Detection of hepatitis B virus DNA in serum with nucleic acid probes labelled with 32P, biotin, alkaline phosphatase or sulphone.

To facilitate the clinical application of dot-blot hybridization for assaying hepatitis B virus (HBV) DNA, we compared the ability of nucleic acid probes labelled with 32P or with various non-radioactive markers to detect HBV DNA in patient serum. Cloned HBV DNA was hybridized with (1) 32P-labelled HBV DNA cloned in M13, (2) the 32P-labelled HBV RNA probe included in the HepProbe kit, (3) an alkaline phosphatase-labelled synthetic oligonucleotide of HBV, (4) biotin-labelled HBV DNA, and (5) sulphonated HBV DNA. Detection was either by autoradiography or an enzymatic colour reaction. The lowest level of detection of cloned HBV DNA was achieved with the 32P-labelled HBV RNA probe (0.3 pg HBV DNA, corresponding to 3 x 10(4) genomes in 50 microliters), followed by the 32P-labelled DNA probe (0.3-2 pg), sulphonated DNA (1-2 pg), biotin-labelled DNA (4 pg), and an alkaline phosphatase-labelled synthetic oligonucleotide (30 pg). Subsequently, sera from 159 patients with various constellations of HBsAg, HBeAg, and anti-HBe were tested with the most sensitive radioactive method (HepProbe) and the corresponding nonradioactive method (sulphonation). The overall concordance rate was 71% (r = 0.42). Compared with HepProbe results, sulphonation showed a sensitivity of 80% and a specificity of 67%. We conclude that radiolabelling (in particular 32P-labelling of HBV RNA) still allows the most sensitive and reliable detection of HBV DNA in patient serum using conventional dot-blot hybridization.