β-Catenin Signaling Regulates the In Vivo Distribution of Hepatitis B Virus Biosynthesis across the Liver Lobule

Abstract

β-Catenin (Ctnnb1) supports high levels of liver gene expression in hepatocytes in proximity to the central vein functionally defining zone 3 of the liver lobule. This region of the liver lobule supports the highest levels of viral biosynthesis in wild-type hepatitis B virus (HBV) transgenic mice. Liver-specific β-catenin-null HBV transgenic mice exhibit a stark loss of high levels of pericentral viral biosynthesis. Additionally, viral replication that does not depend directly on β-catenin activity appears to expand to include hepatocytes of zone 1 of the liver lobule in proximity to the portal vein, a region of the liver that typically lacks significant HBV biosynthesis in wild-type HBV transgenic mice. While the average amount of viral RNA transcripts does not change, viral DNA replication is reduced approximately 3-fold. Together, these observations demonstrate that β-catenin signaling represents a major determinant of HBV biosynthesis governing the magnitude and distribution of viral replication across the liver lobule in vivo. Additionally, these findings reveal a novel mechanism for the regulation of HBV biosynthesis that is potentially relevant to the expression of additional liver-specific genes. IMPORTANCE Viral biosynthesis is highest around the central vein in the hepatitis B virus (HBV) transgenic mouse model of chronic infection. The associated HBV biosynthetic gradient across the liver lobule is primarily dependent upon β-catenin. In the absence of β-catenin, the gradient of viral gene expression spanning the liver lobule is absent, and HBV replication is reduced. Therefore, therapeutically manipulating β-catenin activity in the livers of chronic HBV carriers may reduce circulating infectious virions without greatly modulating viral protein production. Together, these changes in viral biosynthesis might limit infection of additional hepatocytes while permitting immunological clearance of previously infected cells, potentially limiting disease persistence.

Keywords: hepatitis B virus (HBV); liver lobule; transgenic mice; viral biosynthesis; β-catenin.

FIG 1

Effect of β-catenin on transcriptional activation of the HBV promoters. HEK293T cells were transfected with a constitutively active β-catenin and/or an Lrh1 expression vector with a firefly luciferase (LUC) reporter gene construct driven by one of the four HBV promoters, CpLUC (A), SpLUC (B), PS(1)pLUC (C), and XpLUC (D), respectively. A CMV-driven Renilla luciferase reporter gene construct (pRL-CMV) was used as an internal control for transfection efficiency. Means and standard deviations from two replicates are shown. Abbreviations: Cp, nucleocapsid (core) promoter; Sp, major surface antigen promoter; PS(1)p, large surface antigen promoter; Xp, X-gene promoter.

FIG 2

Gene expression levels in control and liver-specific β-catenin-null HBV transgenic mice. (A) NanoString gene expression analysis of β-catenin RNA and β-catenin target gene transcripts from total liver RNA. (B) Representative immunohistochemical analysis of liver samples from HBV transgenic mice. Comparison of control (HBV^+/−Ctnnb1^flox/floxAlbCre^−/−) and constitutive liver-specific β-catenin-null HBV transgenic mice (HBV^+/−Ctnnb1^flox/floxAlbCre^+/−). Immunohistochemical staining indicates the presence of glutamate-ammonia ligase (Glul)/glutamine synthetase (GS) (c, central vein; p, portal vein). (C) NanoString gene expression analysis of cytokine gene transcripts from total liver RNA. Means and standard deviations are presented (n = 9 per group). Statistically significant differences between groups determined by Wilcoxon rank sum test are indicated; *, P < 0.05; **, P < 0.01; ***, P < 0.001; ns, not significant.

FIG 3

HBcAg immunohistochemical analysis of livers from constitutive liver-specific β-catenin-null HBV transgenic mice. Comparison of control (HBV^+/−Ctnnb1^flox/floxAlbCre^−/−) and liver-specific β-catenin-null HBV transgenic mice (HBV^+/−Ctnnb1^flox/floxAlbCre^+/−). Representative liver lobules are expanded below each section with central vein on the left and portal vein on the right. Immunohistochemical staining indicates the presence of HBcAg.

