Development of Recombinant Infectious Hepatitis E Virus Harboring the nanoKAZ Gene and Its Application in Drug Screening

ABSTRACT

Hepatitis E virus (HEV) is a quasi-enveloped virus with a single-stranded positive-sense RNA genome belonging to the family Hepeviridae. Studies of the molecular aspects of HEV and drug screening have benefited from the discovery of bioluminescent reporter genes. However, the stability of large foreign genes is difficult to maintain after insertion into the viral genome. Currently, ribavirin is used to treat HEV-infected patients who require antiviral therapy. This has several major drawbacks. Thus, the development of novel anti-HEV drugs is of great importance. We developed a system consisting of recombinant infectious HEV harboring a small luciferase gene (nanoKAZ) in the hypervariable region (HVR) of the open reading frame 1 (ORF1) (HEV-nanoKAZ). It replicated efficiently in cultured cells, was genetically stable, and had morphological characteristics similar to those of the parental virus. Both membrane-associated (eHEV-nanoKAZ) and membrane-unassociated (neHEV-nanoKAZ) particles were infectious. HEV particles circulating in the bloodstream and attaching to hepatocytes in HEV-infected patients are membrane-associated; thus, eHEV-nanoKAZ was applied in drug screening. The eHEV-nanoKAZ system covers at least the inhibitor of HEV entry and inhibitor of HEV RNA replication. Four drugs with anti-HEV activity were identified. Their effectiveness in cultured cells was confirmed in naive and HEV-producing PLC/PRF/5 cells. Two hit drugs (azithromycin and ritonavir) strongly inhibited HEV production in culture supernatants, as well as intracellular expression of ORF2 protein, and may therefore be candidate novel anti-HEV drugs. The HEV-nanoKAZ system was developed and applied in drug screening and is expected to be useful for investigating the HEV life cycle.

IMPORTANCE Bioluminescent reporter viruses are essential tools in molecular virological research. They have been widely used to investigate viral life cycles and in the development of antiviral drugs. For drug screening, the use of a bioluminescent reporter virus helps shorten the time required to perform the assay. A system, consisting of recombinant infectious HEV harboring the nanoKAZ gene in the HVR of ORF1 (HEV-nanoKAZ), was developed in this study and was successfully applied to drug screening in which four hit drugs with anti-HEV activity were identified. The results of this study provide evidence supporting the use of this system in more variable HEV studies. In addition, both forms of viral particles (eHEV-nanoKAZ and neHEV-nanoKAZ) are infectious, which will enable their application in HEV studies requiring both forms of viral particles, such as in the investigation of unknown HEV receptors and the elucidation of host factors important for HEV entry.

INTRODUCTION

Hepatitis E virus (HEV) is a quasi-enveloped virus with a single-stranded positive-sense RNA genome and is a member of family Hepeviridae (1). The genome length varies from 6.4 to 7.3 kb, and it has a short 5′ untranslated region (5′ UTR) capped at its 5′ end and a short 3′ UTR terminated by the poly(A) tract (2, 3). The HEV genome encompasses three major open reading frames (ORFs)—ORF1, which encodes a nonstructural polyprotein containing multiple functional domains involved in viral replication, including methyltransferase, the Y domain, papain-like cysteine protease, the hypervariable region (HVR) or proline-rich region, the X domain (or macro domain), helicase, and RNA-dependent RNA polymerase (4, 5); ORF2, which encodes capsid protein essential during virion assembly and viral attachment to host cells and is a major target for neutralizing antibodies (6–8); and ORF3, which is important in virion egress (9–11) and is a functional ion channel acting as a viroporin (12). There are two distinct forms of HEV particles. HEV particles present in bile and those shed in the feces are membrane-unassociated, while HEV particles in the bloodstream and culture supernatants are membrane-associated (9, 13–16). HEV is increasingly recognized as the leading cause of acute hepatitis. Although most HEV infections are self-limiting, a chronic course may develop, especially in immunocompromised patients (17). HEV is distributed worldwide, where it is transmitted mainly through the fecal-oral route or zoonotic foodborne route and via other less frequent routes, such as organ transplantation, transfusion of blood products, or from an infected mother to her fetus (18).

In various virological studies, bioluminescent reporter viruses have been applied to investigate the viral life cycle, characterize viral proteins, and perform bioluminescence-based molecular imaging (19–21). In addition, it is an essential part in drug discovery and development, where it allows rapid and sensitive detection of analytes (22). In the early years of their discovery, bioluminescent systems were mostly based on firefly luciferase and Renilla luciferase. However, the size, stability, and luminescence efficiency limit their applications; thus, researchers have gradually switched to those with smaller size, higher stability, and brighter luminescence (19). One of the newest additions to the luciferase enzyme systems is nanoKAZ (23–25). It is a mutated component of Oplophorus luciferase, which has an amino acid sequence identical to that of NanoLuc but a different nucleotide sequence (23, 26). Additionally, the nanoKAZ system has been reported to show enhanced stability and over 150-fold higher luminescence in comparison to firefly and Renilla luciferases (23–25). With a size of only 19 kDa, the insertion of this gene into the HEV genome is advantageous, considering that the insertion of a large foreign gene for fluorescence is often difficult to maintain (19, 20, 27). Recombinant HEVs harboring smaller bioluminescence genes and their application in HEV molecular studies have also been reported, such as the use of recombinant HEV harboring hemagglutinin epitope tag or NanoLuc in the ORF1 protein to study HEV replication complex (21), or HEV replicon expressing Gaussia luciferase (GLuc) to study the function of ORF3 protein or its application in drug screening (12, 28).

Currently, no specific antiviral drugs for HEV are available. Drugs such as ribavirin and pegylated interferon α (peg-IFN-α) have been used in certain cases of clinical HEV infection, such as chronic cases or acute fulminant cases (29, 30). However, they are associated with severe side effects and are contraindicated in pregnant women, who represent a major risk group (31). In addition, ribavirin has been reported to be associated with the development of resistance (32–36). Therefore, the development of novel and specific antiviral drugs against HEV is urgently needed in order to provide more treatment options.

Our recently developed HEV replicon expressing GLuc, was able to screen drugs that inhibit HEV RNA replication (28) and was applied in the analysis of anti-HEV candidates and their combinations in vitro (37). In the present study, to broaden the scope of screening, a system consisting of the entire HEV genome harboring the nanoKAZ gene in the HVR of ORF1 was developed (HEV-nanoKAZ), since the HVR of ORF1 has been reported to accommodate either naturally acquired insertions (e.g., insertions identified in acute or chronic HEV patients and in swine [38–44]) or insertions made by genetic recombination (21). HEV-nanoKAZ replicated efficiently in cultured cells, was genetically stable, and had morphological characteristics similar to the parental virus. In particular, both membrane-associated (eHEV-nanoKAZ) and membrane-unassociated (neHEV-nanoKAZ) particles, which were obtained following treatment of eHEV-nanoKAZ with deoxycholate and trypsin, were infectious. This screening system has the ability to at least monitor the inhibitors of both HEV entry and HEV RNA replication. The four hit drugs selected by drug screening in this study were confirmed to be effective in a cell culture system. The successful development of the HEV-nanoKAZ system and its use in comprehensive drug screening, as well as the availability of both infectious forms (eHEV-nanoKAZ and neHEV-nanoKAZ), have opened the door to expand its application not only in drug discovery and development, but also in studies of the molecular aspects of HEV that require both forms of viral particles, such as investigations of unknown HEV receptors and studies to elucidate host factors that are important for HEV entry.

RESULTS

Construction of the recombinant infectious HEV harboring the nanoKAZ gene.

To construct a recombinant infectious HEV carrying the nanoKAZ gene in ORF1, an infectious cDNA clone of pJE03-1760F/P10 was used as a template. It is based on a cell-culture-adapted HEV-3 strain, generated after 10 consecutive passages of the wild-type strain and has enhanced HEV production in comparison to the wild-type infectious cDNA clone (pJE03-1760F/wt) (45). Insertion of a foreign gene for fluorescence into a virus genome is often difficult to maintain due to the large size of such genes (19, 20, 27); thus, the small luciferase gene (nanoKAZ, 19 kDa) was selected for this study. The 171-amino acid (aa) nanoKAZ gene was inserted into five distinct sites within HVR, where the constructed plasmids were named after their orderly numbered insertion sites (pJE03-1760F/P10-nanoKAZ1, -nanoKAZ2, -nanoKAZ3, -nanoKAZ4, and -nanoKAZ5) (Fig. 1A). The insertion sites within the HVR that have been reported in various HEV-3-infected humans or pigs with at least 30-aa insertion (Table 1) were used as references in the current study.

FIG 1.

(A) Schematic diagram of the full-length cDNA clone of the JE03-1760F/P10 strain (pJE03-1760F/P10) and its derivative recombinants. Arrows indicate five insertion sites of the nanoKAZ gene within the HVR of ORF1. MeT, methyl transferase; Y, Y domain; PCP, papain-like cysteine protease; HVR, hypervariable region; X, macro domain; Hel, helicase; RdRp, RNA-dependent RNA polymerase. (B) Quantification of HEV RNA in culture supernatants. RNA transcripts of pJE03-1760F/P10-nanoKAZ1, -nanoKAZ2, -nanoKAZ3, -nanoKAZ4, and -nanoKAZ5 were transfected to PLC/PRF/5 cells to examine their replication efficiency, along with RNA transcripts of pJE03-1760F/P10 (control) and its replication-defective mutant, pJE03-1760F/P10-GAA (negative control). HEV growth was observed for 28 days. The data are presented as the mean ± standard deviation (SD) for two wells each. The dotted horizontal line represents the limit of detection by real-time RT-PCR used in this study, at 2 × 10¹ RNA copies/mL. (C) Intracellular luciferase activity in the cells 28 days after RNA transfection was measured to examine the specificity of nanoKAZ activity. The data are presented as the mean ± SD for two wells each. (D) Gel electrophoresis of the RT-PCR products, which covered the HVR containing the nanoKAZ insertion, from purified HEV RNA in the culture supernatants on day 28 posttransfection. The transfection experiment was performed in two wells (well 1 [w1] and well 2 [w2]), where the culture supernatants from each well were purified and subjected to RT-PCR. The size of the PCR product with no nanoKAZ insertion was 661 bp, while the size of the PCR product was 1,174 bp when the insertion was intact.

TABLE 1.

