The feasibility of establishing a hamster model for HBV infection: in vitro evidence

Abstract

Chronic hepatitis B virus (HBV) infection remains a significant public health burden with no cure currently available. The research to cure HBV has long been hampered by the lack of immunocompetent small animal models capable of supporting HBV infection. Here, we set out to explore the feasibility of the golden Syrian hamster as an immunocompetent small rodent model for HBV infection. We first started with in vitro assessments of the HBV replication cycle in primary hamster hepatocytes (PHaHs) by adenoviral HBV (Ad-HBV) transduction. Our results demonstrated that PHaHs support HBV reverse transcription and subsequent cccDNA formation via the intracellular recycling pathway. Next, with luciferase reporter assays, we confirmed that PHaHs support the activities of all HBV major promoters. Then, we transduced PHaHs with an adenoviral vector expressing HBV receptor human Na+/taurocholate cotransporting polypeptide NTCP (Ad-huNTCP), followed by HBV inoculation. While the untransduced PHaHs did not support HBV infection, Ad-huNTCP-transduced PHaHs supported de novo cccDNA formation, viral mRNA transcription, and expression of viral antigens. We then humanized the amino acid (aa) residues of hamster NTCP (haNTCP) critical for HBV entry, aa84-87 and aa157-165, and transfected HepG2 cells with constructs expressing wild-type haNTCP and humanized-haNTCP, H84R/P87N and H84R/P87N/G157K/M160V/M165L, respectively, followed by HBV inoculation. The results showed that the humanization of H84R/P87N alone was sufficient to support HBV infection at a level comparable to that supported by huNTCP. Taken together, the above in vitro evidence supports the future direction of humanizing haNTCP for HBV infection in vivo.IMPORTANCEOne of the biggest challenges in developing an HBV cure is the lack of immunocompetent animal models susceptible to HBV infection. Developing such models in mice has been unsuccessful due to the absence of a functional HBV receptor, human NTCP (huNTCP), and the defect in supporting viral cccDNA formation. In search of alternative models, we report herein multiple lines of in vitro evidence for developing a golden Syrian hamster model for HBV infection. We demonstrate that the primary hamster hepatocytes (PHaHs) support HBV replication, transcription, and cccDNA formation, and PHaHs are susceptible to de novo HBV infection in the presence of huNTCP. Furthermore, expressing hamster NTCP with two humanized residues critical for HBV entry renders HepG2 cells permissive to HBV infection. Thus, our work lays a solid foundation for establishing a gene-edited hamster model that expresses humanized NTCP for HBV infection in vivo.

Keywords: NTCP; cccDNA; hamster hepatocytes; hepatitis B virus.

Fig 1

Primary hamster hepatocytes (PHaHs) support HBV cccDNA formation and huNTCP-mediated HBV infection. (A) HepG2 cells were either left untransduced (lane 1) or transduced by Ad-HBV at increasing MOIs of 20 (lane 2), 40 (lane 3), 80 (lane 4), 160 (lane 5), and 320 (lane 6) for 6 days. HBV cytoplasmic core DNA (DNA replicative intermediates) (top panel) and cccDNA (bottom panel) were detected by Southern blot. For cccDNA detection, Hirt DNA samples were heat denatured at 90°C and then digested by EcoRI, by which turned the deproteinated relaxed circular DNA (DP-rcDNA) into single-stranded (SS) DNA and cccDNA was linearized into double-stranded linear (DSL) form, a common practice to highlight cccDNA on Southern blot. The 3.2 kb unit-length linear HBV DNA served as size marker (lane 7). (B) PHaHs were transduced with Ad-HBV at MOI of 200 for 6 days. Intracellular HBV core DNA and Hirt DNA were extracted and subjected to Southern blot analysis. (C) The Hirt DNA sample from (B) was aliquoted and subjected to various treatments as schematically illustrated (top panel) and subsequent Southern blot assay (bottom panel), including untreated (lane 2), 90°C heat denaturation for 5 min (lane 3), 90°C heat denaturation followed by EcoR I linearization (lane 4), or directly digested by exonuclease I and III (ExoI/III) (lane 5). CM-rcDNA: closed minus-strand rcDNA. (D) Nonhepatic 293T cells, PHaHs, and hepatoma HepG2 cells were co-transfected by each indicated HBV promoter firefly luciferase (Luc) reporter plasmid and CMV promoter Renilla luciferase reporter plasmid pRL-CMV (20:1 ratio) for 3 days, followed by dual luciferase assay. The firefly luciferase signals were normalized by Renilla luciferase signals and plotted as fold change against that in 293T cells (mean ± SD, n = 3; *P < 0.05, **P < 0.01, ***P < 0.001). EnII/Cp: enhancer II and core promoter, S1p: S1 promoter; S2p: S2 promoter; EnI/Xp: enhancer I and X promoter. (E, F) PHaH cells were transduced with Ad-huNTCP-HA at MOI of 20 for 2 days. The expression of huNTCP-HA was analyzed by immunofluorescence (E) and Western blot (F). Cell nuclei were counter-stained by DAPI (E) and β-actin served as loading control (F). (G–K) PHaH cells were transduced with Ad-huNTCP-HA at MOI of 20 for 2 days, followed by HBV infection at MOI of 500 for 3 days. The intracellular cccDNA was quantified by qPCR and normalized by hamster mitochondrial DNA (mean ± SD, n = 3; ***P < 0.001) (G); HBV 3.5 kb pc/pgRNA was analyzed by RT-qPCR and normalized by hamster GAPDH mRNA, data are presented as fold change versus mock control (mean ± SD, n = 3; ***P < 0.001) (H); intracellular HBc was detected by immunofluorescence (I); and HBeAg (J) and HBsAg (K) in the supernatant were measured by CLIA (mean ± SD, n = 3; ***P < 0.001).

Fig 2

Humanized hamster NTCP (haNTCP) supports HBV infection in vitro. (A) Membrane topology of NTCP, a nine transmembrane (TM) protein (top panel). The sequence alignment highlights the aa 84–87 and 157–165 domains of NTCP protein among human (NP_003040.1), golden Syrian hamster (XP_005072871.1), mouse (NP_035517.1), and rhesus monkey (ALX38773.1) (bottom panel). The polymorphisms from hamster, mouse, and monkey NTCP in these two domains are indicated in red. Created with Biorender.com. (B–D) HepG2 cells were transfected with empty vector, wild-type haNTCP (haNTCPwt-Flag), humanized haNTCP (haNTCPmut84-87-Flag and haNTCPmut84-87/157-164-Flag), or human NTCP (huNTCP-C9) for 2 days; the cells were subjected to haNTCP-Flag and huNTCP-C9 immunofluorescence (B) and Western blot (C) using anti-Flag and anti-C9 antibodies, respectively. (D) Another set of transfected cells was probed with preS1-TAMRA, followed by haNTCP-Flag and huNTCP-C9 immunofluorescence for detection of preS1 and NTCP colocalization by confocal microscopy. (E–G) The above transfected HepG2 cells were infected with HBV (MOI: 500) for 3 days, the cells were collected for cccDNA qPCR and normalized by human mitochondrial DNA qPCR (mean ± SD, n = 3; ***P < 0.001) (E), and the supernatant samples were subjected to HBeAg CLIA (mean ± SD, n = 3; *P < 0.05, **P < 0.01) (F). (G) Another set of transfected HepG2 cells was left untreated or treated with HBV entry inhibitor MyrB (500 nM) during HBV infection (MOI: 500), and 3 days later, the cells were subjected to HBc immunofluorescence.