Hepatitis B virus (HBV) is a hepatotropic DNA virus that replicates by reverse transcription 
. It chronically infects >350 million people world-wide and kills up to 1.2 million patients annually by inducing liver failure and liver cancer 
. Reverse transcription is catalyzed by a virally-encoded polymerase that has two enzymatic activities: a DNA polymerase that synthesizes new DNA and a ribonuclease H (RNAseH) that destroys the viral RNA after it has been copied into DNA 
. Both activities are essential for viral replication.
HBV infections are treated with interferon α or one of five nucleos(t)ide analogs 
. Interferon α leads to sustained clinical improvement in 20–30% of patients, but the infection is very rarely cleared 
. The nucleos(t)ide analogs are used more frequently than interferon. They inhibit DNA synthesis and suppress viral replication by 4–5 log10
in up to 70–90% patients, often to below the standard clinical detection limit of 300–400 copies/ml 
. However, treatment eradicates the infection as measured by loss of the viral surface antigen (HBsAg) from the serum in only 3–6% of patients even after years of therapy 
. Antiviral resistance was a major problem with the earlier nucleos(t)ide analogs, but resistance to the newer drugs entecavir and tenofovir is very low 
. This has converted hepatitis B from a steadily worsening disease into a controllable condition for most individuals 
. The cost of this control is indefinite administration of the drugs (probably life-long; 
), with ongoing expenses of $400–600/month 
and unpredictable adverse effects associated with decades-long exposure to the drugs.
The key form of the HBV genome in cells that must be eliminated to clear the infection is the nuclear episomal covalently-closed circular DNA (cccDNA) that is the template for transcription of all HBV RNAs 
. Following reverse transcription in the cytoplasm, newly synthesized genomes can either be enveloped and secreted from the cell as virions, or they can be transported into the nucleus to replenish the cccDNA pool () 
. Transfer of newly synthesized viral genomes into the nucleus via “recycling” is the default pathway, and virion secretion occurs only if the cccDNA pool is large enough to support adequate synthesis of the HBsAgs.
The cccDNA pool is very stable, but nucleos(t)ide therapy can suppress cccDNA levels in the liver by ~1 log10
after 1–2 years 
. The indefinite persistence of the cccDNA even in patients whose HBV titres in serum have been suppressed below the limit of clinical detection by the nucleos(t)ide analogs is due to residual viral replication, leading to replenishment of the cccDNA pool by a combination of intracellular recycling and low-level infection of new cells 
. The sequential accumulation of resistance mutations during nucleos(t)ide therapy confirms that cccDNA maintenance by residual viral replication occurs in the absence of clinically detectable viremia 
. A recent genetic analysis of HBV DNA in the liver explicitly demonstrated that low levels of cccDNA replenishment occurs even when nucleos(t)ide analog therapy has reduced viral titres below the clinical detection limit 
RNAseH enzymes hydrolyze RNA in an RNA:DNA heteroduplex 
. They belong to the nucleotidyl transferase superfamily whose members share a similar protein fold and presumably have similar enzymatic mechanisms 
. This family includes E. coli
RNAseH I and II 
, DNA transposases including the Tn5 transposase 
, retroviral integrases including the HIV integrase 
, the RuvC Holliday junction resolvase 
, the Argonaute RNAse 
, and human RNAseH 1 and 2 
. The canonical RNAseH structure contains about 100 aa including four conserved carboxylates (the “DEDD” motif) that coordinate two divalent cations 
. The RNAseH mechanism is believed to involve both divalent cations 
, although a one-ion mechanism has also been proposed 
. The HBV RNAseH domain shares low but recognizable (~20%) sequence identity with the RNAseH domains of reverse transcriptases and other retro-elements 
. Manually optimizing alignment of the HBV RNAseH and the HIV-1 RNAseH yielded 23% identity and 33% similarity (). A similar alignment between the HBV RNAseH and the HIV integrase revealed 19% identity and 33% similarity.
Alignments between the HBV RNAseH and the HIV-1 RNAseH and integrase.
The HBV RNAseH is encoded at the carboxy-terminus of the viral polymerase protein that also encodes the viral DNA polymerase activity (reverse transcriptase). The high hydrophobicity of the HBV polymerase and its existence as a complex with host chaperones 
have severely restricted study of the HBV RNAseH. Furthermore, we demonstrated that the RNAseH in its native context within the polymerase protein is unable to accept exogenous heteroduplex substrates 
, analogous to the inability of the DNA polymerase active site to engage exogenous primer-templates 
. Consequently, most of our limited knowledge of the RNAseH comes from mutational studies of the viral genome in the context of viral replication conducted by us and others 
. These restrictions have prevented biochemical characterization of the RNAseH and blocked biochemical screens for anti-HBV RNAseH drugs to date.
A few reports of recombinant forms of the hepadnaviral RNAseH exist. Wei and co-workers 
expressed the HBV RNAseH domain in E. coli
and purified it by denaturing nickel-affinity chromatography. Following refolding, they found an RNAse activity. Lee et al. 
expressed the HBV RNAseH domain in E. coli
as a dual maltose-binding protein/hexahistidine fusion and purified soluble protein by two-step affinity chromatography; this enzyme had RNAseH activity. Choi and co-workers 
expressed the intact duck hepatitis B virus polymerase in yeast and reported that it had a weak RNAse activity. Finally, Potenza et al. 
expressed the HBV RNAseH domain as a synthetic gene in E. coli
. Following purification from inclusion bodies and refolding, this enzyme had RNAse activity. However, no follow-up reports have appeared with any of these systems, possibly due to the technical difficulties associated with the purification protocols and/or contamination challenges with host RNAseH or other RNAse classes.
Human Immunodeficiency Virus (HIV) reverse transcription also requires a virally encoded RNAseH activity 
, and consequently the RNAseH has attracted much attention as a potential drug target 
. Over 100 anti-HIV RNAseH compounds have been reported, typically with inhibitory concentration-50% (IC50
) values in the low µM range. Most of the compounds inhibit HIV replication in culture, typically with effective concentration-50% (EC50
) values that are ~10-fold higher than the biochemical IC50
values. These compounds are often modestly cytotoxic, leading to therapeutic indices (TI) that are usually <10. Second-generation inhibitors with substantially improved efficacy have been reported, but their TI values were not necessarily improved markedly 
. Despite these limitations, compounds with efficacy and TI values appropriate for a drug exist 
. Most of the compounds inhibit the RNAseH by binding to the enzyme and chelating the divalent cations in the active site 
, but compounds that appear to inhibit the RNAseH by altering the enzyme's conformation or its interaction with nucleic acids have also been reported 
. As predicted from their common membership in the nucleotidyl transferase superfamily, some anti-HIV RNAseH compounds can inhibit the HIV integrase, and some anti-integrase compounds can inhibit the RNAseH 
The ability of the nucleos(t)ide analog drugs to profoundly suppress HBV in most patients and to cure HBV infection in a few patients indicates that they can push the virus to the brink of elimination. This presents an opportunity to cure many more patients by suppressing HBV replication further, but achieving a cure will require novel drugs against targets other than the DNA polymerase active site. These drugs would be used in combination with the nucleos(t)ide analogs to suppress viral replication below the level needed to maintain the cccDNA. A logical target would be the second of HBV's two enzymatic activities, the RNAseH. Here, we report production of enzymatically active recombinant HBV RNAseH suitable for low throughput antiviral drug screening. Using this novel reagent, we demonstrated that the HIV RNAseH and integrase are similar enough to the HBV RNAseH to allow information derived from HIV RNAseH and integrase inhibitors to guide identification of anti-HBV RNAseH compounds.