The insertion sites reported in various HEV-3-infected humans or pigs with at least 30-aa insertion and the inserted sites in the HVR in the current study

Genotypes	GenBank ID (nucleotide)	GenBank ID (ORF1 protein)	ORF1 protein length (aa insertion)a	Hosts	Inserted sites
3a	JN564006	AEX32827.1	1,736 (33) aa	Human	D739/I740 (Fig. 1A, 1)
3a	JQ679013	AFD33683.1	1,765 (62) aa	Human	E748/E749 (Fig. 1A, 2)
3f	KC166971	AGT96625.1	1,759 (56) aa	Human	E748/E749 (Fig. 1A, 2)
3f	EU495148	ACD62590.1	1,733 (30) aa	Human	E748/E749 (Fig. 1A, 2)
3f	JN906976	AFH35002.1	1,733 (30) aa	Swine	E748/E749 (Fig. 1A, 2)
3c	KC618402	AHC54587.1	1,765 (62) aa	Human	G765/L766 (Fig. 1A, 3)
3f	AB850879	BAQ95549.1	1,733 (30) aa	Human	P772/P773 (Fig. 1A, 4)
3f	EU723514	ACH87959.1	1,733 (30) aa	Swine	P773/P774 (Fig. 1A, 5)
3f	EU723516	ACH87965.1	1,733 (30) aa	Swine	P773/P774 (Fig. 1A, 5)

^aaa, amino acids.

To monitor virus production, HEV RNA levels in the culture supernatants of PLC/PRF/5 cells transfected with RNA transcripts of the five recombinant HEVs were quantitated. The HEV RNA levels in the culture supernatants gradually increased until day 28 posttransfection (Fig. 1B). However, in comparison to those in the culture supernatants of cells transfected with RNA transcripts of the parental clone (pJE03-1760F/P10), which reached 2.6 × 10⁸ copies/mL on day 28 posttransfection, the HEV RNA levels were 10-fold lower in culture supernatants of cells transfected with RNA transcripts of pJE03-1760F/P10-nanoKAZ1, -nanoKAZ2, and -nanoKAZ3 (1.7 × 10⁷, 2.2 × 10⁷, and 1.8 × 10⁷ copies/mL, respectively) and 1,000-fold lower in culture supernatants of cells transfected with RNA transcripts of pJE03-1760F/P10-nanoKAZ4 and -nanoKAZ5 (1.3 × 10⁵ and 3.2 × 10⁵ copies/mL, respectively) (Fig. 1B). Meanwhile, a gradual decrease in HEV RNA titer was observed in culture supernatants of cells transfected with RNA transcripts of a replication-defective mutant (pJE03-1760F/P10-GAA), which expressed functionally disrupted RNA-dependent RNA polymerase (RdRp) (Fig. 1B).

To evaluate the specificity of nanoKAZ activity, cells transfected with RNA transcripts of the five recombinant HEVs were lysed on day 28 posttransfection, and the intracellular luciferase activity was determined. Consistent with the HEV RNA levels in the culture supernatants, the intracellular luciferase activity in cells transfected with pJE03-1760F/P10-nanoKAZ1, -nanoKAZ2, and -nanoKAZ3 RNAs (2.0 × 10⁷, 3.2 × 10⁷, and 2.7 × 10⁷ relative light units [RLU], respectively) was 100-fold higher in comparison to cells transfected with pJE03-1760F/P10-nanoKAZ4 RNA (2.0 × 10⁵ RLU) and 10-fold higher in comparison to cells transfected with pJE03-1760F/P10-nanoKAZ5 RNA (1.3 × 10⁶ RLU), while no luciferase activity was detected in the lysates of cells transfected with either parental virus or its GAA mutant RNAs (Fig. 1C). Taken together, among the five recombinant HEVs, pJE03-1760F/P10-nanoKAZ1, -nanoKAZ2, and -nanoKAZ3 demonstrated not only higher replication efficiency in comparison to pJE03-1760F/P10-nanoKAZ4 and -nanoKAZ5, but also higher intracellular luciferase activity.

To confirm the stability of the nanoKAZ insertion, purified HEV RNAs from culture supernatants of cells transfected with pJE03-1760F/P10-nanoKAZ1, -nanoKAZ2, and -nanoKAZ3 RNAs (day 28 posttransfection) were subjected to reverse transcriptase PCR (RT-PCR), which covered HVR containing the nanoKAZ insertion (Fig. 1D). The size of the PCR product with no nanoKAZ insertion is 661 bp, while it is 1,174 bp if the nanoKAZ insertion is intact. Gel electrophoresis of the PCR products indicated that the band size for two representative wells (well 1, w1; well 2, w2) of pJE03-1760F/P10-nanoKAZ1 was similar to that of pJE03-1760F/P10 (Fig. 1D), suggesting that the nanoKAZ insertion was no longer intact. For pJE03-1760F/P10-nanoKAZ2, although the PCR product of w1 banded at the expected size, indicating an intact nanoKAZ insertion, the PCR product of w2 banded at the same size as that of pJE03-1760F/P10 (Fig. 1D). At the same time, the PCR product for two representative wells of pJE03-1760F/P10-nanoKAZ3 banded at the expected size for intact nanoKAZ insertion (Fig. 1D). Sequence analyses were performed on all wells to confirm these results. Those with the same size as JE03-1760F/P10 lost the nanoKAZ insertion, while those with a longer size had an intact nanoKAZ insertion, in addition to pJE03-1760F/P10-nanoKAZ4 and -nanoKAZ5 which also failed to maintain the nanoKAZ insertion (data not shown). Based on these results, only pJE03-1760F/P10-nanoKAZ3 exhibited stable nanoKAZ insertion.

To evaluate the association between replication ability and the expression of nanoKAZ, a replication-defective mutant (pJE03-1760F/P10-nanoKAZ3-GAA), in which the catalytic residues Asp1561 and Asp1562 of the GDD motif in ORF1 were substituted with alanine into GAA, was constructed. RNA transcripts of pJE03-1760F/P10-nanoKAZ3 and pJE03-1760F/P10-nanoKAZ3-GAA were transfected to PLC/PRF/5 cells. During 28 days of observation, the HEV RNA levels in the culture supernatants of cells transfected with pJE03-1760F/P10-nanoKAZ3 RNA continued to increase, reaching 8.6 × 10⁶ copies/mL on the final day of observation, while those in culture supernatants of cells transfected with pJE03-1760F/P10-nanoKAZ3-GAA RNA gradually decreased (Fig. 2A). Culture supernatants of cells transfected with pJE03-1760F/P10-nanoKAZ3 and pJE03-1760F/P10-nanoKAZ3-GAA RNAs were then subjected to Western blotting to examine the expression of HEV proteins. At 28 days posttransfection, the specific band of ORF2 protein was only detected in the culture supernatants of cells transfected with pJE03-1760F/P10-nanoKAZ3 RNA and was undetectable in the culture supernatants of the pJE03-1760F/P10-nanoKAZ3-GAA RNA-transfected cells (Fig. 2B, left panel). In line with this result, the ORF3 protein was detected only in the culture supernatants of pJE03-1760F/P10-nanoKAZ3 RNA-transfected cells and was undetectable in those of the pJE03-1760F/P10-nanoKAZ3-GAA RNA-transfected cells (Fig. 2B, right panel). To check the intracellular expression of the ORF2 protein, cells at 28 days posttransfection were subjected to an immunofluorescence assay (IFA). ORF2 protein was expressed abundantly in the cells transfected with pJE03-1760F/P10-nanoKAZ3 RNA (Fig. 2C, upper panel), in contrast to the pJE03-1760F/P10-nanoKAZ3-GAA RNA-transfected cells, in which the expression of ORF2 protein was undetectable (Fig. 2C, lower panel). The specificity of the nanoKAZ expression was determined using cell lysates at 2, 4, 8, 12, 16, 20, and 28 days posttransfection. Significant luciferase activity was detected in the lysates of pJE03-1760F/P10-nanoKAZ3 RNA-transfected cells, which increased over time and reached 1.6 × 10⁷ RLU at day 28 posttransfection, while it was at a steadily low level in the lysates of pJE03-1760F/P10-nanoKAZ3-GAA RNA-transfected cells (Fig. 2D). These results demonstrated the replication ability of JE03-1760F/P10-nanoKAZ3 and the virus replication-dependent expression of nanoKAZ.

FIG 2.

Association between replication ability and the nanoKAZ expression of JE03-1760F/P10-nanoKAZ3. (A) HEV growth during 28 days of observation after transfection of RNA transcripts of pJE03-1760F/P10-nanoKAZ3 along with its replication-defective mutant, pJE03-1760F/P10-nanoKAZ3-GAA. The data are presented as the mean ± SD for two independent experiments. The dotted horizontal line represents the limit of detection by real-time RT-PCR used in this study, at 2 × 10¹ RNA copies/mL. (B) Western blot analysis of the culture supernatants of cells transfected with pJE03-1760F/P10-nanoKAZ3 RNA or the pJE03-1760F/P10-nanoKAZ3-GAA RNA to examine the expression of HEV ORF2 (left panel) and ORF3 proteins (right panel). (C) Immunofluorescence staining of the cells transfected with pJE03-1760F/P10-nanoKAZ3 RNA (upper panel) or the pJE03-1760F/P10-nanoKAZ3-GAA RNA (lower panel) on day 28 following transfection, to examine the HEV ORF2 protein expression. Results representative of one of two experiments are shown. (D) Intracellular luciferase activity of the cells transfected with pJE03-1760F/P10-nanoKAZ3 RNA or the pJE03-1760F/P10-nanoKAZ3-GAA RNA on day 2, 4, 8, 12, 16, 20, 24, and 28 posttransfection. dpt, days posttransfection.

Characterization of recombinant membrane-associated and membrane-unassociated JE03-1760F/P10-nanoKAZ3 (eHEV-nanoKAZ3 and neHEV-nanoKAZ3).

