Published online Nov 14, 2012. doi: 10.3748/wjg.v18.i42.6096
Revised: July 25, 2012
Accepted: July 28, 2012
Published online: November 14, 2012
AIM: To investigate the variability of the main immunodominant motifs of hepatitis B virus (HBV) core gene by ultra-deep-pyrosequencing (UDPS).
METHODS: Four samples (2 genotype A and 2 genotype D) from 4 treatment-naïve patients were assessed for baseline variability. Two additional samples from one patient (patient 4, genotype D) were selected for analysis: one sample corresponded to a 36-mo treatment-free period from baseline and the other to the time of viral breakthrough after 18 mo of lamivudine treatment. The HBV region analyzed covered amino acids 40 to 95 of the core gene, and included the two main epitopic regions, Th50-69 and B74-84. UDPS was carried out in the Genome Sequencer FLX system (454 Life Sciences, Roche). After computer filtering of UDPS data based on a Poisson statistical model, 122 813 sequences were analyzed. The most conserved position detected by UDPS was analyzed by site-directed mutagenesis and evaluated in cell culture.
RESULTS: Positions with highest variability rates were mainly located in the main core epitopes, confirming their role as immune-stimulating regions. In addition, the distribution of variability showed a relationship with HBV genotype. Patient 1 (genotype A) presented the lowest variability rates and patient 2 (genotype A) had 3 codons with variability higher than 1%. Patient 3 and 4 (both genotype D) presented 5 and 8 codons with variability higher than 1%, respectively. The median baseline frequencies showed that genotype A samples had higher variability in epitopic positions than in the other positions analyzed, approaching significance (P = 0.07, sample 1 and P = 0.05, sample 2). In contrast, there were no significant differences in variability between the epitopic and other positions in genotype D cases. Interestingly, patient 1 presented a completely mutated motif from amino acid 64 to 67 (E64LMT67), which is commonly recognized by T helper cells. Additionally, the variability observed in all 4 patients was particularly associated with the E64LMT67 motif. Codons 78 and 79 were highly conserved in all samples, in keeping with their involvement in the interaction between the HBV virion capsid and the surface antigens (HBsAg). Of note, codon 76 was even more conserved than codons 78 and 79, suggesting a possible role in HBsAg interactions or even in hepatitis B e antigen conformation. Sequential analysis of samples from patient 4 (genotype D) illustrated the dynamism of the HBV quasispecies, with strong selection of one minor baseline variant coinciding with a decrease in core variability during the treatment-free and lamivudine-treated period. The drop in variability seemed to result from a “steady state” situation of the HBV quasispecies after selection of the variant with greatest fitness.
CONCLUSION: Host immune pressure seems to be the main cause of HBV core evolution. UDPS analysis is a useful technique for studying viral quasispecies.