Interferon stimulated immune profile changes in a humanized mouse model of HBV infection

Abstract

The underlying mechanism of chronic hepatitis B virus (HBV) functional cure by interferon (IFN), especially in patients with low HBsAg and/or young ages, is still unresolved due to the lack of surrogate models. Here, we generate a type I interferon receptor humanized mouse (huIFNAR mouse) through a CRISPR/Cas9-based knock-in strategy. Then, we demonstrate that human IFN stimulates gene expression profiles in huIFNAR peripheral blood mononuclear cells (PBMCs) are similar to those in human PBMCs, supporting the representativeness of this mouse model for functionally analyzing human IFN in vivo. Next, we reveal the tissue-specific gene expression atlas across multiple organs in response to human IFN treatment; this pattern has not been reported in healthy humans in vivo. Finally, by using the AAV-HBV model, we test the antiviral effects of human interferon. Fifteen weeks of human PEG-IFNα2 treatment significantly reduces HBsAg and HBeAg and even achieves HBsAg seroconversion. We observe that activation of intrahepatic monocytes and effector memory CD8 T cells by human interferon may be critical for HBsAg suppression. Our huIFNAR mouse can authentically respond to human interferon stimulation, providing a platform to study interferon function in vivo. PEG-IFNα2 treatment successfully suppresses intrahepatic HBV replication and achieves HBsAg seroconversion.

Fig. 1. Generation and functional evaluation of a type I Interferon receptor-humanized (huIFNAR) mouse.

a Schematic diagram of huIFNAR knock-in strategy in the exon 2 of mouse Ifnαr2 gene. The upper, from left to right, consists of mouse Ifnαr2 promoter (deep blue), Exons (orange), the insert, and end. The chimeric receptor structure (below) is indicated. Hu: human; Mo: Mouse. R2: interferon receptor 2, R1: interferon receptor 1. Linker-2A: linker region and 2A self-cutting protease. PolyA: transcription termination signal. b Topological structure of humanized interferon receptor. Blue, human origin; orange, mouse origin. c Relative huIFNAR mRNA expression to mouse Gapdh (mGapdh) by reverse transcription quantitative PCR (qPCR), n = 3 biologically independent WT mice and huIFNAR mice, respectively. d Flow cytometry determination of huIFNAR protein on the cell surface. e–g huIFNAR response to human IFN stimulation in vitro. Mouse PBMCs isolated from blood was stimulated for 5 h ex vivo with human IFNα2 (800 ng/ml) or plus huIFNAR blocking antibodies. anti-hR1 and anti-hR2 represent anti-human IFNAR1 and IFNAR2 antibodies (10 µg/ml), respectively. The mMx1 and mIsg15 normalized to mGapdh or mActin are indicated. n = 3 biologically independent mice in each group. h–j Functional assess of huIFNAR response to PEG-IFNα2 (2 µg/ml for 16 h, subcutaneous injection) in vivo. The relative levels of mMx1 and mIsg15 mRNA to mGapdh in blood PBMC, liver and spleen are indicated. For two group comparison, continuous variables were represented as mean ± SEM or median and interquartile range (IQR), and compared by two-sided unpaired t-test or Mann–Whitney U test as appropriate. For multiple comparison, ANOVA test or Kruskal–Wallis test was applied. n = 3 biologically independent mice in each group. Source data are provided as a Source Data file.

Fig. 2. Similarity analysis between the mouse huIFNAR and human PBMCs in response to human IFN stimulation.

a Experiment protocol for comparing the immune response of PBMCs to PEG-IFNa2 (0.8 µg/ml for 8 h) stimulation between health humans (n = 3) and huIFNAR mice (n = 3). The gene expression profile was analyzed by next-generation sequencing. b Correlation of differentially expressed ISGs between human and huIFNAR mice. Each dot represents one gene. Number and percentage are indicated in each quadrant. c Venn diagrams of shared enriched KEGG pathways induced by IFN between human and huIFNAR mouse. Number and percentage are indicated. d The top 30 most enriched GO terms in all biological processes (BP). Circle, human. Triangle, huIFNAR mouse. e, f The immune and metabolic-related sub GO terms as in (d). Adjusted p values in (d–f) are calculated using two-sided hypergeometric distribution and adjusted by Benjamini–Hochberg method. Source data are provided as a Source Data file.