To confirm that the morphological characteristics of virus particles generated in culture supernatants of pJE03-1760F/P10-nanoKAZ3 RNA-transfected PLC/PRF/5 cells are similar to those of their parental virus, the culture supernatants were subjected to equilibrium centrifugation in a sucrose density gradient. The peak density was observed at 1.16 g/mL (Fig. 3A), similar to the density of membrane-associated HEV particles derived from culture supernatants (11, 13). To clarify this finding, culture supernatants of the pJE03-1760F/P10-nanoKAZ3 RNA-transfected cells were treated with sodium deoxycholate (DOC-Na) and trypsin to remove the lipid membrane and ORF3 protein. After treatment, the peak density shifted to 1.27 g/mL (Fig. 3A), which was similar to that of membrane-unassociated HEV particles (14, 46), suggesting that the virus particles in culture supernatants of cells transfected with pJE03-1760F/P10-nanoKAZ3 RNA are membrane-associated. To validate this observation, the culture supernatants of pJE03-1760F/P10-nanoKAZ3 RNA-transfected cells were purified and then negatively stained and subjected to transmission electron microscopy (TEM). Virus-like particles in the culture supernatants were observed with identical morphology and size (diameter, ∼40 nm) (Fig. 3B, left panel) to membrane-associated HEV particles (16). In addition, following treatment with DOC-Na and trypsin, the virus-like particles (Fig. 3B, middle panel) in the culture supernatants were observed to have a morphology and size (diameter, ∼30 nm) that were comparable to membrane-unassociated HEV particles (16). To further clarify that these particles are membrane-unassociated HEV, they were subjected to immunoelectron microscopy (IEM) using immunogold labeling. The virus-like particles were clearly coated with anti-human HEV ORF2 monoclonal antibody (MAb) (Fig. 3B, right panel), confirming that they are membrane-unassociated HEV particles. Collectively, these data indicated that the insertion of nanoKAZ gene into HVR does not affect the particle formation of HEV.

FIG 3.

Characterization of recombinant membrane-associated and membrane-unassociated JE03-1760F/P10-nanoKAZ3 (eHEV-nanoKAZ3 and neHEV-nanoKAZ3). (A) Culture supernatants of PLC/PRF/5 cells transfected with pJE03-1760F/P10-nanoKAZ3 RNA were subjected to equilibrium centrifugation in a sucrose density gradient. The peak density observed at 1.16 g/mL was of the eHEV-nanoKAZ3 (closed circle), while the peak density at 1.27 g/mL was of the neHEV-nanoKAZ3 (closed diamond) generated following treatment of the culture supernatants with sodium deoxycholate (DOC-Na) and trypsin. (B) Transmission electron microscopy (TEM) of the negatively stained eHEV-nanoKAZ3 (left panel) and neHEV-nanoKAZ3 (middle panel) and immunogold labeling of the neHEV-nanoKAZ3 with anti-HEV ORF2 monoclonal antibody (MAb) (right panel). Bars, 50 nm. All images are at 20,000× magnification. Results representative of one of two experiments are shown. (C) Immunofluorescence staining of cells inoculated with eHEV-nanoKAZ3 and neHEV-nanoKAZ3 in comparison to cells inoculated with nonrecombinant (nr)-eHEV or nr-neHEV, using anti-HEV ORF2 MAb at 4 days postinoculation. Results representative one of two experiments are shown. (D) Intracellular luciferase activity in cells inoculated with either eHEV-nanoKAZ3 (2 × 10⁶ copies/well) or neHEV-nanoKAZ3 (2 × 10⁵ copies/well) at 2, 4, and 6 days postinoculation was compared to that of nr-eHEV- or nr-neHEV-inoculated cells (2 × 10⁵ copies/well). The data are presented as the mean ± SD for three wells each.

To determine whether the recombinant membrane-associated HEV with nanoKAZ3 (eHEV-nanoKAZ3) and recombinant membrane-unassociated HEV with nanoKAZ3 (neHEV-nanoKAZ3) are infectious, they were inoculated to PLC/PRF/5 cells. After 4 days, the cells were subjected to IFA to assess the ORF2 protein expression. The ORF2 proteins were detected in the cells inoculated with either eHEV-nanoKAZ3 or neHEV-nanoKAZ3, similar to the cells inoculated with nonrecombinant (nr)-eHEV or nr-neHEV (16), while no ORF2 expression was detected in the mock-inoculated cells (Fig. 3C). Furthermore, the luciferase activity was determined in the cells inoculated with eHEV-nanoKAZ3 or neHEV-nanoKAZ3 at 2, 4, and 6 days postinoculation. Luciferase activity gradually increased over time in the lysates of cells inoculated with eHEV-nanoKAZ3 or neHEV-nanoKAZ3, while luciferase activity in the lysates of nr-eHEV- or nr-neHEV-inoculated cells was at a background level (Fig. 3D). These results confirmed that both eHEV-nanoKAZ3 and neHEV-nanoKAZ3 are infectious and that the intracellular luciferase activity increased over time. Here, the recombinant membrane-associated JE03-1760F/P10-nanoKAZ3 will be called eHEV-nanoKAZ, and the membrane-unassociated form will be called neHEV-nanoKAZ.

Genetic stability of eHEV-nanoKAZ.

To monitor the genetic stability of eHEV-nanoKAZ, it was subjected to five consecutive passages. eHEV-nanoKAZ generated in the culture supernatants of pJE03-1760F/P10-nanoKAZ3 RNA-transfected cells (passage 0, p0) was used as inoculum for p1. Although less efficient than the virus growth shown by eHEV, the HEV RNA levels for eHEV-nanoKAZ p1, p2, p3, p4, and p5 increased until day 28 postinoculation, reaching 3.4 × 10⁷ copies/mL, 3.3 × 10⁷ copies/mL, 1.1 × 10⁸ copies/mL, 5.0 × 10⁸ copies/mL, and 4.5 × 10⁸ copies/mL, respectively (Fig. 4A). The intracellular luciferase activity detected in the lysates of each representative well (w1, w2, and w3) at 28 days postinoculation was also stably maintained over 10⁷ RLU during the five consecutive passages (Fig. 4B). To examine the nanoKAZ insertion, purified HEV RNA from the culture supernatants on day 28 postinoculation of each passage was subjected to RT-PCR, as described above (Fig. 1D). In gel electrophoresis, the PCR products from w1, w2, and w3 of p1, p2, p3, p4, and p5 were observed to be the same size as the PCR product of p0 and longer than that of the parental virus (Fig. 4C). Furthermore, sequence analyses confirmed that the nanoKAZ insertions are intact in w1, w2, and w3 of the five consecutive passages (data not shown). These results indicated that eHEV-nanoKAZ is genetically stable.

FIG 4.

Genetic stability of eHEV-nanoKAZ. (A) The HEV RNA titer in culture supernatants of eHEV-nanoKAZ-inoculated cells during five consecutive passages (passage 1 [p1], passage 2 [p2], passage 3 [p3], passage 4 [p4], and passage 5 [p5]). The data are presented as the mean ± SD of three wells each. The dotted horizontal line represents the limit of detection by real-time RT-PCR used in this study, at 2 × 10¹ RNA copies/mL. (B) Intracellular luciferase activity of each representative well in the lysates of PLC/PRF/5 cells inoculated with eHEV-nanoKAZ during five consecutive passages. Cells were lysed 28 days postinoculation. (C) Gel electrophoresis of the RT-PCR products, which covered HVR containing the nanoKAZ insertion, from purified HEV RNA in the culture supernatants on day 28 postinoculation. Five consecutive passages were performed in three wells (w1, w2, and w3), where the culture supernatants of each well were purified and subjected to RT-PCR. The size of the PCR product with intact nanoKAZ insertion was 1,174 bp.

Capability of eHEV-nanoKAZ to replicate in multiple cell lines.

To examine the capability of eHEV-nanoKAZ to replicate in cell lines other than the PLC/PRF/5 cells, it was inoculated into human hepatocarcinoma cell HepG2/C3A (a clonal derivative of HepG2, a human hepatocellular carcinoma cell) and into the non-liver cancer cell lines A549 (human lung adenocarcinoma cell) and Caco-2 (human colorectal adenocarcinoma cell). The three cell lines have been previously reported to support HEV replication (47, 48). The HEV RNA titer in the culture supernatants of either HepG2/C3A (Fig. 5A), A549 (Fig. 5B), and Caco-2 (Fig. 5C) cells inoculated with the eHEV-nanoKAZ was steadily maintained during 28 days of observation, despite lower replication efficiency in comparison to that of eHEV. To examine the nanoKAZ expression over time in the inoculated cells, the lysates were collected at 0, 8, 16, 24, and 28 days postinoculation. The nanoKAZ expression was detectable in the HepG2/C3A (Fig. 5D), A549 (Fig. 5E), and Caco-2 (Fig. 5F) cells inoculated with eHEV-nanoKAZ, while it was on the background level in the cells inoculated with eHEV, indicating that these three cell lines can also support the propagation of eHEV-nanoKAZ, although the replication efficiency was lower in these cell lines than in PLC/PRF/5 cells.

FIG 5.

Inoculation of eHEV-nanoKAZ to multiple cell lines which support HEV replication. (A to C) The HEV RNA titer in the culture supernatants of eHEV-nanoKAZ-inoculated (A) HepG2/C3A, (B) A549, and (C) Caco-2 cells, where the eHEV served as the control. The data are presented as the mean ± SD of two wells each. The dotted horizontal line represents the limit of detection by real-time RT-PCR used in this study, at 2 × 10¹ RNA copies/mL. (D to F) Luciferase activity in the eHEV-nanoKAZ-inoculated HepG2/C3A (D), A549 (E), and Caco-2 (F) cells. Cells were lysed on day 0, 8, 16, 24, and 28 postinoculation. dpi, days postinoculation.

To elucidate whether the eHEV-nanoKAZ can replicate in not only cancer cell lines, but also in non-cancer cell lines, it was inoculated into PXB-cells, which were derived from fresh human hepatocytes isolated from chimeric mice with highly humanized liver (PXB-mouse). This cell was previously reported to support hepatitis A virus (HAV) replication (49) and hepatitis B virus (HBV) replication as well (50). To our knowledge, this is the first report on the use of PXB-cells in HEV propagation; thus, in addition to eHEV-nanoKAZ, neHEV-nanoKAZ along with eHEV and neHEV were inoculated into the PXB-cells. During 28 days of observation, eHEV and neHEV replicated in the PXB-cells, reaching 8.0 × 10⁵ copies/mL and 1.3 × 10⁶ copies/mL, respectively (Fig. 6A), by the final observation day, indicating that the PXB-cells support the HEV propagation. Despite the lower replication efficiency in comparison to eHEV and neHEV, eHEV-nanoKAZ and neHEV-nanoKAZ were also able to grow in the PXB-cells, as shown by their ability to maintain the virus production in culture supernatants during the observation period (Fig. 6A). The cell lysates were collected at 28 days postinoculation to confirm the specificity of luciferase activity. The nanoKAZ expression was detectable in the cells inoculated with eHEV-nanoKAZ and neHEV-nanoKAZ, while it was on the background level in the cells inoculated with eHEV and neHEV (Fig. 6B), demonstrating the specificity of nanoKAZ expression.

