Cell binding tropism of rat hepatitis E virus is a pivotal determinant of its zoonotic transmission to humans

Abstract

Classically, all hepatitis E virus (HEV) variants causing human infection belong to the genus Paslahepevirus (HEV-A). However, the increasing cases of rat HEV infection in humans since 2018 challenged this dogma, posing increasing health threats. Herein, we investigated the underlying mechanisms dictating the zoonotic potentials of different HEV species and their possible cross-protection relationships. We found that rat HEV virus-like particles (HEV^VLPs) bound to human liver and intestinal cells/tissues with high efficiency. Moreover, rat HEV^VLPs and infectious rat HEV particles penetrated the cell membrane and entered human target cells postbinding. In contrast, ferret HEV^VLPs showed marginal cell binding and entry ability, bat HEV^VLPs and avian HEV^VLPs exhibited no binding and entry potency. Structure-based three-dimensional mapping identified that the surface spike domain of rat HEV is crucial for cell binding. Antigenic cartography indicated that rat HEV exhibited partial cross-reaction with HEV-A. Intriguingly, sera of HEV-A infected patients or human HEV vaccine Hecolin® immunized individuals provided partial cross-protection against the binding of rat HEV^VLPs to human target cells. In summary, the interactions between the viral capsid and cellular receptor(s) regulate the distinct zoonotic potentials of different HEV species. The systematic characterization of antigenic cartography and serological cross-reactivity of different HEV species provide valuable insights for the development of species-specific diagnosis and protective vaccines against zoonotic HEV infection.

Keywords: cell binding tropism; hepatitis E virus; virus entry; zoonosis.

Fig. 1.

Generation and characterization of VLPs of HEV A-D. (A) The Neighbor-Joining tree was constructed based on the amino acid sequences of ORF2 protein of the reference HEV sequences (26, 27). The scale bar indicated the distances of ORF2 amino acid sequences. The colors on the branches indicate HEV genotypes. The colors of the outermost ring indicate HEV genera (A-D). Silhouettes represent the different hosts. (B) C-terminally his-tagged HEV-A ORF2^T was expressed in E. coli and purified. Nondenatured (N) and denatured (D) HEV-A ORF2^T proteins were analyzed by western blots with anti-his antibody. The dimer and multimeric proteins were indicated with black arrows. (C) Electron microscopic images of the self-assembled VLPs, named HEV-A^VLP. (D) Human liver tissue slides were incubated with PBS (NC), HEV-A^VLP, HEV-A^VLP + NC serum (1:20), HEV-A^VLP + anti-HEV-A^VLP (1:20) overnight at 4 °C. The binding of HEV-A^VLP to human liver tissue slides was detected by immunohistochemistry. (E–G) Same as (B–D) for the generation and characterization of HEV-B ORF2^T and HEV-B^VLP. (H–J) Same as (B–D) for the generation and characterization of HEV-C1 ORF2^T and HEV-C1^VLP. (K–M) Same as (B–D) for the generation and characterization of HEV-C2 ORF2^T and HEV-C2^VLP. (N–P) Same as (B–D) for the generation and characterization of HEV-D ORF2^T and HEV-D^VLP. The images presented are the representative of three independent experiments.

Fig. 2.

Characterization of cell-binding capability of HEV A-D VLPs. (A) The binding capability of HEV-A^VLP, HEV-B^VLP, HEV-C1^VLP, HEV-C2^VLP, and HEV-D^VLP to human liver cell lines (HepG2 and HuH7), human intestinal cell line (Caco2), and rat live cell line (RH-35) was measured by immunofluorescence assay. (B) The relative fluorescent intensity was quantified by ImageJ. The fluorescent intensity of HEV-A VLP was set as 100. The significance of difference between HEV-A^VLP and the other groups was calculated based on three independent experiments. (C) Human liver tissue slides were incubated with PBS (NC), HEV-C1^VLP, HEV-C1^VLP + NC serum (1:20), HEV-C1^VLP + anti-HEV-C1^VLP (1:20) overnight at 4 °C. The binding capacity of HEV-C1^VLP was detected by immunohistochemistry. (D) Same as (C) for testing the binding of HEV-C1^VLP to human intestinal tissue slides. The images presented are representative of three independent experiments.

Fig. 3.

Characterization of cell entry capability of HEV-C1^VLP and rat HEV. (A) HepG2 and RH-35 cells were inoculated with the same amounts of VLPs. The entry of HEV-A^VLP, HEV-B^VLP, HEV-C1^VLP, HEV-C2^VLP, and HEV-D^VLP to HepG2 and RH-35 cells was detected by confocal immunofluorescence assay. Bright-field microscopy channel was used to determine outlines of adherent cells. Cell boundary was illustrated with white dashed lines. (B and C) The intracellular positive fluorescent dots of five HEV VLPs were quantified based on three independent experiments. (D) HepG2 cells were inoculated with mock or human HEV-3 virus (~3.3 × 10³ copy/cell) with rat NC serum (1:20), or anti-HEV-A^VLP serum (1:20) overnight. The levels of intracellular HEV-3 RNA were quantified by RT-qPCR assay based on three independent experiments. n.d., not detected. (E) Same as (D), RT-qPCR detection of intracellular levels of rat HEV RNA in HepG2 cells post rat HEV (~1.3 × 10³ copy/cell) inoculation. (F) Same as (E) for the detection of intracellular rat HEV RNA in RH-35 cells by RT-qPCR. (G–I) Same as (D–F), ORF2 protein immunostaining and DAPI counterstaining was performed 6 d post inoculation. The percentage of ORF2 positive cells were quantified based on three independent experiments.

