B-BEST: Profiling the intrahepatic HBV reservoir at single-cell resolution

This is the authors' abstract. We don't add key points for this paper.

HBV infection affects more than 250 million people worldwide and is one of the most common etiologies associated with the development of liver complications such as cirrhosis and HCC.1 Although current therapy based on the use of nucleos(t)ide analogues is able to control viral replication, it is rarely able to achieve a cure. Thus, individuals with chronic hepatitis B (CHB) require lifelong treatment, as discontinuation often leads to viral rebound. Although patients with CHB can be monitored for viral load and markers of liver damage in the blood, evaluating the intrahepatic viral reservoir and its impact on the host is relevant for understanding disease progression and developing novel HBV cure strategies. In this context, single-cell technologies have in recent years provided the opportunity to map the liver microenvironment in health and disease at high resolution. Moreover, these approaches have been employed not only to characterize the host response in patients with CHB,2,3 but also to specifically identify cells presenting viral transcripts.4,5 This additional information provides the possibility to compare the transcriptomic profiles between HBV-infected and bystander hepatocytes. However, HBV-infected hepatocytes can present a variety of replicative states, characterized by specific DNA intermediates and RNAs, with the latter potentially generated from episomal covalently closed circular (ccc)DNA or integrated HBV sequences. Considering that widely used oligo(dT)-based capture methods allow only to detect polyadenylated transcripts, capturing both HBV DNA and RNA could improve the detection of infected hepatocytes within the liver microenvironment. In a recent article published in Hepatology, Huang et al,6 describe the development of B-BEST (HBV can BE Seen on the host Transcriptome), a single-cell (sc)/single-nuclei (sn) RNA sequencing (RNA-seq) approach that combines oligo(dT) with custom HBV probes targeting 5 different viral regions, which the authors named: 3.5-kb, S, X, relaxed circular DNA, and covalently closed circular DNA (cccDNA) (Figure 1). Thus, the method can potentially capture both HBV RNA and DNA sequences. Compared to oligo(dT)-based capture alone, the B-BEST approach considerably increased the number of HBV reads (sevenfold) and demonstrated a comparable specificity when using the hepatic cell line HepAD38, which replicates HBV in the absence of tetracycline treatment.FIGURE 1: Schematic representation of the B-BEST method, which is based on the use of a dual-capture probe panel that combines oligo(dT) for polyadenylated mRNA enrichment and HBV-specific probes that target 5 viral regions. This approach enables the simultaneous characterization of the host transcriptome and identification of HBV-positive hepatocytes.Next, the authors analyzed liver samples from healthy controls and patients with CHB by sc/snRNA-seq. This analysis showed that snRNA sequencing outperformed scRNA-seq for the detection of hepatocytes, which is consistent with previous reports.7 Therefore, this method was selected for the subsequent evaluation of B-BEST. As expected, the characterization of CHB liver samples showed that HBeAg-positive patients presented a higher proportion of HBV-infected hepatocytes (defined as having at least one HBV target region detected) than HBeAg-negative individuals. Notably, HBV-positive hepatocytes were enriched in subpopulations exhibiting an increased expression of synthetic transporters and genes associated with mitochondrial processes, suggesting a preferential viral replication in metabolically active hepatocytes. This is consistent with published data showing that HBcAg-positive hepatocytes are enriched in pericentral areas.8 Then, the authors performed a differential expression analysis to study the host response taking place in HBV-infected hepatocytes. The results showed an upregulation of the main HBV entry factor, NTCP (SLC10A1), as well as genes involved in stress and immune signaling. such as FN1 and TRAIL. Additionally, transcription factors that have been described as favoring HBV replication (eg, YBX1 and RXRA) were also found to be upregulated. In contrast, splicing regulators (eg, DDX5 and DDX17) and chromatin-modifying enzymes (eg, SETD5 and KMT2C) exhibited impaired expression. Hepatocytes from patients with HBV-associated liver failure were characterized by a considerable suppression of HBV replication and increased expression of inflammatory markers (eg, CXCL1 and CXCL6). This suggests that the liver failure–associated transcriptomic profile is characterized by a robust antiviral response in these individuals. Although a slightly enhanced interferon (IFN) signaling pathway activity was observed in HBV-positive hepatocytes, only a minority presented it. This finding is in line with previous work describing a modest difference in the expression of inflammatory markers.4 To determine whether HBV transcripts were derived from cccDNA or integrated viral sequences, the authors performed PacBio single-cell long-read sequencing and explored the presence of HBV/host chimeric transcripts. This analysis revealed a higher proportion of transcripts derived from integrated sequences in HBeAg-negative patients than in HBeAg-positive individuals, consistent with previous reports.9 Finally, the authors employed the B-BEST method to characterize the response to antiviral therapy in human liver-chimeric Hu-URG mice. Analysis of these samples showed a considerable reduction in HBV-positive hepatocytes in mice treated with entecavir or pegylated-IFN. Furthermore, entecavir treatment increased the expression of genes associated with lipid metabolism (eg, SCP2 and ACACA), while pegylated-IFN increased the levels of IFN-stimulated genes, HLA class I antigens, and factors involved in the heat shock response. The work described in this article represents a technical step forward regarding the characterization of HBV persistence within the liver microenvironment. By enabling a more sensitive identification of infected hepatocytes and their transcriptomic profiles, this approach opens new possibilities for dissecting the mechanisms related to disease progression, immune evasion, and treatment response. However, since each probe targeting an HBV region can potentially detect multiple viral RNAs or replication intermediates, one limitation of the approach is to distinguish “silent” (ie, DNA-positive, RNA-negative) or “replication incompetent” (cccDNA-negative, with RNAs derived from integration) HBV-infected hepatocytes. Thus, future refinements aimed at specifically distinguishing them will be essential. This will require improving the methods to quantify cccDNA at single-cell resolution, as it remains a technical challenge for our scientific community.10 Overall, the development of B-BEST is an important advance for precisely assessing the intrahepatic viral reservoir. Using it in combination with complementary strategies, such as spatial transcriptomics,11 could provide an even more detailed characterization and facilitate the future development of curative strategies for CHB.