FIG 6.

Inoculation of eHEV, eHEV-nanoKAZ, neHEV, and neHEV-nanoKAZ in PXB-cells. (A) The HEV RNA titer in the culture supernatants of the PXB-cells inoculated with eHEV, eHEV-nanoKAZ, neHEV, and neHEV-nanoKAZ. The data are presented as the mean ± SD of two wells each. The dotted horizontal line represents the limit of detection by real-time RT-PCR used in this study, at 2 × 10¹ RNA copies/mL. (B) Luciferase activity in the PXB-cells inoculated with eHEV, eHEV-nanoKAZ, neHEV, and neHEV-nanoKAZ. Cells were lysed on day 28 postinoculation. The data are presented as the mean ± SD of two wells each.

Taken together, these results indicate that eHEV-nanoKAZ is capable of replicating in multiple cell lines, including the normal hepatocyte. In addition, the normal hepatocyte used in the present study was able to support not only the growth of eHEV-nanoKAZ, but also that of eHEV, neHEV, and neHEV-nanoKAZ.

Sensitivity of eHEV-nanoKAZ to anti-HEV reagents.

Previously, our JE03-1760F/P10-GLuc replicon system was able to screen drugs that inhibit HEV RNA replication (28). To examine the sensitivity of eHEV-nanoKAZ to treatment with known anti-HEV reagents, PLC/PRF/5 cells were inoculated with eHEV-nanoKAZ in the presence of various concentrations of sucrose or ribavirin (Fig. 7A). Sucrose was used to represent HEV entry inhibitor, as HEV entry is known to depend on clathrin-mediated endocytosis, not on caveola-mediated endocytosis (15). Ribavirin is an inhibitor of HEV RNA replication and has been used to treat certain cases of clinical HEV infection, such as chronic infection or acute fulminant infection (30). Genistein (inhibitor of caveola-mediated endocytosis) (51) and lomibuvir (inhibitor of RNA polymerase NS5B of hepatitis C virus [HCV], which has been used as a negative control for ribavirin in previous HEV studies [28, 37]), served as negative controls for sucrose and ribavirin, respectively (Fig. 7A). Intracellular luciferase activity was determined at 4 days postinoculation. It decreased to 84.1%, 48.5%, and 15.0% after treatment with 50, 100, and 250 mM sucrose, respectively (P < 0.001, Fig. 7A), in comparison to the control treated with 1% dimethyl sulfoxide (DMSO). In contrast, the intracellular luciferase activity was not affected by treatment with various doses of genistein (Fig. 7A), showing that eHEV-nanoKAZ is sensitive to treatment with sucrose. In addition, intracellular luciferase activity decreased to 81.3%, 31.2%, and 10.1% (in comparison to the control treated with 1% DMSO) after treatment with 8, 40, and 160 μM ribavirin, respectively (P < 0.05 or P < 0.001, Fig. 7A), while treatment with various doses of lomibuvir did not decrease the intracellular luciferase activity (Fig. 7A), suggesting that eHEV-nanoKAZ is also sensitive to the treatment of ribavirin. None of the drugs affected cellular proliferation or survival, as confirmed by cell viability assay (Fig. 7B), indicating that the decreased luciferase activity was not due to drug cytotoxicity. These results indicate that this reporter system can at least cover drugs that inhibit HEV entry, represented by sucrose, and drugs that inhibit HEV RNA replication, represented by ribavirin.

FIG 7.

Capability of eHEV-nanoKAZ to screen anti-HEV drugs. (A) Sensitivity of eHEV-nanoKAZ to sucrose (inhibitor of clathrin-mediated endocytosis) and ribavirin (inhibitor of HEV RNA replication). Genistein (inhibitor of caveola-mediated endocytosis) and lomibuvir (inhibitor of RNA polymerase NS5B of hepatitis C virus [HCV]) served as negative controls to sucrose and ribavirin, respectively. Cells were lysed 4 days after the drug treatment. (B) A cell viability assay to check the drug’s effect on cellular proliferation and survival, performed 4 days after the drug treatment. All data represent the mean ± SD of two independent experiments. *, P < 0.05; **, P < 0.001.

Application of eHEV-nanoKAZ to drug screening.

To confirm the utility of eHEV-nanoKAZ in drug screening, it was applied to screen a commercially available drug library consisting of 765 Food and Drug Administration (FDA)-approved drugs. Sucrose and ribavirin served as reference drugs, while genistein and lomibuvir served as negative controls for sucrose and ribavirin, respectively. In the primary screening and cell viability assay, all drugs were tested at a concentration of 20 μM. PLC/PRF/5 cells were inoculated with eHEV-nanoKAZ in the presence of the drugs. On day 4 postinoculation, the cell lysates were collected, and their intracellular luciferase activity was determined. The reference drugs, sucrose and ribavirin, inhibited luciferase activity 77.6% and 81.7%, respectively, without affecting cell viability, while genistein and lomibuvir did not decrease the intracellular luciferase activity (Fig. 8A). A total of 91 drugs inhibited intracellular luciferase activity >80% while maintaining >80% cell viability (Fig. 8A, red square). These drugs were then subjected to secondary screening and tested at lower concentrations of 10 μM (Fig. 8B, left panel) and 1 μM (Fig. 8B, right panel). Six drugs, which maintained cell viability at >80% and caused >70% inhibition of luciferase activity when tested at 10 μM (Fig. 8B, left panel, red square) and >50% when tested at 1 μM (Fig. 8B, right panel, red square), passed the secondary screening. A summary of the results of primary and secondary screening is presented in Fig. 8C. The inhibition of intracellular luciferase activity by the drugs that passed primary and secondary screening was comparable to that exhibited by the reference drugs sucrose and ribavirin, suggesting that eHEV-nanoKAZ is useful for screening drugs with anti-HEV activity.

FIG 8.

Application of eHEV-nanoKAZ to drug screening. (A) Results from primary screening and the cell viability assay. A total of 91 drugs with >80% inhibition of luciferase activity and for which >80% cell viability was maintained are indicated inside the red square and passed primary screening. (B) The 91 drugs that passed the primary screening were subjected to secondary screening and tested at lower doses of 10 μM (left panel) and 1 μM (right panel). The drugs that inhibit >70% and >50% of luciferase activity when tested at 10 μM and 1 μM, respectively, while maintaining >80% cell viability passed secondary screening and are indicated inside the red square. (C) Flowchart of the primary and secondary screenings. Six drugs passed the secondary screening. Cells were lysed 4 days after the drug treatment. A cell viability assay was performed 2 days after the drug treatment. Cells treated with only 1% dimethyl sulfoxide (DMSO) were used as controls in all experiments. Sucrose (represented by the red square) and ribavirin (represented by the green circle) served as reference drugs in all experiments, while genistein (represented by the yellow diamond) and lomibuvir (represented by the blue triangle) served as negative controls for sucrose and ribavirin, respectively. Luciferase activity of >200 RLU was set at the 200 mark for the scatterplots (y axis).

Anti-HEV activity of the hit drugs.

Following the successful application of eHEV-nanoKAZ in primary and secondary drug screenings, further evaluation of the hit drugs was performed. Among six hit drugs, the dose of two drugs was not suitable for administration to humans; thus, they were excluded from further analyses. The final four drugs were gefitinib, chlorpromazine, azithromycin, and ritonavir. To confirm the inhibitory activity of the four hit drugs against eHEV-nanoKAZ and the range at which inhibition occurs, eHEV-nanoKAZ was inoculated to PLC/PRF/5 cells in the presence of various concentrations of the drugs. Intracellular luciferase activity was determined on day 4 postinoculation. Cells treated with 1% DMSO were used as a control. Although the lowest dose of gefitinib (0.001 μM) did not affect intracellular luciferase activity, doses of 0.01, 0.1, 0.5, 1, 5, and 20 μM decreased the luciferase activity to 86.2%, 71.6%, 48.3%, 35.8%, 9.3%, and 8.9%, respectively (P < 0.05, P < 0.01, or P < 0.001, Fig. 9A). Chlorpromazine decreased the intracellular activity to 96.0%, 81.3%, 48.2%, 28.7%, 3.1%, 0.7%, and 0.8% after treatment with the doses of 0.01, 0.1, 0.5, 1, 2.5, 10, and 20 μM, respectively (P < 0.01 or P < 0.001, Fig. 9A). Intracellular luciferase activity decreased to 72.5%, 62.2%, 36.6%, 12.3%, 6.6%, 2.0%, and 2.1% after treatment with 0.001, 0.01, 0.1, 0.5, 1, 5, and 50 μM azithromycin, respectively (P < 0.05, P < 0.01, or P < 0.001, Fig. 9A). Ritonavir decreased intracellular luciferase activity to 85.0%, 59.7%, 25.6%, 9.5%, 3.4%, and 4.1% after treatment with doses of 0.05, 0.1, 1, 5, 20, and 35 μM, respectively (P < 0.01 or P < 0.001, Fig. 9A), where the lowest dose of 0.01 μM demonstrated no effect on luciferase activity (Fig. 9A). The anti-HEV activity of the four hit drugs was confirmed in a dose-dependent manner (Fig. 9A), while the drug treatment had no significant effect on cell viability (Fig. 9B).

FIG 9.

Anti-HEV activity of the hit drugs. (A) The inhibitory activity of gefitinib, chlorpromazine, azithromycin, and ritonavir against HEV was tested at various concentrations. Cells were lysed 4 days after the drug treatment. (B) Cell viability assay to confirm the cellular proliferation and survival of cells after treatment with various concentrations of the four hit drugs, performed 4 days after the drug treatment. All data represent the mean ± SD of two independent experiments. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Assessment of the effects of hit drugs on HEV RNA replication by JE03-1760F/P10-GLuc replicon system.