Fig. 3. The tissue-specific transcriptome atlas of response to the human PEG-IFNα2.

a Schematic diagram tissue collection as indicated at 16 h after subcutaneous inoculation of human PEG-IFNα2 (2 µg/mouse, in 200 µl buffer). b The huIFNAR2 mRNA levels in 9 tissues. n = 3 biologically independent huIFNAR mice. CPM counts per million reads. Data are presented as mean values ± SEM. c Heatmap of differential expression genes (DEGs). DEGs were defined as significance under an FDR threshold of 0.05 in the IFN group versus the control group. The DEGs numbers were indicated under each tissue. d Venn diagrams of the DEGs among PBMC, liver, and spleen. e Heatmap of differentially expressed ISGs. f Alluvial plot of Tissue-specific GO analysis. A total of 15 GO terms from the top 5 significant biological process terms in each organ was shown. The stream wideness represents the size of annotated DEGs enriched in the GO term. Source data are provided as a Source Data file.

Fig. 4. Human PEG-IFNα2 suppresses HBV in the huIFNAR AAV-HBV mouse model.

a Experimental protocol. A huIFNAR mouse model of chronic AAV-HBV infection was established by tail vein injection of a 1.3-fold HBV genome (AAV-1.3×HBV). After that, Peg-IFNα2(2 µg/ mouse) was injected and blood was collected from tail vein for virologic index test every week. According to the type of mice and different treatments, the mice were divided into three groups: C57BL/6J AAV-HBV PEG-IFNα2-treated wildtype group (WT-IFNα2, n = 6), huIFNAR AAV-HBV mock-treated group (huIFNAR-Mock, n = 5) and PEG-IFNα2-treated group (huIFNAR-IFNα2, n = 6). In this figure, n = 5 or 6 biologically independent mice. (b) Kinetics of serum HBsAg (IU/ml). c Serum HBsAg at termination. d Kinetics of serum HBeAg (IU/ml). e Serum HBeAg measurement at termination. f HBsAg and HBeAg in the #3 huIFNAR mouse treated by PEG-IFNα2. g Serum of HBsAb levels. Intrahepatic HBV biomarkers at termination. Intrahepatic HBV pgRNA (h), HBV total mRNA (i) and HBV DNA ( j) levels were measured by qPCR. Two-sided ANOVA test or Kruskal–Wallis test was performed in (b–e) and (h–j). The p values in black represent comparisons between the huIFNAR-IFNα2 group and huIFNAR-Mock group, while the ones in blue are between the huIFNAR-IFNα2 group and WT-IFNα2 group, both adjusted for multiple comparisons. Data in (b–e) and (g–j) are presented as mean values ± SEM and their replicates same to (a). In c, e, and g–j, WT, Mock and IFNα2 represent WT-IFNα2, huIFNAR-Mock and huIFNAR-IFNα2, respectively. Source data are provided as a Source Data file.

Fig. 5. Characteristics of the intrahepatic monocytes post a 15-week interferon treatment at single-cell level.

a The uniform manifolds approximation projection (UMAP) of myeloid cell clusters. Eight populations visualized from 941 myeloid cells were indicated with different colors. b Population constitution analysis between MOCK (black) and IFNα2 (blue) treated groups, n = 2 cell samples examined independently, which were obtained and pooled from 2 or 3 mice. Data are presented as mean values ± SEM. c Volcano plots of differentially expressed genes in monocytes between MOCK and IFNα2 mice. Negative log2 fold change indicates downregulation (blue), positive log2 fold change indicates upregulation in IFNα2 (red) relative to MOCK mice. Genes with a log2 fold change between −0.5 and 0.5 are shown in gray. d Kyoto Encyclopedia of Genes and Genomes (KEGG) and gene ontology (GO) analysis of differentially expressed genes in IFNα2 versus MOCK livers. Top 4 significantly altered pathways are presented. e Heatmap of the enriched genes in the IFNα2-treated group. Biological processes are indicated. f–j Violin plots of gene set enrichment analysis scores between the huIFNAR-Mock and huIFNAR-IFNα2 groups. Data were analyzed using two-sided t-test. Only p values less than 0.05 are indicated, and n = total number of cells in each group. In e–j, WT, Mock and IFNα2 represent WT-IFNα2, huIFNAR-Mock and huIFNAR-IFNα2, respectively. Source data are provided as a Source Data file.

