The characteristics of binding of hepatitis B core antigen (HBcAg) and hepatitis B e antigen (HBeAg) polypeptides to hepatitis B virus (HBV) DNA were analyzed. HBcAg polypeptide from recombinant HBV core particles and HBeAg polypeptide from partially purified serum HBeAg were prepared and verified to have molecular weights of 21,500 (P21.5) and of 17,000 (P17) and 18,000 (P18), respectively, by immunoblot analysis. By reaction of these proteins on a nitrocellulose membrane with cloned 32P-HBV DNA, it was revealed that the HBeAg polypeptide, which lacks the C-terminal 34 amino acids of P21.5, as well as the HBcAg polypeptide, bound to the DNA. The secondary structures of nucleocapsid proteins of HBV, woodchuck hepatitis virus, and ground squirrel hepatitis virus were predicted by the Garnier algorithm. Amino acid sequences which, in addition to those of the C-terminal regions, may contribute to binding were proposed to be the 21-amino-acid residues located at amino acids 100 to 120 of the nucleocapsid proteins of these hepadnaviruses.

The interrelation between HBeAg subtypes, HBeAg/1 and HBeAg/2, in sera was examined immunochemically. The detection of HBeAg subtypes in immunodiffusion (ID) depended upon the amount of HBeAg determined by reversed passive haemagglutination (RPHA), ie, the titres of HBeAg in sera positive for both HBeAg/1 and HBeAg/2, positive for only HBeAg/1 and negative for both HBeAg/1 and HBeAg/2 were 2(9.8) +/- 1.5, 2(7.0) +/- 1.6, and 2(5.6) +/- 1.3, respectively. When the sera belonging to the latter two groups were concentrated up to 2(10) RPHA titre, the precipitin line corresponding to that of HBeAg/2-anti-HBeAg/2 was visualized in ID. Monoclonal anti-HBeAg antibody that absorbed only the precipitin line of HBeAg/1-anti-HBeAg/1 in ID was prepared for the characterization of HBeAg subtypes. A linear correlation (r = 0.91) between titres of HBeAg determined by the RPHA cells prepared with monoclonal and polyclonal antibodies was found in almost all HBeAg-positive sera. The reactivities of this monoclonal anti-HBeAg antibody to both HBeAg/1 and HBeAg/2 were demonstrated in affinity chromatography experiments using a Sepharose 4B column conjugated with this antibody. These results suggest that both HBeAg/1 and HBeAg/2 are constantly present in HBeAg-positive sera and that they are closely associated. Based upon these results, a hypothetical model for the elucidation of the immunological relationship between HBeAg/1 and HBeAg/2 is proposed.

No abstract

In order to establish the virological significance of HBeAg subtypes (HBeAg/1 and HBeAg/2) during hepatitis B virus infection, HBsAg, HBeAg and hepatitis B virus DNA in serum and HBcAg in liver were determined quantitatively in relation to the detection of HBeAg subtypes in agar gel diffusion. Thirty-eight chronic HBsAg carriers with HBeAg, including 16 non-specific reactive hepatitis, 8 chronic persistent hepatitis, 11 chronic active hepatitis and 3 liver cirrhosis, who were seen at Tohoku University Hospital from 1983 to 1985, were examined. Significantly larger amounts of HBsAg, HBeAg and hepatitis B virus DNA in serum and HBcAg in liver were found in patients positive for both HBeAg/1 and HBeAg/2 in serum than in those positive for only HBeAg/1 or negative for both subtypes. These results suggest that the presence of HBeAg/2 in serum may reflect the occurrence of active viral replication. When the detection pattern of HBeAg subtypes was examined during serial follow-up for at least 1 year, three groups of patients were classified with respect to the presence of HBeAg/2, i.e., Type I, consistently positive for HBeAg/2; Type II, consistently negative for HBeAg/2, and Type III, intermittently positive for HBeAg/2. More than 80% of Type I patients were histologically diagnosed having as nonspecific reactive hepatitis, while more than 80% of Type II and III patients had more progressive liver diseases such as chronic persistent hepatitis, chronic active hepatitis and liver cirrhosis. These results suggest that the serial examination of HBeAg subtypes in serum may be important for more detailed evaluations of type B hepatitis.

The molecular interrelation between hepatitis B e Ag 1 (HBeAg/1) and HBeAg/2 was examined immunochemically. Major polypeptides with m.w. around 17,000 (P17) and one minor polypeptide with m.w. 18,000 (P18) that had HBeAg activity were consistently detected in 12 different serum samples by immunoblotting analysis. To examine the polypeptide composition of HBeAg/1 and HBeAg/2, precipitin lines of HBeAg/1-anti-HBeAg/1 and HBeAg/2-anti-HBeAg/2 were separately cut from the agarose gel and analyzed by immunoblotting. HBeAg/1 and HBeAg/2 showed similar P17 and P18 compositions. Monomeric forms of P17 and P18 in a solution of 0.1% SDS showed HBeAg/1 activity, whereas polymeric forms of these polypeptides showed HBeAg/2 activity. The stability of HBeAg subspecificities to SDS and 2-ME was examined. When HBeAg was incubated in a buffer containing 0.1% SDS and 0.1% 2-ME, only the HBeAg/2 precipitin line disappeared in agarose gel concurrently with the conversion of HBeAg molecules to a monomeric from a polymeric form. In contrast, neither monomerization nor the disappearance of HBeAg/2 was seen when SDS or 2-ME was used by itself. These results indicate that hydrophobic forces and disulfide bonds may be involved in the association of P17 and P18 in serum and that HBeAg/2 activity depends upon association of HBeAg-polypeptides but that HBeAg/1 activity does not. The possibility that the epitopes of HBeAg/2 are specific for the conformation of polymerized molecules is discussed.