The four hit drugs in the current study were not covered in our previous study on screening of the same FDA-approved drug library using the pJE03-1760F/P10-GLuc replicon (Fig. 10A) (28), where the hit drugs in the previous study showed inhibitory activity against HEV RNA replication. This suggests that instead of inhibiting HEV RNA replication, the four hit drugs in our current study might inhibit other steps in the HEV life cycle. To confirm this hypothesis, the four hit drugs were applied to pJE03-1760F/P10-GLuc RNA-transfected PLC/PRF/5 cells at 4 h posttransfection. Ribavirin, an HEV RNA replication inhibitor, served as the reference drug, while untreated pJE03-1760F/P10-GLuc RNA-transfected cells served as a control, and the untreated replication-defective mutant (pJE03-1760F/P10-GLuc-GAA) RNA-transfected cells served as a negative control. GLuc activity was determined in the culture supernatants at 24, 48, 72, and 96 h after the drug treatment. The GLuc activity increased in the culture supernatants of the control, reaching 3.7 × 10⁷ RLU at the end of the observation time (Fig. 10B). The GLuc activity at 96 h after ribavirin treatment was 2.0 × 10⁵ RLU (Fig. 10B), which was 100-fold lower than that of the control. Meanwhile, the GLuc activity in the culture supernatants of pJE03-1760F/P10-GAA-GLuc RNA-transfected cells was steady at a low level of ∼10³ RLU (Fig. 10B). The GLuc activity increased over time to 3.1 × 10⁷ RLU, 3.0 × 10⁷ RLU, 2.5 × 10⁷ RLU, and 4.1 × 10⁷ RLU at 96 h in the pJE03-1760F/P10-GLuc RNA-transfected cells treated with gefitinib (5 μM), chlorpromazine (10 μM), azithromycin (5 μM), or ritonavir (20 μM), respectively, which was comparable to the GLuc activity of the control (Fig. 10B). These results clarified that the four hit drugs did not affect HEV RNA replication.

FIG 10.

Assessment of the effect of gefitinib, chlorpromazine, azithromycin, or ritonavir on HEV RNA replication by the replicon system consisting of HEV expressing Gaussia luciferase (JE03-1760F/P10-GLuc). (A) Schematic illustration of the pJE03-1760F/P10-GLuc replicon where ORF2 was replaced with GLuc. (B) PLC/PRF/5 cells transfected with the pJE03-1760F/P10-GLuc RNA were treated with gefitinib, chlorpromazine, azithromycin, or ritonavir, at 4 h posttransfection. GLuc activity was determined in culture supernatants at 24, 48, 72, and 96 h. Ribavirin served as the reference drug. Untreated cells transfected with pJE03-1760F/P10-GLuc RNA served as a control, while untreated cells transfected with pJE03-1760F/P10-GLuc-GAA RNA served as a negative control. All data represent the mean ± SD of two independent experiments.

Evaluation of the effectiveness of the four hit drugs against HEV in cell culture.

To evaluate the effectiveness of the four candidate drugs, the virus growth kinetics in PLC/PRF/5 cells inoculated with nr-eHEV in the presence of various concentrations of candidate drugs (where ribavirin served as the reference drug) were observed. The HEV RNA levels in the culture supernatants were quantitated, and virus growth was observed for 28 days. The HEV RNA titers of the untreated control wells in all drug groups increased and reached ∼10⁸ copies/mL on day 28 postinoculation (Fig. 11A). Initially, HEV growth was inhibited by 1, 5, and 10 μM gefitinib; however, the HEV RNA titers gradually increased and were finally similar to those of control wells with no drug treatment at (∼10⁸ copies/mL) by the end of the observation period (Fig. 11A, upper left panel). Although the wells treated with 1 μM and 5 μM chlorpromazine finally reached a similar titer to that of the control wells on day 28 (∼10⁸ copies/mL), the HEV RNA titer of cells treated with a 10-μM dose was ∼10⁵ copies/mL on day 28 (Fig. 11A, upper middle panel), which was 1,000-fold lower than that of untreated control cells. Strong inhibition of HEV growth was shown in the wells treated with 5 μM or 10 μM azithromycin (Fig. 11A, upper right panel), and in those treated with 20 μM ritonavir (Fig. 11A, lower left panel), similar to in those treated with 160 μM the reference drug, ribavirin (Fig. 11A, lower right panel). On day 28, cells treated with the highest dose of each hit drug were subjected to IFA to examine the ORF2 expression. Over 85% of untreated control cells expressed ORF2 protein (Fig. 11B, upper left panel). The ORF2 protein was also expressed by a similarly high percentage (86.7%) of cells treated with 10 μM gefitinib (Fig. 11B, upper middle panel). In cells treated with 10 μM chlorpromazine, only 17.1% expressed ORF2 protein (Fig. 11B, upper right panel). Meanwhile, the expression of ORF2 was not detected in either cells treated with 10 μM azithromycin (Fig. 11B, lower left panel) or those treated with 20 μM ritonavir (Fig. 11B, lower middle panel) on day 28. Similarly, the ORF2 expression was also undetectable in the cells treated with 160 μM reference drug, ribavirin (Fig. 11B, lower right panel).

FIG 11.

Effectiveness of gefitinib, chlorpromazine, azithromycin, or ritonavir against HEV in cell culture. (A) HEV growth kinetics were observed for 28 days in cells inoculated with nr-eHEV in the presence of various concentrations of the drugs. Ribavirin served as the reference drug. The data are presented as the mean ± SD of two independent experiments. The dotted horizontal line represents the limit of detection by real-time RT-PCR used in this study, at 2 × 10¹ RNA copies/mL. (B) Immunofluorescence staining of the drug-treated cells from day 28 in comparison to untreated cells using anti-ORF2 MAb. Two visual fields were used to quantitate the HEV-infected cells, where at least 200 cells/visual field were calculated. Results representative one of two experiments are shown.

Taken together, the dose-dependent inhibition of HEV growth by the hit drugs was confirmed in cell culture. In particular, azithromycin and ritonavir were demonstrated to have strong inhibition activity, which was demonstrated by both the decreased HEV RNA titer in culture supernatants to around the detection limit and the undetectable expression of ORF2.

Effectiveness of the four hit drugs in the inhibition of HEV growth in a culture system consisting of HEV-producing cells.

To mimic the real conditions of HEV-infected patients, cells that robustly produce HEV (PLC/PRF/5 cells infected with HEV for a long period of time, where they are able to continuously produce virus in high titer) were used to further evaluate the effectiveness of the four hit drugs. The HEV RNA titer on the day of seeding was ∼10⁴ copies/mL in all wells (Fig. 12A). Various concentrations of the drugs were applied to HEV-producing PLC/PRF/5 cells coseeded with naive cells. Ribavirin served as the reference drug. Observation was performed for 28 days following drug treatment, and the HEV RNA titers of the culture supernatants were quantitated. The HEV RNA levels of the untreated control cells continued to increase and reached ∼10⁸ copies/mL by the end of the observation period in all drug groups (Fig. 12A). Although gefitinib initially demonstrated inhibition of HEV growth, all wells had the same HEV RNA titer of ∼10⁸ copies/mL on day 28, which was similar to that of untreated control cells (Fig. 12A, upper left panel). Chlorpromazine, at doses of 10 μM and 20 μM, was able to inhibit HEV growth to ∼10⁶ copies/mL and ∼10⁵ copies/mL (100-fold and 1,000-fold lower in comparison to untreated control cells, respectively) on day 28 (Fig. 12A, upper middle panel). The highest suppression of HEV growth among the four hit drugs was demonstrated by 50 μM azithromycin (Fig. 12A, upper right panel) and 35 μM ritonavir (Fig. 12A, lower left panel); by day 28, the HEV RNA titer finally declined to (∼10^³ copies/mL) below that at the starting point, at which point it was 100,000-fold lower than that of untreated control cells. Meanwhile, at a dose of 160 μM, the reference drug ribavirin (Fig. 12A lower right panel) demonstrated strong inhibition of HEV growth. These results show that HEV growth is inhibited by the four hit drugs in a dose-dependent manner, even in the cell culture system consisting of HEV-producing cells.

FIG 12.

Effectiveness of gefitinib, chlorpromazine, azithromycin, or ritonavir in inhibition of HEV growth in the culture system consisting of HEV-producing cells. (A) HEV growth kinetics were observed for 28 days in the presence of various concentrations of the hit drugs. Ribavirin served as the reference drug. The data are presented as the mean ± SD of three wells each. The dotted horizontal line represents the limit of detection by real-time RT-PCR used in this study, at 2 × 10¹ RNA copies/mL. (B) Immunofluorescence staining of the drug-treated cells from day 28 in comparison to untreated cells using anti-ORF2 MAb. Two visual fields were used to quantitate the HEV-infected cells, where at least 200 cells/visual field were calculated. Results representative of one of two experiments are shown.

To support this result, on day 28 after the start of the drug treatment, cells treated with the highest dose of each drug were subjected to IFA to examine the ORF2 protein expression. The ORF2 protein was expressed by 92.8% of untreated control cells (Fig. 12B, upper left panel). The percentage of cells expressing ORF2 protein following treatment with 20 μM gefitinib was similar to that of the untreated control cells (90.8%) (Fig. 12B, upper middle panel). The percentage of cells expressing ORF2 protein following treatment with 20 μM chlorpromazine was lower than that of untreated cells (<20%) (Fig. 12B, upper right panel). In addition, <5% of cells treated with 50 μM azithromycin (Fig. 12B, lower left panel) or 35 μM ritonavir (Fig. 12B, lower middle panel) expressed ORF2 protein on day 28. Meanwhile, the ORF2 expression was undetectable in the cells treated with 160 μM ribavirin (Fig. 12B lower right panel).

Taken together, although not as strong as the inhibition effect of ribavirin, azithromycin and ritonavir were able to suppress HEV growth, with the HEV RNA titer becoming lower than that at the starting point; this was also supported by the extremely low expression of ORF2 protein, as observed by IFA. These results suggest that azithromycin and ritonavir, which were selected by drug screening, effectively inhibit HEV growth, even in a culture system consisting of cells that robustly produce HEV.

Evaluation of the effectiveness of azithromycin and ritonavir in the HEV-infected cells producing high virus titer in culture supernatants.