FIG 4

HBcAg immunohistochemical analysis of livers from inducible liver-specific β-catenin-null HBV transgenic mice. Comparison of control (HBV^+/−Ctnnb1^flox/floxSA-Cre-ER^T2−/− with and without tamoxifen treatment plus HBV^+/−Ctnnb1^flox/floxSA-Cre-ER^T2+/− without tamoxifen treatment) and inducible liver-specific β-catenin-null HBV transgenic mice (HBV^+/−Ctnnb1^flox/floxSA-Cre-ER^T2+/− with tamoxifen treatment). Mice were either tamoxifen or vehicle treated as indicated. Representative liver lobules are expanded below each section with central vein on the left and portal vein on the right. Immunohistochemical staining indicates the presence of HBcAg.

FIG 5

Quantitation of immunohistochemical analysis of livers from wild-type plus constitutive and inducible liver-specific β-catenin-null HBV transgenic mice. The intensity of HBcAg staining spanning 5 to 10 liver lobules from each mouse was measured and proportionally distributed into 9 bins, where bins 1 and 9 were central vein (CV) and portal vein (PV) proximal, respectively. The means and standard deviations of staining intensity are displayed. (A) Control and constitutive β-catenin-null mice: 2 male and 2 female control [Cre(−)] HBV transgenic mice; 3 male and 3 female constitutive β-catenin-null [Cre(+)] HBV transgenic mice. (B) Control and inducible β-catenin-null mice: 7 male and 7 female control HBV transgenic mice comprising 2 Cre⁻ with vehicle treatment [Cre(−)+V], 2 Cre⁻ with tamoxifen treatment [Cre(−)+T], and 3 Cre⁺ with vehicle treatment [Cre(+)+V]; 3 male and 3 female inducible β-catenin-null HBV transgenic mice {Cre⁺ with tamoxifen treatment [Cre(+)+T]}.

FIG 6

Analysis of HBV RNA expression in liver-specific β-catenin-null HBV transgenic mice. (A) Counts of HBV 3.5-kb transcripts in total liver RNA from control (Cre⁻) and constitutive liver-specific β-catenin-null HBV transgenic mice (Cre⁺) by NanoString gene expression analysis (n = 9 per group). (B) RNA filter hybridization analysis on total liver RNA from control (Cre⁻) and constitutive liver-specific β-catenin-null HBV transgenic mice (Cre⁺). The glyceraldehyde 3-phosphate dehydrogenase (Gapdh) transcript was used as an internal control for the quantitation of the HBV 3.5-kb RNA. Noncontiguous lanes from a single analysis are presented. (C) Quantitation of viral 3.5-kb RNA from the RNA filter hybridization analysis. Means and standard deviations are presented (male Cre⁻, n = 23; male Cre⁺, n = 28; female Cre⁻, n = 14; female Cre⁺, n = 13). Statistically significant differences between HBV transgenic mice by Wilcoxon rank sum test are indicated; *, P < 0.05; **, P < 0.01; ***, P < 0.001; ns, not significant.

FIG 7

Analysis of viral DNA replication intermediates of liver-specific β-catenin-null HBV transgenic mice. (A) DNA filter hybridization analysis on total liver DNA from control (Cre⁻) and constitutive liver-specific β-catenin-null HBV transgenic mice (Cre⁺). The HBV transgene was used as an internal control for the quantitation of the HBV replication intermediates. Abbreviations: Tg, transgene; RC, relaxed circular replication intermediates; SS, single-stranded replication intermediates. Noncontiguous lanes from a single analysis are presented. (B) Quantitation of viral DNA replication intermediates from the DNA filter hybridization analysis. Means and standard deviations are presented (male Cre⁻, n = 23; male Cre⁺, n = 28; female Cre⁻, n = 14; female Cre⁺, n = 13). Statistically significant differences between HBV transgenic mice by Wilcoxon rank sum test are indicated; *, P < 0.05; **, P < 0.01; ***, P < 0.001; ns, not significant.

FIG 8

Model for the role of β-catenin in regulating HBV biosynthesis in HBV transgenic mice. HBV replication occurs within the assembling capsids in the cytoplasm of hepatocytes. Thus, cytoplasmic staining of HBcAg serves as a readout of the lobular distribution of viral replication (14). β-Catenin activity directs robust HBV transcription immediately proximal to the central vein via endothelial cell-derived Wnt signaling (15). This poorly diffusible signal rapidly diminishes across the liver lobule (15). Lower levels of viral replication present in midzone and periportal zone 1 hepatocytes appear to be supported by viral transcription that is independent of direct β-catenin activity. In the absence of β-catenin activity, robust HBV biosynthesis is lost from pericentral zone 3 hepatocytes with compensating viral transcription and hence enhanced capsid biosynthesis expanding to include periportal zone 1 hepatocytes.