Fig 1.
The FEN1 inhibitor, PTPD, reduces cccDNA production.
Effect of FEN1 inhibition on HBV-replicating cells. Hep38.7-Tet cells were treated with dimethylsulfoxide (DMSO) as a vehicle control, PTPD (5 μM), or 3TC (50 μM) in the absence of tetracycline for 5 days. At day 5, levels of HBV DNA, HBV RNA (pgRNA normalized by HPRT) and pre-C mRNA were analyzed. qPCR analysis of HBV DNA in (A) culture supernatant, (B) cytoplasmic NC-DNA, (C) cccDNA, and (D) pgRNA. Each result represents the mean ± SEM of three independent experiments. Asterisks indicate statistically significant differences; *P < 0.05, **P < 0.01, ***P < 0.001 compared with DMSO (A–D). (E) RT-PCR analysis for pre-C mRNA transcribed from cccDNA in Hep38.7-Tet cells. β-actin was used as a loading control.
Fig 2.
FEN1 siRNA knockdown and CRISPR/Cas9-mediated gene editing reduce cccDNA production.
(A–D) Hep38.7-Tet cells were transfected with FEN1-specific siRNA (siFEN1 #1 or #2) or control (siCtrl), and cultured without tetracycline. Four days after transfection, FEN1 mRNA/protein and HBV DNA were analyzed. (A) FEN1 mRNA quantified by RT-qPCR (normalized by HPRT) (upper panel) and Western blotting of FEN1 protein (lower panel). GAPDH expression is shown as a loading control. FEN1 protein expression gives rise to two bands in our study, which may be due to post-translational modification. (B–C) Levels of HBV DNA in cytoplasmic nucleocapsid (NC) (B) and cccDNA (C). (D) Efficiency of cccDNA formation (cccDNA levels normalized by cytoplasmic NC-DNA) was calculated. Each result represents the mean ± SEM of three independent experiments. Asterisks indicate statistically significant differences; *P < 0.05, **P < 0.01, ***P < 0.001 compared with siCtrl. (E–G) The pX330-FEN1 plasmid, which expresses Cas9 mRNA and sgRNA for exon 2 of the FEN1 gene, was transfected with a blasticidin-resistant gene expression vector into Hep38.7-Tet cells. After drug selection, two resistant clones (#1 and #2) were analyzed. (E) RT-qPCR and Western blotting analyses of FEN1 expression levels. (F) Efficiency of cccDNA formation. Each result represents the mean ± SEM of three independent experiments. Asterisks indicate statistically significant differences; *P < 0.05, **P < 0.01 compared with parental groups. (G) Southern blot analysis of cytoplasmic NC-DNA and Hirt-extracted HBV DNA. Hirt-extracted HBV DNA was analyzed with or without heat treatment and subsequent EcoRI digestion. Densitometric analysis of the EcoRI-digested cccDNA signal is shown below the Southern blot images. The signal for parental cell line is taken as 100%.
Fig 3.
PTPD reduces HBV replication in HBV infection models.
(A) HepG2-hNTCP-C4 cells were pretreated with PTPD (5 μM) for 1 day, and subsequently infected with HBV in PTPD-containing culture medium. cccDNA levels at 3 dpi were analyzed by Southern blotting. (B) PXB cells (primary human hepatocytes) were pretreated with PTPD (5 μM) or 3TC (50 μM) from 1 day before infection to 7 dpi. Secreted HBV DNA levels were quantified at the indicated time points (left side). HBV RNA was quantified by RT-qPCR at 21 dpi (right side). Asterisks indicate statistically significant differences; *P < 0.05 compared with DMSO.
Fig 4.
Requirement of nuclease activity and the C-terminus of FEN1 for cccDNA production.
(A) Schematic presentation of FEN1 protein. D181A: nuclease-deficient mutant. ΔC: deletion mutant unable to bind WRN protein. (B) FEN assay. Flap endonuclease activity of immunoprecipitated FEN1 protein (wt, D181A, or ΔC) was determined as in Fig 1A. (C) pResQ lentiviral vectors carrying FEN1 shRNA and the FEN1 transgene (wt, D181A, or ΔC, see also S8 Fig) were transduced into Hep38.7-Tet cells. After puromycin selection, NC-DNA was analyzed by Southern blotting and qPCR, respectively. Each result represents the mean ± SEM of three independent experiments. Asterisks indicate statistically significant differences; ****P < 0.0001, **P < 0.01 compared with Mock (B), *P < 0.05 compared with shCtrl (C). (D) Two clones of the FEN1+/− Hep38.7-Tet cells (#1 and #2 in Fig 2) were transduced with pResQ lentiviral vectors carrying FEN1 shRNA. wt, shCtrl. Hep38.7-Tet cells were transduced with the pResQ mock vector. After puromycin selection, FEN1 protein, cytoplasmic NC-DNA, and Hirt-extracted HBV DNA were analyzed by Southern blotting. Hirt-extracted HBV DNA was analyzed with or without heat treatment and subsequent EcoRI digestion. Densitometric analysis of the EcoRI-digested cccDNA signal is shown below the Southern blot images. The signal for the FEN1 wt cell line with control shRNA is considered 100%.
Fig 5.
Deletion of the C-terminus disrupts the nuclear localization and reduces HBV DNA association of FEN1 protein.
(A) Expression vectors of FEN1-GFP, FEN1ΔC-GFP, or mock vector (pcDNA4/TO) were transfected into Hep38.7-Tet cells. The nucleus was visualized by co-transfection of the nuclear localization signal (NLS)-tagged DsRed vector. (B–C) Myc-FEN1 or Myc-FEN1ΔC vector was transfected into Hep38.7-Tet cells. (B) Myc-tagged protein expression (before cross linkage) shown by Western blot. Two blots with different protein loadings are shown. (C) Myc-FEN1-transfected Hep38.7-Tet cells were cross-linked, and the lysates were immunoprecipitated with either control IgG or anti-Myc antibody. The immunoprecipitants were subjected to qPCR analysis using a primer pair to detect the core region of HBV. Each result represents the mean ± SEM of three independent experiments. Asterisks indicate statistically significant differences; ***P < 0.001.
Fig 6.
FEN1 protein facilitates cccDNA formation in vitro.
(A) Schematic presentation of in vitro cccDNA formation assay. Purified NC-DNA (108 copies) was incubated with recombinant FEN1, Bst DNA polymerase, and Taq DNA ligase. Following incubation, the DNA was purified and analyzed (B–F). Regions for qPCR amplification (E and F) were indicated as p. The 5.4-kb PstI fragment in HBV plasmid (Control) has a partial HBV sequence but does not have core and intact P genes. (B) cccDNA-selective qPCR. Each result represents the mean ± SEM of three independent experiments. Asterisks indicate statistically significant differences; ***P < 0.001 compared with negative control (no enzyme). (C) Southern blot analysis of in vitro cccDNA formation assay. The DNA was analyzed directly (left), or treated with T5 exonuclease (middle) or T5 exonuclease, then subsequently digested with EcoRI (right). (D) Detection of RCA products. DNA treated with indicated enzymes was subjected to RCA and then digested with indicated restriction enzymes. Arrow indicates the 5.4-kb fragment of the HBV plasmid, used as a replication-defective control for transfection in E and F. (E and F) Equal amounts of digested RCA product generated in (D) were self-circularized and then transfected into HepG2 cells. Three days after transfection, HBV DNA was analyzed by qPCR. Each result represents the mean ± SEM of three independent experiments. Asterisks indicate statistically significant differences compared with the control; * P < 0.05, ** P < 0.01.