzol reagent to purify and extract total RNA. Liver tissues were also collected to analyze HBV core antigen expression by using immunohistochemical staining.

All experiments were repeated at least three times. The results are presented as means ± SD unless stated otherwise. The statistical significant differences were determined by using one-way ANOVA analysis with multiple comparison test and independent Student’s t-test. Statistical analyses were achieved using the Prism 8 software (GraphPad Software Inc., San Diego, CA, United States). A P ≤ 0.05 was considered statistically significant.

IFI27 or ISG12a is one of the most highly induced genes following treatment of cells with type I IFNs (Rosebeck and Leaman, 2008). To confirm this speculation, we cultivated HepG2 and Huh7 cells separately in a 12-well plate for 24 h and then treated them with IFN-α (100 ng/ml) in a time-dependent manner (0, 6, and 12 h). The samples were collected at indicated time points. The protein levels of IFI27 in both types of cell lines were examined with western blotting (Figures 1A,B). The total RNA was extracted from the cells, and the levels of mRNA of IFI27 were determined by qRT-PCR (Figures 1C,D). As expected, IFI27 was significantly upregulated in HepG2 and Huh7 following IFN-α treatment. The results revealed that IFI27 is one of the endogenous IFN-α-inducible genes in liver cells and may be involved in the molecular mechanism of IFN-α-mediated restriction of HBV.

Next, to judge the physiological role of IFI27 in IFN-elicited anti-HBV response, we use HepG2.2.15 cells, which are derived from the human hepatoblastoma cell line HepG2 and are characterized by having stable HBV expression and replication in the culture system (Sells et al., 1987). As a cell source, HepG2.2.15 cells can stably express HBV and these cells have been frequently used in studies of HBV infection assays (Zhao et al., 2011). The HepG2.2.15 cells were transfected with pCAGGS-HA-IFI27 plasmid or empty vector. After 24 h of incubation, the cells were treated with 100 ng/ml IFN-α. The expression levels of IFI27 mRNA were analyzed by qRT-PCR (Figure 1E). The total HBV mRNAs were efficiently reduced by IFN-α-treated cells as revealed by qRT-PCR (Figure 1F). As a result of the upregulation of IFI27, HBV intracellular DNA expression levels were also reduced as determined by qPCR (Figure 1G). These results indicate that following successful transfection, IFI27 potentiated the anti-HBV activity upon IFN-α treatment in HepG2.2.15 cells.

The inhibitory effect of IFI27 on HBV replication has not been described, and we initially examined the role of IFI27 in HBV gene expression and replication. HepG2 cells were co-transfected with an HBV plasmid (pHBV1.3) and pSV-β-gal along with different doses of IFI27 expression vector (pCAGGS-HA-IFI27). The expression levels of IFI27 mRNA and protein were analyzed by qRT-PCR and western blotting, respectively (Figures 2A,B). The HBV RNA transcripts (3.5, 2.4, and 2.1 kb) were subjected to northern blot. The 3.5-kb transcript containing the pgRNA and the longer precore mRNAs were decreased remarkably upon IFI27 overexpression (Figure 2C). Besides, with the reduction of HBV RNAs, qRT-PCR indicated that the overexpression of IFI27 decreases the levels of HBV 3.2 kb mRNA significantly (Figure 2D). We observe the inhibitory effect of IFI27 against HBV replication by further measuring HBsAg and HBeAg in cell culture supernatant using ELISA (Figures 2E,F). We also illustrate the suppressive role of IFI27 on HBV gene expression and replication, and the core-associated HBV DNA in transfected HepG2 cells and HBV DNA (secreted) in the culture medium were extracted and observe by qPCR (Figures 2G,H). Both HBV intracellular core-associated and extracellular DNAs were highly restricted upon overexpression. Hence, our data suggested that IFI27 plays an inhibitory role in HBV gene expression and replication.

To validate further the inhibition role of IFI27 on HBV gene expression, we performed gene knockdown or silencing in HepG2 cells. For knockdown, the IFI27 expression in HepG2 cells was downregulated by transducing with shRNA-lentiviral vectors such as shIFI27-4 and shIFI27-5 or scramble control (shControl), targeting IFI27. HepG2 cells were transfected with pHBV1.3 and pSV-β-gal plasmids and incubated for 24 h. After 24 h, the medium was discarded and the cells were transduced with shIFI27-4 and shIFI27-5 or scramble control (shGFP) and incubated for further 48 h. The two independent shRNAs, shIFI27-4 and shIFI27-5, showed strong IFI27 knockdown effects in the cells. The levels of IFI27 mRNA were examined by qRT-PCR (Figure 3A), and western blotting confirmed the expression of IFI27 protein (Figure 3B). Both mRNA and protein expression levels of IFI27 were significantly suppressed as compared with the scramble control, which indicates that shIFI27-4 and shIFI27-4 are effective. The HBV transcripts (3.5, 2.4, and 2.1 kb) in transduced HepG2 cells were evaluated through northern blot (Figure 3C). Here, we also investigated whether IFI27 affects the level of HBV 3.2 kb mRNA, and the qRT-PCR result revealed that the HBV 3.2-kb mRNA levels increased significantly upon silencing of IFI27 (Figure 3D). Concordantly, knockdown of IFI27 in HepG2 cells also enhanced the secretion of HBsAg and HBeAg proteins in the culture supernatant, as confirmed by ELISA (Figures 3E,F). As a result of the downregulation of IFI27, the levels of HBV intracellular core-associated and extracellular HBV DNA were also increased, as determined by qPCR (Figures 3G,H). Overall, these results suggested that the silencing of IFI27 significantly enhanced HBV gene expression in hepatocytes.