In order to imitate the condition of chronic HEV infection, particularly in the immunocompromised patients presenting with a high plasma level of HEV RNA, HEV-infected cells with virus titer in culture supernatants that plateaued at ∼10⁹ copies/mL were used to further evaluate the effectiveness of azithromycin and ritonavir (Fig. 13). The highest dose of azithromycin (50 μM) and ritonavir (35 μM) was used to treat the HEV-infected cells and observed for 28 days. Ribavirin was used as the reference drug. Azithromycin slowly decreased the virus titer to 3-fold lower in comparison to that of the starting titer on day 28 after the start of the treatment (Fig. 13). Ritonavir inhibited the virus growth more efficiently, where the virus titer decreased faster and finally reached 50-fold lower compared to that of the starting titer, on day 28 after the start of the treatment (Fig. 13). Although the inhibition efficiency was lower than that of 160 μM ribavirin, ritonavir demonstrated a faster inhibition rate in comparison to that shown by 40 μM ribavirin (Fig. 13). These results suggested that ritonavir is capable of maintaining its inhibition efficiency on HEV production in vitro even in the state of high virus titer.

FIG 13.

Inhibition of virus growth by azithromycin and ritonavir in the HEV-infected cells with high HEV RNA titer in culture supernatants, which plateaued at ∼10⁹ copies/mL. HEV growth kinetics were observed for 28 days. Ribavirin served as the reference drug. The data are presented as the mean ± SD of three wells each. The dotted horizontal line represents the limit of detection by real-time RT-PCR used in this study, at 2 × 10¹ RNA copies/mL.

DISCUSSION

Studies on the molecular aspects of HEV as well as drug screening to identify new anti-HEV drugs have greatly benefited from the discovery of bioluminescent reporter genes, particularly because the resulting reporter viruses enable rapid yet sensitive assays. Despite the various advantages offered by bioluminescent reporter viruses, several problems remain as major drawbacks. In addition to the obstacles in finding functional insertion sites, inserting a foreign gene into a viral genome, particularly a large gene, can adversely impact viral replication efficiency, viral particle formation, and the genetic stability of the inserted gene (20, 52). In the present study, we attempted to overcome the drawbacks by constructing recombinant infectious HEV harboring a small luciferase gene, nanoKAZ, in the HVR of ORF1. The HVR has been reported to accommodate either naturally acquired insertions, such as those identified in patients with acute HEV infection (38, 39), chronically HEV-infected patients (40–43), and swine (44), or insertions made by genetic recombination, such as the recently reported insertion of the luciferase gene in HVR (21). However, among the five recombinant HEVs constructed in the current study, only one (JE03-1760F/P10-nanoKAZ3) could replicate and maintain stable nanoKAZ insertion, suggesting that the HVR has limited flexibility in tolerating foreign gene insertion.

In the further characterization of JE03-1760F/P10-nanoKAZ3 (here, HEV-nanoKAZ), we showed that, although lower than the parental virus, it can replicate efficiently in cultured cells, is genetically stable, and has morphological characteristics similar to the parental virus (Fig. 1 to 4). In addition, both forms of viral particles (eHEV-nanoKAZ and neHEV-nanoKAZ) were confirmed to be infectious (Fig. 3), which will enable their application as functional tools in various HEV studies requiring both forms of viral particles, such as in the investigation of unknown HEV receptors and the elucidation of host factors that are important for HEV entry. Other than its ability to efficiently propagate in PLC/PRF/5 cells, the eHEV-nanoKAZ was also able to replicate in a wide variety of cells, ranging from the cancer cell lines HepG2/C3A, A549, and Caco-2 (Fig. 5A to C) to the normal hepatocyte, PXB-cells (Fig. 6A), where nanoKAZ is produced specifically (Fig. 5D to F and 6B), supporting its potential use in a variety of cell culture conditions in future HEV studies. The parental virus (JE03-1760F/P10) of eHEV-nanoKAZ had adapted to growth in PLC/PRF/5 cells during 10 consecutive passages (45), which may be associated with more efficient replication of eHEV-nanoKAZ in PLC/PRF/5 cells than in the three other cell lines tested. It is likely that progenies of eHEV-nanoKAZ would propagate more efficiently through consecutive passages in these cell lines as evidenced by cultivation of the JE03-1760F strain in A549 cells (53).

Recently, Szkolnicka et al. (21) reported the construction of an HEV replicon with a small luciferase (NanoLuc) tag in the HVR, allowing quantitative monitoring of HEV replication. The nanoKAZ used in our current study is a mutated 19-kDa component of Oplophorus luciferase, which has an amino acid sequence identical to NanoLuc but a different nucleotide sequence (23, 26). The HVR insertion site in the recombinant HEV used for drug screening assays (JE03-1760F/P10-nanoKAZ3) in the current study was previously reported in a chronically HEV-3-infected kidney transplant patient undergoing immunosuppressive treatment (62-aa insertions found in the HVR were derived from two parts of its ORF1); the virus demonstrated efficient replication in cultured cells (43).

HEV infections are generally self-limiting. In certain cases, such as acute fulminant or chronic cases, the administration of antiviral treatment is required. Ribavirin, an off-label drug, is given in such clinical settings. Ribavirin is a guanosine analogue with broad-spectrum in vitro activity against both RNA and DNA viruses and is indicated in several viral infections, such as those caused by HCV (in combination with interferon-α) or respiratory syncytial virus (RSV) (54). However, the use of ribavirin as an anti-HEV drug has increasingly been associated with severe side effects (55) and the development of resistance (32–34, 36), and it is contraindicated in pregnant women, who represent a major risk group (31). Therefore, novel drugs are urgently needed as additional options for HEV-infected patients who require antiviral treatment.

To evaluate the utility of HEV-nanoKAZ as a functional tool in HEV studies, we applied this system to screen commercially available FDA-approved drugs. HEV particles circulating in the bloodstream and attaching to the hepatocytes in HEV-infected patients are membrane-associated (14); thus, eHEV-nanoKAZ was employed for the drug screening assays. The Z-factor (the parameter used to evaluate the suitability of an assay [56]) for the eHEV-nanoKAZ system using an inoculum titer of 3 × 10⁶ copies/well in a 96-well plate is 0.934, suggesting that this system is suitable for application in drug screening. Four drugs were hit in the primary and secondary screenings, namely, gefitinib, chlorpromazine, azithromycin, and ritonavir. Subsequent evaluation confirmed their dose-dependent inhibition of the luciferase activity of eHEV-nanoKAZ-inoculated PLC/PRF/5 cells (Fig. 9A). On the other hand, these drugs did not affect HEV RNA replication (Fig. 10B), suggesting that they possibly inhibit another step in the HEV life cycle. Our preliminary data (Fig. 7A) showed that the eHEV-nanoKAZ system is able to cover the inhibitor of HEV entry (represented by sucrose, an inhibitor of clathrin-dependent endocytosis) and the inhibitor of HEV RNA replication (represented by ribavirin, which inhibits HEV replication in vitro by increasing the error rate of viral RdRp) (33). Solely based on these results, the four hit drugs might inhibit earlier steps of the HEV life cycle, such as attachment to the cell receptor, internalization, or uncoating.

Among the four drugs hit by screening in the current study, chlorpromazine and azithromycin have been reported to inhibit HEV infection. Chlorpromazine inhibits HEV entry through clathrin-mediated endocytosis (15), while azithromycin has recently been reported to be a potent inhibitor of HEV replication and viral protein expression in cell culture models (57), confirming the reliability of the eHEV-nanoKAZ system in identifying potential anti-HEV drugs. On the other hand, the remaining two drugs have been reported to inhibit other viral infections. Gefitinib prevents infectious spleen and kidney necrosis virus (ISKNV) infection by blocking virus-mediated endocytosis through suppression of epidermal growth factor receptor (EGFR)/phosphatidylinositol 3-kinase (PI3K) signaling and cytoskeletal rearrangement (58), while ritonavir is a protease inhibitor indicated in combination with other antiretroviral agents for the treatment of human immunodeficiency virus (HIV)-1-infected patients (59). The four hit drugs also demonstrated the inhibition of HEV growth, as observed in cell culture evaluations using wild-type HEV-3, not only in naive PLC/PRF/5 cells (Fig. 11) but also in HEV-producing PLC/PRF/5 cells (Fig. 12), clarifying the results from drug screening assays. Further studies are warranted to elucidate the precise mechanism through which the four hit drugs inhibit HEV infection.

In cell culture evaluation, gefitinib managed to inhibit HEV growth, even though the inhibition effect ended earlier in comparison to the other three drugs (Fig. 11 and 12). Although the inhibition displayed by chlorpromazine on HEV growth is not as strong as that of azithromycin and ritonavir, by the final day of observation, the highest virus titer was only ∼10⁵ copies/mL, which is 1,000-fold lower than that of untreated control cells, suggesting its moderate inhibition activity, as indicated by the IFA results (Fig. 11 and 12). Azithromycin and ritonavir had similarly strong inhibition activity on HEV growth, as shown by the decreased HEV RNA titer in culture supernatants of the naive PLC/PRF/5 cells (Fig. 11A) and supported by the results of the IFA, where the intracellular ORF2 expression was undetectable (Fig. 11B and 12B), confirming their potent inhibition activity in a long-term cell culture, even in cells that robustly produce HEV. Although they strongly inhibited HEV growth, the HEV RNA titer in culture supernatants was relatively unchanged in comparison to the starting virus titer in the culture system consisting of HEV-producing cells (Fig. 12A). This is different than the effect shown by ribavirin. As ribavirin is an inhibitor of HEV RNA replication, it is able to inhibit virus production by coseeded virus-producing cells and, therefore, to decrease the HEV RNA titer in culture supernatants. On the other hand, as the mechanisms underlying the effects of the four hit drugs in the present study do not involve the inhibition of HEV RNA replication (Fig. 10B), it is understandable why the virus titer in culture supernatants of azithromycin- and ritonavir-treated coseeded virus-producing cells was relatively similar to that of the starting point. Azithromycin is generally safe to use in pregnancy and is designated pregnancy category B by the FDA (60). As for ritonavir, the Antiretroviral Pregnancy Registry (APR) reports no evidence of increased risk in human teratogenicity (61). Furthermore, in addition to its ability to inhibit HEV growth in the coseeded virus-producing cells, ritonavir is able to maintain its inhibition activity even in the state of high virus titer in vitro (Fig. 13), the condition similar to chronic HEV infection (particularly in immunocompromised patients), therefore further supporting its potential therapeutic use.

In conclusion, we developed the HEV-nanoKAZ system as a functional investigation tool with successful application in drug screening assays. The results of this study have emphasized the benefit of using a small-sized luciferase gene in maintaining genetic stability for the successful development of a functional bioluminescent reporter system. The utility of this novel system can be further expanded beyond drug screening to study molecular aspects of the HEV life cycle, taking into consideration the availability of both particle forms (membrane-associated and membrane-unassociated). In addition, regarding the major drawbacks associated with the current treatment for HEV infections using the off-label drug ribavirin, the drugs identified in the current study may be potential treatment options for HEV infections.