Fig. 4.

Mapping of amino acid residues crucial for the cell-binding of HEV-C1^VLP. (A) Amino acid sequences of HEV-C1 ORF2^T. 24 HEV-C1 ORF2^T mutants were created. The amino acid substitutions at the surface of the P2 domain were labeled with red color. (B) His-tagged HEV-C1 ORF2^T mutants were expressed and purified. Nondenatured (N) and denatured (D) mutants were analyzed by SDS-PAGE. The images presented are the representative of three independent experiments. (C) Structural model of HEV-C1 ORF2^T dimer (Top view). The substitutions in the P2 domain essential for VLP formation were shown in cyan. The P2 and truncated P1 domains of the Right-side monomer are violet and yellow, whereas the domains on the Left-side monomer are light pink and lemon. (D) Same as (B) for the analysis of nondenatured (N) and denatured (D) mutants. (E) The cell-binding of HEV-C1^VLP mutants to RH-35 cells was measured by immunofluorescence assay based on three independent experiments. (F) The 3D structural model of HEV-C1 ORF2^T dimer (side view). The substitutions in the P2 domain essential for target cell-binding are shown in red.

Fig. 5.

The binding specificity of HEV-C1^VLP on live and intestinal tissues. HEV-C1^VLP was preincubated with rat NC serum (dilution 1:20) or anti-HEV-C1^VLP (dilution 1:20) for 45 min at room temperature. (A) rat duodenum, jejunum, ileum, or colon tissue slides were incubated with PBS (NC), HEV-C1^VLP (WT), HEV-C1^VLP (WT)+ rat NC serum, HEV-C1^VLP (WT)+ anti-HEV-C1^VLP, HEV-C1^VLP M21, M2, or M10 overnight at 4 °C. The binding specificity of VLPs was detected by Immunohistochemistry. The images presented are the representative of three independent experiments. (B) Scheme of rat infection experiment. Group 1 (n = 5) is given high-dose immunosuppressive drug regimen from −10 d to day 0 and inoculated intravenously with 200ul rat HEV (SRN250811, 10⁶ copies/mL) stool filtrate on day 0, Group 2 (n = 2) rats were administered PBS only on day 0. (C) Stool samples were collected on days 0, 7, 14, 21, and 28. The rat HEV viral load were determined by RT-qPCR (n = 5). (D) The liver and gut tissues were collected at day 28. Rat HEV viral load was determined by RT-qPCR (n = 5). (E) The gut tissues were collected from rat HEV infected rats (n = 5) or mock-infected rats (n = 2) at day 28 and subjected to immunohistochemistry analysis of rat HEV ORF2 protein.

Fig. 6.

HEV-C1^VLP and HEV-A^VLP possess closer mutual immune cross-reactivity and partial bilateral cross-protection. (A) Two-dimensional antigenic mapping of HEV-A^VLP, HEV-C1^VLP, HEV-C2^VLP, HEV-B^VLP, and HEV-D^VLP with rat sera revealed clustering relationships among different HEV genera. The antigen–antibody cross-reactivity of HEV-A^VLP and HEV-C1^VLP partially overlaps. The vertical and horizontal axes both represent antigenic distance. One unit of antigenic distance denotes a twofold difference in sera titers. The solid diamond, hexagon, circle, square, and pentagon represent A, C1, C2, B, and D, respectively, whereas the hollow ones correspond to the immunized rat serum of each antigen. (B and C) HEV-A^VLP was preincubated with PBS, rat NC serum (dilution 1:20), rat anti-HEV-A^VLP (dilution 1:180 or 1:20), or rat anti-HEV-C1^VLP (dilution 1:20). The ability to block the binding of HEV-A^VLP to HuH7 cells was detected by immunofluorescence assay. The relative fluorescent intensity was quantified by ImageJ based on three independent experiments. Green, HEV-A ^VLP; blue, nucleus. (D and E) HEV-A^VLP was preincubated with rat NC serum (dilution 1:20), anti-HEV-C1^VLP (dilution 1:20), or anti-HEV-A^VLP (dilution 1:20). The ability to block the binding of HEV-A^VLP to HepG2 cells was measured by immunofluorescence assay. The relative fluorescent intensity was quantified by ImageJ based on three independent experiments. Red, HEV-A ^VLP; blue, nucleus. (F and G) Same as (B and C) for testing the ability to block the binding of HEV-C1^VLP to HuH7 cells. (H and I) Same as (D and E) for testing the ability to block the binding of HEV-C1^VLP to RH-35 cells.