Based on the results of overexpression and silencing of IFI27 on replication of HBV, we further confirm our results by using CRISPR/Cas9 gene-editing technology to generate a stable HepG2 cell line in which the IFI27 gene was completely knocked out (HepG2 IFI27-KO). The HBV replication-competent plasmid (pHBV1.3) was transfected into HepG2 (wild-type) and HepG2 IFI27-KO cells. The suppression level of IFI27 protein was analyzed with western blotting (Figure 4A). These results suggested that the level of IFI27 protein expression was significantly reduced if we compare it with wild-type (parent cells), which determines the efficiency of knocking out of IFI27 in HepG2 cells. The northern blot result revealed that IFI27-KO HepG2 cells highly increased the level of HBV transcripts (2.5, 2.4, and 2.1 kb) (Figure 4B). In addition, the HBV 3.2-kb mRNA in HepG2 IFI27-KO cells more potently increased as revealed by qRT-PCR (Figure 4C). With the enhancement of HBV RNAs, as compared with WT cells, secreted HBsAg and HBeAg protein levels in the supernatant were efficiently enhanced in IFI27-KO HepG2 cells (Figures 4D,E). Our result showed that the level of HBV intracellular core-associated and extracellular DNA of HBV in supernatant was also increased in HepG2 IFI27-KO cells relative to parent cells as revealed by qPCR (Figures 4F,G). Collectively, these findings demonstrate that IFI27 knockout in hepatocytes increased HBV gene expression and replication.

To examine the mechanism of IFI27-mediated HBV repression, the role of IFI27 in the regulation of HBV promoters were assessed. HepG2 cells were co-transfected with reporter plasmids pGL3-EnhI/Xp, pGL3-EnhII/Cp, pGL3-SP1, and pGL3-SP2 along with pCAGGS–IFI27 or pCAGGS. The results from the luciferase-based reporter assay revealed that EnhII/Cp activity was suppressed by IFI27, whereas IFI27 did not affect the activities of the EnhI/Xp, SP1, and SP2 promoters (Figure 5A). To confirm the suppressive effect of IFI27 in the inhibition of HBV EnhII/Cp promoter, HepG2 cells were co-transfected with pGL3-EnhII/Cp together with IFI27 expression plasmid or control plasmid at different concentrations. According to the reporter assay, the activity of EnhII/Cp was inhibited by IFI27 in a dose-dependent manner (Figure 5B). To further validate the specific inhibitory effect of IFI27 on HBV EnhII/Cp, we replaced EnhII/Cp with pCMV (cytomegalovirus promoter). HepG2 cells were co-transiently transfected with pGL3-pCMV plasmid along with IFI27 expression plasmid, or control plasmids were subjected to dual-luciferase activity. Importantly, IFI27 did not affect the activity of the pCMV promoter (Figure 5C). This result indicated that IFI27 is solely responsible for the inhibition of EnhII/Cp. We further examined the mechanisms of IFI27 inhibition on the HBV EnhII/Cp promoter and identified which region is targeted by IFI27. A series of EnhII/Cp deletion mutants were constructed and then these constructs were subcloned into the pGL3 basic vector to generate five reporters (Figure 5D). The deletion region (nt 1715–1815) of EnhII/Cp is reduced potently by IFI27. Hence, the region from nt 1715 to 1815 of HBV genome could be responsible for the suppressive effect of IFI27 (Figure 5E). Then, the study conducted a ChIP assay to investigate whether IFI27 could bind to the region from nt 1715 to 1815 in HBV EnhII/Cp promoter. Fragments of HBV EnhII/Cp promoter were detected in anti-HA-IFI27 antibody immunoprecipitated candidates in HepG2 cells transfected with pHBV1.3 plasmids, but were not detected in cells transfected with control vector (Figure 5F). Taken together, these data indicated that IFI27 binds to the EnhII/Cp region and displays an inhibitory effect on HBV replication and transcription.

Because IFI27 was shown to restrict HBV gene expression and replication in HepG2 cells, we also investigated the role of IFI27 expression on HBV replication in the mouse model system in vivo. C57BL/6 male mice were co-delivered with pHBV1.3 and pSV-β-gal, together with pCAGGS-HA-IFI27 plasmids or control vector (pCAGGS) through hydrodynamic injection and kept for 4 days. After scarifying all the mice, the effect of IFI27 on HBV replication was assessed. The IFI27 expression level was detected with western blot. The result showed a high expression of IFI27 protein level in mice liver (Figure 6A). In the sera of treated mice, HBsAg, HBeAg, and HBV DNA were tested. The levels of HBsAg and HBeAg were measured by ELISA (Figures 6B,C) and HBV DNA was evaluated through qPCR (Figure 6D). These data suggested that the expression of HBsAg and HBeAg and HBV DNA were remarkably diminished via IFI27 expression. To further validate HBV transcription in vivo, HBV 3.5 kb mRNA levels were determined by qRT-PCR (Figure 6E), which was significantly suppressed by IFI27. Moreover, the immunohistochemical (IHC) staining of the mice liver tissues reveals that the HBV core proteins (HBc) were reduced effectively in the presence of IFI27 (Figure 6F). Together, these data indicated that IFI27 has an inhibitory effect on HBV transcription and replication in vivo.