MATERIALS AND METHODS

Cell culture.

PLC/PRF/5 (CRL-8024) and HepG2/C3A (HB-8065) cells obtained from the American Type Culture Collection (ATCC) were grown in Dulbecco’s modified Eagle medium (DMEM) (Thermo Fisher Scientific) containing 10% heat-inactivated fetal bovine serum (FBS) (Thermo Fisher Scientific) (growth medium). The growth medium for HEV-infected cells was supplemented with 1% dimethyl sulfoxide (DMSO) (Fujifilm Wako). A549 cells obtained from ATCC (CCL-185) were grown in 50% DMEM and 50% medium 199 (Thermo Fisher Scientific) containing 2% heat-inactivated FBS and 30 mM MgCl₂. Caco-2 cells obtained from ATCC (HTB-37) were grown in DMEM containing 20% heat-inactivated FBS and 1× MEM nonessential amino acid solution (MEM NEAA) (Thermo Fisher Scientific). PXB-cells purchased from PhoenixBio were grown in the medium specific for the PXB-cells (PhoenixBio) according to the manufacturer’s recommendation. The culture was performed at 37°C under a humidified 5% CO₂ atmosphere.

Viruses.

Culture supernatants containing cell culture-adapted membrane-associated JE03-1760F strain (passage 26) (62) (nonrecombinant [nr]-eHEV) were used in this study. To generate the membrane-unassociated particles (nr-neHEV), the culture supernatants were treated with 0.2% sodium deoxycholate (DOC-Na) and 0.2% trypsin and incubated at 37°C for 3 h to remove the lipid membrane and ORF3 protein, as described previously (16). JE03-1760F/P10 (eHEV) is based on a cell culture-adapted HEV-3 strain, generated after 10 consecutive passages of the wild-type strain. It has enhanced HEV production in comparison to JE03-1760F.

Plasmids.

The constructs used in this study were prepared using pJE03-1760F/P10 of an HEV-3 strain (GenBank accession number LC126332) (45) as a template. To determine the site of insertion for the nanoKAZ DNA sequence in ORF1, we compared the amino acid length of the ORF1 protein of the JE03-1760F/P10 strain with 161 other HEV strains. Nine HEV strains demonstrated an ORF1 protein of >30 amino acids longer than that of the JE03-1760F/P10 in five distinct sites of the hypervariable region (HVR) (Table 1). We inserted the nanoKAZ gene into the five sites (D739/I740, E748/E749, G765/L766, P772/P773, and P773/P774 of the ORF1 protein) to construct plasmids pJE03-1760F/P10-nanoKAZ1 to -nanoKAZ5 (Table 1 and Fig. 1A). The HEV infectious cDNA clone (pJE03-1760F/P10) was digested with NotI (nucleotide [nt] 2181) (New England Biolabs) and BstBI (nt 2820) (New England Biolabs). The artificially synthesized nanoKAZ gene (Genscript) was used as a template and amplified by PCR with nanoKAZ-specific forward and reverse primers (Table 2). PCR product 1 (starts at NotI [nt 2181] to just before the nanoKAZ insertion) and PCR product 2 (starts just after the nanoKAZ insertion to BstBI [nt 2820]) were amplified by PCR with insertion site-specific forward and reverse primers (with 15-bp complementary ends of a nanoKAZ insert and linearized cloning vector), using the pJE03-1760F/P10 plasmid as a template (Table 2). NotI- and BstBI-digested linearized plasmid DNA, the PCR-generated nanoKAZ gene, PCR product 1, and PCR product 2 were fused using an In-Fusion HD cloning kit according to the manufacturer’s protocol (TaKaRa Bio). All constructs were verified by sequence analyses.

TABLE 2.

Primers used to construct pJE03-1760F/P10-nanoKAZ1 to -nanoKAZ5

PCR product	Forward primer (5′–3′)	Reverse primer (5′–3′)
nanoKAZ	atggtcttcaccctggaggacttcgtcggcgac	ggccaggattctctcgcacagtctccagccgg
nanoKAZ1 PCR product 1	ccccctgaggcggccgcccccgcaccggctgctgc	cagggtgaagaccatatcactgactggtggagtgg
nanoKAZ1 PCR product 2	gagagaatcctggccatctgggtgttaccaccgcc	ggcttgttagcttcgaaccaggccgcagcggg
nanoKAZ2 PCR product 1	ccccctgaggcggccgcccccgcaccggctgctgc	cagggtgaagaccatttctgagggcggtgg
nanoKAZ2 PCR product 2	gagagaatcctggccgagcttcaggttgatgcgg	ggcttgttagcttcgaaccaggccgcagcggg
nanoKAZ3 PCR product 1	ccccctgaggcggccgcccccgcaccggctgctgc	cagggtgaagaccatcccgacgggcccgggggcgg
nanoKAZ3 PCR product 2	gagagaatcctggccctgcccagccttcctgagcc	ggcttgttagcttcgaaccaggccgcagcggg
nanoKAZ4 PCR product 1	ccccctgaggcggccgcccccgcaccggctgctgc	cagggtgaagaccataggctcaggaaggctggg
nanoKAZ4 PCR product 2	gagagaatcctggccccaccccccgtgcgtaagcc	ggcttgttagcttcgaaccaggccgcagcggg
nanoKAZ5 PCR product 1	ccccctgaggcggccgcccccgcaccggctgctgc	cagggtgaagaccattggaggctcaggaaggc
nanoKAZ5 PCR product 2	gagagaatcctggccccccccgtgcgtaagccacc	ggcttgttagcttcgaaccaggccgcagcggg

In addition, as negative controls, pJE03-1760F/P10-GAA and pJE03-1760F/P10-nanoKAZ3-GAA were generated by mutating the conserved GDD motif in the RNA-dependent RNA polymerase (RdRp) region of ORF1 to GAA; the catalytic residues Asp1561 and Asp1562 were substituted with alanine, as described previously (45).

In vitro transcription and transfection of RNA transcripts to PLC/PRF/5 cells.

Each of the full-length genome plasmids, including pJE03-1760F/P10 and pJE03-1760F/P10-nanoKAZ1 to -nanoKAZ5, as well as pJE03-1760F/P10-GAA and pJE03-1760F/P10-nanoKAZ3-GAA, were linearized with NheI, and RNA transcripts were synthesized with T7 RNA polymerase using the AmpliScribe T7-Flash transcription kit (Epicentre Biotechnologies). Following in vitro transcription, RNA transcripts were capped using a ScriptCap m7G capping system (Epicentre Biotechnologies). The integrity and yield of the synthesized RNAs were determined by agarose gel electrophoresis. An aliquot (0.025 μg) of each capped RNA was transfected into PLC/PRF/5 cells (with >80% confluence) in a well of a 24-well plate (Thermo Fisher Scientific) using the TransIT-mRNA transfection kit (Mirus Bio) according to the manufacturer’s recommendations. Following incubation at 37°C under a humidified 5% CO₂ atmosphere for 48 h, the cells were washed with phosphate-buffered saline (pH 7.5) without Mg²⁺ and Ca²⁺ [PBS(–)], and the culture medium was replaced with 0.5 mL of growth medium supplemented with 1% DMSO, and the cells were incubated at 35.5°C under a humidified 5% CO₂ atmosphere. Every other day, half of the culture medium (0.25 mL) was replaced with fresh growth medium containing 1% DMSO. The collected culture supernatants were stored at −80°C until use. The nanoKAZ insertion was confirmed by RT-PCR and a sequence analysis as described below.

For inoculum in the drug screening assays, culture supernatants of PLC/PRF/5 cells transfected with 0.025 μg/well of pJE03-1760F/P10-nanoKAZ3 RNA in a six-well plate (Iwaki) were collected. Quantitation of the HEV RNA titer and confirmation of nanoKAZ insertion were performed as described below. Before being used as inoculum for drug screening assays, the collected culture supernatants were filtered through a microfilter with a pore size of 0.22 μm (Millex-GV; Millipore Corp.).

Quantification of HEV RNA.

Total RNA was extracted from culture supernatants using TRIzol-LS reagent (Thermo Fisher Scientific). The quantitation of HEV RNA was performed by RT-PCR using a LightCycler apparatus (Roche Diagnostics KK) with a QuantiTect Probe RT-PCR kit (Qiagen), a primer set, and a probe targeting the overlapping region of ORF2 and ORF3, according to a previously described method (63).

Luciferase assay.

To examine the nanoKAZ luciferase activity, PLC/PRF/5 cells seeded in a 96-well plate (Thermo Fisher Scientific) were washed twice in PBS(–) and then lysed with 20 μL of cell lysis buffer (JNC Corporation). The plate was agitated using plate shaker N-704 series B (Nissin) at 1,500 rpm for 10 min, followed by centrifugation using a plate centrifuge (As One) at 2,200 rpm for 3 min. For 24-well plate, cells were washed and then pelleted in a 1.5-mL tube (Eppendorf). Following addition of 100 μL cell lysis buffer, the tube was vortexed and briefly centrifuged. The resulting supernatants were collected, where 10 μL were then diluted in 30 μL of assay buffer for coelenterazine (CTZ)-type luciferase (JNC Corporation) in a 96-well microplate (Berthold Technologies). Then, 40 μL of 250 ng/mL h-CTZ (JNC Corporation) in assay buffer for CTZ-type luciferase was added to each well, and the luminescence kinetics were measured with a TriStar² LB942 multimode plate reader (Berthold Technologies). The measured luminescence intensity values were represented as relative light units (RLUs) and normalized to that of the DMSO (vehicle) control. For HepG2/C3A, A549, Caco-2, and PXB-cells, all experiments were carried out in 24-well plates, and luciferase activity was measured as described above.

RT-PCR and sequence analysis.

