An Oxford Nanopore Technology-Based Hepatitis B Virus Sequencing Protocol Suitable for Genomic Surveillance Within Clinical Diagnostic Settings

Abstract

Chronic Hepatitis B Virus (HBV) infection remains a significant public health concern, particularly in Africa, where the burden is substantial. HBV is an enveloped virus, classified into ten phylogenetically distinct genotypes (A-J). Tests to determine HBV genotypes are based on full-genome sequencing or reverse hybridization. In practice, both approaches have limitations. Whereas diagnostic sequencing, generally using the Sanger approach, tends to focus only on the S-gene and yields little or no information on intra-patient HBV genetic diversity, reverse hybridization detects only known genotype-specific mutations. To resolve these limitations, we developed an Oxford Nanopore Technology (ONT)-based HBV diagnostic sequencing protocol suitable for clinical virology that yields both complete genome sequences and extensive intra-patient HBV diversity data. Specifically, the protocol involves tiling-based PCR amplification of HBV sequences, library preparation using the ONT Rapid Barcoding Kit (Oxford nanopore Technologies, Oxford, OX4 4DQ, UK), ONT GridION sequencing, genotyping using genome detective software v1.132/1.133, a recombination analysis using jpHMM (26 October 2011 version) and RDP5.61 software, and drug resistance profiling using Geno2pheno v2.0 software. We prove the utility of our protocol by efficiently generating and characterizing high-quality near full-length HBV genomes from 148 residual diagnostic samples from HBV-infected patients in the Western Cape province of South Africa, providing valuable insights into the genetic diversity and epidemiology of HBV in this region of the world.

Keywords: HBV; NGS; ONT; recombination; whole-genome sequencing.

Figure 1

(A) Comparison of genome coverage for different detected HBV genotypes. The boxes indicate the lower quartile, median, and upper quartile minimum and maximum values by the whiskers. Significant differences between genome coverage (paired Wilcoxon test) values between each genotype are denoted above the box and whisker plots. (B) Scatter plot for log viral load against the genome coverage for detected HBV genotypes. A total of 124 genomes with >80% coverage was produced (36 with a log viral load of <7.6 IU/mL and 88 with a log viral load of >7.6 IU/mL). HBV genotypes are represented by different colors and shapes. The different background color shades represent different quality control groups. The blue and purple shades represent ≤7.6 IU/mL log viral load and coverage of ≥80% and <80%, respectively. The green and pink shades represent a log viral load of >7.6 IU/mL and coverage of ≥80% and <80%, respectively.

Figure 2

Flowchart showing the identification of recombinants using the jpHMM HBV tool and RDP5.46. The blue section shows the number of genomes produced by the genome detective, the green section shows genotype variation as assessed by the genome detective hepatitis B phylogenetic typing tool; the orange section highlights the jpHMM recombination classifications of the unclassified sequences, and the red section shows the RDP5.46 recombination analysis. Sample IDs are shown below the flowchart sections, and IDs that are shown in red highlight recombinants that both jpHMM and RDP5.46 detected. Sample IDs marked with a “*” were classified as different recombinants by jpHMM and RDP5.46.

Figure 3

jpHMM genome maps for A/D recombinant viruses. The query isolates identifier names which are listed below each jumping profile map. Genome maps presented here were created using the software package, Circos [28]. The colored shadings represent different HBV genotypes (red = A, yellow = D, gray = unknown). Regions of orange shading represent recombination breakpoint intervals, e.g., region 405 ± 40 (outer ring). All sequence position numbers are given relative to the HBV reference genome AM282986. Positions of genes in the genome are marked with gray and black bars (inner ring). The color legend is in the middle, and the “NA” denotes “not assigned”.

Figure 4

(A) jpHMM genome maps and (B) genome detective bootscan plots for non-A/D recombinant viruses. Bootscan analysis was performed with a window size of 400 and a step size of 100. (C) Genome coverage maps highlighting HBV sequencing depth. The query isolates are listed below each genome coverage map.

Figure 5

Recombination region and breakpoint distributions (A) recombinant region count matrix highlighting areas of the genome that are most and least commonly transferred during recombination events. Unique recombination events were mapped onto a region count matrix based on determined breakpoint positions. Each cell in the matrix represents a pair of genome sites, with the colors of cells indicating the number of times recombination events separated the represented pairs of sites. (B) Breakpoint distribution across HBV genomes. All detectable breakpoint positions are represented as black lines above the graph. The green areas show the 99% confidence interval for breakpoint clustering under random recombination. The upper dotted line represents the global 99% confidence interval for a breakpoint clustering under random recombination, and the lower dotted line is the global 95% confidence interval for breakpoint clustering under random recombination. The black line represents the number of breakpoints within a 200-nucleotide window moved along the genome. Areas in red where the black line emerges above the green area are considered recombination warm spots, and those that traverse the dotted global 95% confidence interval line are considered statistically supported recombination hotspots. An HBV gene map is plotted between the figures.

Figure 6

(A) Prevalence of predicted drug resistance based on mutation patterns in the RT/HBsAg overlapping region for HBV genomes sampled in South Africa between September 2022 and November 2023. (B) Prevalence of HBsAg vaccine escape mutations in the RT/HBsAg overlapping region for these same genomes.