Quantitative dynamics of hepatitis B basal core promoter and precore mutants before and after HBeAg seroconversion

Abstract

Background & aims: Hepatitis B e antigen (HBeAg) seroconversion is an important clinical and virological "landmark" during chronic hepatitis B virus (HBV) infection. Mutant viruses carrying the precore G1896A and/or the basal core promoter (BCP) A1762T/G1764A mutations are associated with HBeAg seroconversion. However, the exact role of these mutants in HBeAg seroconversion remains unclear, partly because the evolution of these mutant viruses before and after seroconversion has not been well studied.

Methods: Using our novel mutant quantification methods, the percentage of the mutant viruses was analyzed both cross-sectionally and longitudinally, before and after seroconversion.

Results: Cross-sectional analysis showed that the percentage of both precore and BCP mutants gradually increased with age in the HBeAg-positive population. Follow-up of 18 HBeAg-positive patients revealed that the mutant percentage may stay low and stable for many years, followed by a steady increase in the percentage of G1896A and/or A1762T/G1764A mutants, from <10% to 50-100%, within about 3 years prior to seroconversion. In all cases, increase of mutant percentage was preceded or accompanied by elevated serum alanine aminotransferase. After the seroconversion, the mutant percentage could remain high or decrease significantly, sometimes to below 20%.

Conclusions: Levels of G1896A and A1762T/G1764A mutants (of genotypes B and C) in the HBeAg-positive patients may predict the time of HBeAg seroconversion. The dominance of these mutants in the HBeAg-positive phase is more likely the result of immune selection rather than the enhanced replication capability of the mutants. However, anti-HBe antibody may not be a major selection force for these mutants.

Fig. 1. The total viral loads and mutant percentages in HBeAg-positive and -negative populations

Total HBV DNA titers transformed by log10 were plotted against different age groups (see Table 1) that are either HBeAg-positive (e+) or HBeAg-negative (e−) (A). The distribution and summary of precore mutant percentage (B) and BCP mutant percentage (C) are also shown. The means diamond indicates the mean (the line cross the diamond) and the 95% confidence interval (the vertical span of the diamond). The short horizontal lines flanking the means are overlap marks computed as 0.3536 × confidence interval above and below the mean.

Fig. 2. Longitudinal analysis of precore and BCP mutants in 9 individuals who underwent seroconversion

Shown in each case, from top to bottom, are gender and HBV genotype, ALT status (ALT levels >40 U/L are shown by an upward triangle, while ALT ≤ 40 U/L are shown by a downward triangle), Log₁₀ total viral load (open square line graph), and mutant percentage (open circle for the precore mutant and solid circle for the BCP mutant). The data is plotted against the patient’s age at which the sample was taken. The vertical dashed lines are in between the last HBeAg-positive sample and the first HBeAg-negative sample, and indicate the presumed time of seroconversion. The last case (I) had a second seroconversion. The HBeAg-positive samples before the second seroconversion are indicated by asterisks.

Fig. 3. Longitudinal analysis of precore and BCP mutants in 5 individuals who did not undergo seroconversion

Shown in each case, from top to bottom, are gender and HBV genotype, ALT status (ALT levels >40 U/L are shown by an upward triangle, while ALT ≤ 40 U/L are shown by a downward triangle), Log₁₀ total viral load (open square line graph), and mutant percentage (open circle for the precore mutant and solid circle for the BCP mutant). The data is plotted against the patient’s age at which the sample was taken.

Fig. 4. Temporal correlation between an increase of mutant percentage and seroconversion

A, Evolutionary patterns of HBV precore and BCP mutants before and after HBeAg seroconversion. B and C, Correlation between the mutant percentage levels and the time of seroconversion. The data points were extracted from the longitudinal study of 13 cases in which seroconversion occurred. Linear regressions are also shown.