HEV RNAs purified from the culture supernatants using TRIzol-LS reagent were subjected to RT-PCR (PrimeScript one-step RT-PCR kit v2; TaKaRa Bio), which covers the HVR containing the nanoKAZ insertion, using primers HE290 (5′-GGTATAAYAGATTCACCCAG-3′; plus the strand sequence from nt 1986 to 2005 of the HEV genome) and HE104 (5′-GACTATAGGGTTGAGCAGAAYCC-3′; minus the strand sequence from nt 2624 to 2646 of the HEV genome) (Y = T/C), followed by gel electrophoresis of the PCR products (50°C for 30 min, 94°C for 2 min, 30 cycles of 94°C for 30 s, 52°C for 30 s, and 72°C for 1 min and then 72°C for 5 min). The size of the PCR product with no nanoKAZ insertion was 661 bp; with the nanoKAZ insertion intact, the size would be 1,174 bp. For a sequence analysis, the amplified products were sequenced on both strands using a BigDye Terminator v3.1 cycle sequencing kit on an ABI PRISM 3130xl genetic analyzer (Thermo Fisher Scientific). The sequence analysis and generation of multiple alignments were performed using the Genetyx software program (Genetyx Mac v19).

Western blotting.

To check the expression of ORF2 and ORF3 proteins in the cells transfected with RNA transcripts of pJE03-1760F/P10-nanoKAZ3, cells were lysed in lysis buffer as previously described (16). The proteins in the cell lysates were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and blotted onto polyvinylidene difluoride (PVDF) membranes (0.45 μM) (Merck-Millipore), immunodetected with specific antibodies, and visualized with a chemiluminescence assay using SuperSignal West Femto chemiluminescent substrate (Thermo Fisher Scientific) with an ImageQuant LAS 500 instrument (GE Healthcare), as described previously (64). The primary antibodies used were anti-HEV ORF2 MAb (H6210) (63) and anti-HEV ORF3 MAb (TA0536) (13), both of which were made in-house.

Immunofluorescence assay.

HEV-infected PLC/PRF/5 cells seeded into four-well chamber slides (Thermo Fisher Scientific) were subjected to immunofluorescence staining according to a previously described method (64). The primary antibody used was anti-HEV ORF2 MAb (H6225) (63), and the secondary antibody was Alexa-Fluor 488-conjugated anti-mouse IgG (Thermo Fisher Scientific). Nuclei were counterstained with 4′,6-diamidino-2-phenylindole dihydrochloride (DAPI; Thermo Fisher Scientific). Slide glasses were mounted with Fluoromount/Plus medium (Diagnostic BioSystems) and then viewed under an FV1000 confocal laser microscope (Olympus). Two visual fields were used to quantitate the HEV-infected cells, where at least 200 cells/visual field were calculated. Data are presented as the mean ± standard deviation (SD) from two visual fields each.

Sucrose density gradient centrifugation.

The culture supernatants containing HEV progenies (3.0 × 10⁶ copies/mL) collected from PLC/PRF/5 cells transfected with RNA transcripts of pJE03-1760F/P10-nanoKAZ3 were subjected to equilibrium centrifugation in a sucrose density gradient, as described previously (13). The gradients were fractionated, and the density of each fraction was measured using refractometry. By a similar method, culture supernatants containing HEV progenies (2.8 × 10⁶ copies/mL) collected from PLC/PRF/5 cells transfected with RNA transcripts of pJE03-1760F/P10-nanoKAZ3 were treated with 0.2% DOC-Na and 0.2% trypsin in PBS(–) at 37°C for 3 h and pelleted according to the previously described method (16) and then subjected to equilibrium ultracentrifugation in a sucrose density gradient.

Transmission electron microscopy (TEM).

Cell culture-generated eHEV-nanoKAZ3 particles were purified by ultracentrifugation as previously described (16). neHEV-nanoKAZ3 particles were generated by treating the culture supernatants containing eHEV-nanoKAZ with DOC-Na and trypsin, as described above. TEM was performed as previously described (16).

Immunoelectron microscopy (IEM).

Immunogold labeling was performed in accordance with the previously described method (16). The primary antibody used was anti-HEV ORF2 MAb (H6225). Goat anti-mouse IgG conjugated with 12 nm colloidal gold was used as a secondary antibody (Jackson ImmunoResearch Laboratories). The grids were negatively stained and observed by TEM.

Virus inoculation and consecutive passages.

Monolayers of PLC/PRF/5 cells in a 24-well plate were inoculated with 1 × 10⁶ copies of HEV progenies in the culture supernatants of pJE03-1760F/P10-nanoKAZ3 RNA-transfected cells and diluted in PBS(–) containing 0.2% bovine serum albumin (BSA) (Sigma-Aldrich Merck). Following incubation at room temperature for 1 h, the cells were washed five times with PBS(–), and 0.5 mL of growth medium containing 1% DMSO was added to each well; then the cells were incubated at 35.5°C under a humidified 5% CO₂ atmosphere. Every other day, half of the culture medium (0.25 mL) was replaced with fresh growth medium containing 1% DMSO. The collected culture supernatants were stored at −80°C until use.

For consecutive passages, culture supernatants were filtered through a microfilter (as described above), and 1 × 10⁶ copies of HEV progenies were inoculated onto a monolayer of PLC/PRF/5 cells in a 24-well plate. Quantitation of the HEV RNA titer, confirmation of nanoKAZ insertion, and preparation of cell lysates for luciferase assay were performed as described above.

Drugs.

Sucrose (Fujifilm Wako), ribavirin (Fujifilm Wako), genistein (Fujifilm Wako), lomibuvir (Selleckchem), Screen-Well Food and Drug Administration (FDA)-approved drug library v2 (Japanese version) (Enzo Life Sciences), gefitinib (Tokyo Chemical Industry), chlorpromazine (Tokyo Chemical Industry), azithromycin (Tokyo Chemical Industry), and ritonavir (Tokyo Chemical Industry), were purchased from the indicated sources.

Drug screening assay.

PLC/PRF/5 cells at 1.5 × 10⁴ cells/well of a 96-well plate were incubated at 37°C under a humidified 5% CO₂ atmosphere for 72 h. For primary screening, monolayers of PLC/PRF/5 cells were washed twice with PBS(–) and then inoculated with JE03-1760F/P10-nanoKAZ3 at 3 × 10⁶ copies/well in a growth medium without FBS, containing 20 μM drug with a final concentration of 1% DMSO. At 37°C under a humidified 5% CO₂ atmosphere, 4 h after incubation, growth medium containing 20 μM drug with a final concentration of 1% DMSO was added to each well, followed by incubation at 37°C under a humidified 5% CO₂ atmosphere for 96 h. Secondary screening was performed with 1 μM or 10 μM drug using the same protocol. Luciferase activities were measured as described above.

Cell viability assay.

Cell viability was measured using a Cell Counting kit-8 (Dojindo Laboratories) according to the manufacturer’s protocol. In brief, PLC/PRF/5 cells were seeded in a 96-well plate and incubated at 37°C under a humidified 5% CO₂ atmosphere for 48 h. The indicated doses of drugs with a final concentration of 1% DMSO were then added to each well, followed by incubation of the cells at 37°C under a humidified 5% CO₂ atmosphere for 48 h or 96 h. Subsequently, highly water-soluble tetrazolium salt (WST-8) solution was added to each well, and the cells was incubated at 37°C under a humidified 5% CO₂ atmosphere for 2 h. Absorbance was measured at 490 nm using an iMark microplate reader (Bio-Rad Laboratories). Measured values were normalized to the value of the DMSO (vehicle) control.

Measurement of GLuc activity.

Monolayers of PLC/PRF/5 cells in a 48-well plate (Sigma-Aldrich Merck) were transfected with 0.25 μg of pJE03-1760F/P10-GLuc RNA, as described previously (28). After incubation at 37°C under a humidified 5% CO₂ atmosphere for 4 h, the cells were washed twice with PBS(–), and 0.25 mL of growth medium containing the indicated drug dose with a final concentration of 1% DMSO was added to each well. The cells were then incubated at 35.5°C under a humidified 5% CO₂ atmosphere. Culture supernatants at 0, 24, 48, 72, and 96 h posttransfection were collected and stored at −80°C until use. Measurement of the GLuc activity in the culture supernatants of the transfected cells was performed as described previously (28).

Evaluation of drug effectiveness in cell culture system.

Monolayers of PLC/PRF/5 cells in a 24-well plate were inoculated with 5 × 10⁴ copies of nr-eHEV in growth medium without FBS, containing the indicated drug dose with a final concentration of 1% DMSO, and then subsequently incubated at 37°C under a humidified 5% CO₂ atmosphere for 2 h. After incubation, the cells were washed five times with PBS(–), and 0.5 mL of growth medium containing the indicated drug dose with a final concentration of 1% DMSO was added to each well, followed by incubation at 35.5°C under a humidified 5% CO₂ atmosphere. Every other day, half of the culture medium was replaced with fresh growth medium containing 1% DMSO and the indicated concentration of the drug. Collected culture supernatants were stored at −80°C until use.

Evaluation of drug effectiveness using PLC/PRF/5 cells that robustly produce HEV.

The nr-eHEV was inoculated into PLC/PRF/5 cells. After the HEV RNA titer in culture supernatants of the infected cells reached the plateau stage, they were collected. A mixture of naive PLC/PRF/5 cells at 1 × 10⁵ cells/well and 1 × 10² HEV-infected cells/well were prepared and seeded into a 24-well plate and then incubated at 35.5°C under a humidified 5% CO₂ atmosphere. After 48 h, the culture supernatants were removed, the cells were washed with PBS(–) five times, and 0.5 mL of growth medium containing the indicated drug dose with a final concentration of 1% DMSO was added to each well. The cells were incubated at 35.5°C under a humidified 5% CO₂ atmosphere. Every other day, half of the culture medium was replaced with fresh growth medium containing the indicated drug dose with a final concentration of 1% DMSO. The collected culture supernatants were stored at −80°C until use.

Evaluation of the effectiveness of azithromycin and ritonavir in the HEV-infected cells producing high-titer HEV in culture supernatants.

The nr-eHEV was inoculated into PLC/PRF/5 cells. After the HEV RNA titer in culture supernatants of the infected cells reached the plateau stage at ∼10⁹ copies/mL, the culture supernatants were removed, the cells were washed with PBS(–) five times, and 0.5 mL of growth medium containing the indicated drug dose with a final concentration of 1% DMSO was added to each well. The cells were incubated at 35.5°C under a humidified 5% CO₂ atmosphere. Every other day, half of the culture medium was replaced with fresh growth medium containing the indicated drug dose with a final concentration of 1% DMSO. The collected culture supernatants were stored at −80°C until use.

Statistical analysis.

The results are presented as the mean ± SD. Statistical significance was assessed by Student’s t test. P values of <0.05 were considered statistically significant.