Discussion

In this work, for the first time, we assessed how HBV rcDNA affects antiviral and, specifically, cccDNA-targeting activity of major APOBEC/AID enzymes. Experiments with transcriptionally silenced cccDNA revealed markedly increased deamination of cccDNA by A3A, A3B, and AID (Figure 1). This prompted us to perform experiments with modern antiviral therapeutics that block or terminate HBV rcDNA production, namely lamivudine and siRNA targeting X gene of HBV. These experiments revealed increased deamination of the remaining rcDNA and, more importantly, increased cccDNA deamination (Figure 2, 3) with massive deamination demonstrated by NGS analysis (Figure 4). These observations suggest a new, background defense strategy utilized by the virus to counter innate immunity and to preserve cccDNA integrity by saturating antiviral enzymes with its genome intermediate, HBV rcDNA. This unique mechanism of immune evasion is somewhat similar to what HBV does to dampen adaptive immunity in CHB patients, when HBV produces excessive amounts of HBsAg compared to virions to establish antigen overload that exhausts antigen-presenting cells and leads to immune tolerance57,58. In this particular “hide-abide” strategy, HBV produces abundant HBV rcDNA, the single-stranded regions of which are the primary targets for APOBEC/AID cytidine deaminases. Overloading and saturating APOBEC/AID with ssDNA targets minimizes the effect of these enzymes on cccDNA, contributing to viral persistence. To the best of our knowledge, this is a newly characterized mechanism of HBV immune evasion that has not been reported previously for HBV or other viruses, adding to the already complex interaction of viruses with the host immune system.
Our results also discourage from further use of APOBEC/AID as potential anti-HBV therapeutics. APOBEC/AID enzymes have long been known to be important drivers of human cancers. Widespread APOBEC/AID-mediated mutations in genomic DNA of cancer patients have been detected56. More recent evidence adds knowledge about RNA-targeting properties of some APOBEC/AID enzymes dysregulating the host cell functioning59 and inducing death in animals with, specifically, A3B overexpression20. The promise of APOBEC/AID as safe molecular tools came from the pioneering study by Lucifora et al .42 who demonstrated that A3A and A3B do not deaminate the host genome due to binding HBc protein of HBV. However, we recently observed that A3A, A3B, and AID exhibit no off-site deamination only in human cells with very active HBV replication, whereas in cells with low viral loads, deamination of the host genome occurs31. In this study, we imitated the real-life therapeutic situation of suppressing HBV replication by antivirals, thereby reducing HBV viral loads, and then activated APOBEC/AID expression. In this scenario, A3A and A3B per se were not mutagenic, but the use of siRNA or LAM provoked host mutagenesis of the cancer-relatedTP53 gene by A3A and A3B. Thus, using CRISPR deaminases able to directly interact with cccDNA60 together with improved versions of cytidine deaminases lacking RNA-targeting properties may represent a more rational therapeutic approach61.
At the same time, CRISPR activation and CRISPR interference approaches demonstrate high efficacy in highly specific and transient activation of antiviral genes for suppressing viral replication19,31,62,63. The search for the critical antiviral factors for CRISPR activation is necessary to further its use as an antiviral therapeutic18.
To conclude, this study provides evidence of a new mechanism of immune evasion unique to HBV that works by saturating APOBEC/AID immunity with the rcDNA genomic form to preserve cccDNA pool integrity. Analysis of host genome deamination indicates severely hazardous activity of APOBEC/AID on host DNA when HBV viral loads are reduced, compromising its use as a potential therapy.