Fig. 6. Flow cytometry analysis of the intrahepatic monocyte and macrophages.

a, c Representative gating of specific immune cells population of MHCII⁺ Ly6c⁺Monocytes, proinflammatory (MHCII⁺ Ly6c^high) Macrophages, respectively. b, d The cell frequencies among C57BL/6J AAV-HBV PEG-IFNα2-treated wildtype group (WT-IFNα2, n = 6), huIFNAR AAV-HBV mock-treated group (huIFNAR-Mock, n = 5) and PEG-IFNα2-treated group (huIFNAR-IFNα2, n = 6). Data were analyzed using ANOVA test or Kruskal–Wallis test as appropriate. Data are presented as mean values ± SEM. In b, d, WT, Mock and IFNα2 represent WT-IFNα2, huIFNAR-Mock and huIFNAR-IFNα2, respectively. Source data are provided as a Source Data file.

Fig. 7. Characteristics of the intrahepatic CD8⁺ T Cells post a 15-week interferon treatment at single-cell level.

a The uniform manifolds approximation projection (UMAP) of CD8⁺ T-cell clusters. Six populations designated by specific gene expression visualized from 1045 intrahepatic CD8⁺ T cells were indicated with different colors. b Population constitution analysis between huIFNAR-Mock (black) and huIFNAR-IFNα2 (blue) treated groups, n = 2 cell samples examined independently, which were obtained and pooled from 2 or 3 mice. Data are presented as mean values ± SEM. c Volcano plots of differentially expressed genes in effector CD8⁺ T cells between huIFNAR-Mock and huIFNAR-IFNα2 mice. Negative log2 fold change indicates downregulation (blue), positive log2 fold change indicates upregulation in IFNα2 (red) relative to Mock mice. Genes with a log2 fold change between −0.5 and 0.5 are shown in gray. d GO and KEGG Biological Process analysis of differentially expressed genes in huIFNAR-IFNα2 group. Top 3–4 significantly altered pathways are presented. e Heatmap of the enriched genes in the huIFNAR-IFNα2-treated group. Biological processes are indicated. f, g Individual cell effector score and exhaust score overlay for selected differential canonical pathway activities (effect and depletion phenotype) in huIFNAR-Mock and huIFNAR-IFNα2 groups. h Violin plots of gene set enrichment analysis scores. Data were analyzed using two-sided t-test. Only p values less than 0.05 are indicated, and n = total number of cells in each group. In e–h, WT, Mock and IFNα2 represent WT-IFNα2, huIFNAR-Mock and huIFNAR-IFNα2, respectively. Source data are provided as a Source Data file.

Fig. 8. Flow cytometry analysis of the intrahepatic CD8⁺ T cells.

a, c, e Representative gating of specific immune cells population of effector memory CD8⁺ T (TEM), PD1⁺ CD8⁺T and PD1⁺CD8⁺TEM Cells, respectively. b, d, f The cell frequencies among C57BL/6J AAV-HBV PEG-IFNα2-treated wildtype group (WT-IFNα2, n = 6), huIFNAR AAV-HBV mock-treated group (huIFNAR-Mock, n = 5) and PEG-IFNα2-treated group (huIFNAR-IFNα2, n = 6). Data were analyzed using ANOVA test or Kruskal–Wallis test as appropriate. Data are presented as mean values ± SEM. In b, d, f, WT, Mock and IFNα2 represent WT-IFNα2, huIFNAR-Mock and huIFNAR-IFNα2, respectively. Source data are provided as a Source Data file.