CRISPR/Cas9 cleavage of viral DNA efficiently suppresses hepatitis B virus

Abstract

Chronic hepatitis B virus (HBV) infection is prevalent, deadly, and seldom cured due to the persistence of viral episomal DNA (cccDNA) in infected cells. Newly developed genome engineering tools may offer the ability to directly cleave viral DNA, thereby promoting viral clearance. Here, we show that the CRISPR/Cas9 system can specifically target and cleave conserved regions in the HBV genome, resulting in robust suppression of viral gene expression and replication. Upon sustained expression of Cas9 and appropriately chosen guide RNAs, we demonstrate cleavage of cccDNA by Cas9 and a dramatic reduction in both cccDNA and other parameters of viral gene expression and replication. Thus, we show that directly targeting viral episomal DNA is a novel therapeutic approach to control the virus and possibly cure patients.

Figure 1. Transiently transfected CRISPR constructs exhibit anti-HBV activity.

(a) Schematic of HBV life cycle and putative anti-HBV effect of CRISPR constructs; Cas9-mediated DSB formation should linearize the small, episomal cccDNA repeatedly, potentially leading to indel formation (generating less-fit viral mutants) or even degradation. (b) (left) HBV genome organization and location of target sequences for several tested guide RNA constructs. (right) Table of all possible CRISPR target sites in each HBV ORF, including number of possible target sites in conserved genomic regions. (c) Experimental schematic for (d-e): HepG2 cells are co-transfected with 1.3x WT HBV and sgRNA/Cas9-2A-mCherry construct, and (d) intracellular HBV pregenomic RNA and (e) secreted HBsAg are quantified after 72 hours. Data shown were generated in one representative experiment, with intracellular pgRNA harvested from one pellet and HBsAg collected from replicate wells per group; all data are consistent across three independent experiments. (f) Experimental schematic for (g-h): 1.3x WT HBV and sgRNA/Cas9-2A-mCherry are delivered to the livers of immunodeficient NRG mice via hydrodynamic injection, and (g) HBsAg and (h) secreted HBV titer are quantified in mouse blood at 2 and 4 days post injection. 21M: guide RNA with 5 bp mismatch from g21. Data shown are from one representative experiment, and consistent across multiple experiments. UT: ‘untargeted’ guide RNA (no target sequence in HBV genome). *p < 0.05 for selected comparison; ^**p < 0.01 for selected comparison; ***p < 0.001 for selected comparison as assessed by two-tailed t-test.

Figure 2. Sustained expression of CRISPR machinery enables large reductions in HBV DNA and cccDNA.

(a) Schematic of lentiviral vector and experimental strategy for sustained CRISPR expression. (b-c) CRISPR constructs targeting HBV cause progressive reduction in (b) cccDNA and (c) total HBV DNA levels dependent on successful targeting of viral DNA and Cas9 nuclease activity; data shown are from one representative experiment pooled across 3 separate HBV-targeting guides (sg6, sg17, sg21), and consistent across multiple independent transduction experiments. (d) Southern blot of HBV DNA forms using Hirt’s extraction (to deplete high-molecular weight DNA), shows HBV-targeted sgRNAs with nuclease-active Cas9 generate near-total reduction in cccDNA. (e) Surveyor assay to detect indel formation in total HBV DNA (top) and episomal HBV DNA, enriched by treatment with plasmid-safe DNase (bottom); lentiviral transduction enables high levels of cutting of HBV. Arrowheads depict surveyor digestion products. Expected PCR product sizes for sg6, sg17 and sg21 are respectively 599, 946 and 507 bp. Approximate sizes of surveyor digestion products for sg6, sg17 and sg21 are respectively: 429 + 170, 570 + 376, 275 + 232. (f) Immunofluorescent imaging of HBV Core protein demonstrates large reduction in Core staining upon targeting by sg17 specifically against the Core ORF. (g-h) Cas9/gRNA-transduced Hep-NTCP cells are cocultured with HepG2.2.15 cells to infect them with HBV followed by depletion of HepG2.2.15 cells using puromycin selection (Schematic in Fig S6 left). (g) From left to right, HBsAg secretion, cccDNA copies, levels of HBV 3.5kb RNA relative to 5 bp mismatch control, and titer of HBV DNA in culture medium show that Cas9/sg17 reduce HBV infection in de novo infection. 17M: 5 bp mismatch control. 17D: dead Cas9 with g17. Data shown are from one representative experiment, and consistent across experiments. (h) Surveyor assay performed on DNA untreated (left) or treated (right) with Plasmid-Safe DNase to remove non-episomal viral forms. Arrowheads indicate indel formation. (b-c) ^*p < 0.05 for selected comparison; ^**p < 0.01 for selected comparison, as assessed by one-way ANOVA with Tukey post-hoc test.