The RNase H domain of human hepatitis B virus (HBV) polymerase is an attractive molecular target for the development of new anti-HBV drugs. In this study, a synthetic gene coding for HBV RNase H was assembled from 12 oligonucleotides and expressed in Escherichia coli. The encoded protein was then recovered from inclusion bodies, purified, and refolded by a dilution-dialysis procedure in the presence of a low concentration of lauroylsarcosine (0.01%). The presence of the detergent was an absolute requirement for solubility, suggesting that the untagged RNase H might have exposed hydrophobic regions that need to be shielded from the solvent. The structural identity of the protein was confirmed by N-terminal amino acid sequence analysis and mass spectrometry. The enzymatic activity of HBV RNase H was then tested by a recently developed fluorometric assay and was found to be only slightly lower than that registered with the entire HIV-1 reverse transcriptase. Finally, a structural model of the enzyme showed that H715, R744 and K745 may be involved in substrate recognition.